Immigration and the Transformation of Europe

This page intentionally left blank

Immigration and the Transformation of Europe

A new kind of historic transformation is underway in twenty-firstcentury Europe. Twentieth-century Europeans were no strangers to social, economic, and political change, but their major challenges focused mainly on the intra-European construction of stable, prosperous, capitalist democracies. Today, by contrast, one of the major challenges is flows across borders – and particularly in-flows of non-European people. Immigration and minority integration consistently occupy the headlines. The issues which rival immigration – unemployment, crime, terrorism – are often presented by politicians as its negative secondary effects. Immigration is also intimately connected to the profound challenges of demographic change, economic growth and welfare-state reform. Both academic observers and the European public are increasingly convinced that Europe’s future will largely turn on how it admits and integrates non-Europeans. This book is a comprehensive stocktaking of the contemporary situation and its policy implications.       .       is Assistant Professor of Political Science, University of Oregon.        .         is Maxwell Professor of Public Policy and Professor of Economics and Public Administration, at the Maxwell School of Syracuse University.

Immigration and the Transformation of Europe Edited by

Craig A. Parsons University of Oregon

Timothy M. Smeeding Maxwell School of Syracuse University

cambridge university press Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo Cambridge University Press The Edinburgh Building, Cambridge cb2 2ru, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9780521861939 © Cambridge University Press 2006 This publication is in copyright. Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published in print format 2006 isbn-13 isbn-10

978-0-511-24554-1 eBook (EBL) 0-511-24554-8 eBook (EBL)

isbn-13 isbn-10

978-0-521-86193-9 hardback 0-521-86193-4 hardback

Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

For Kati Foley with great appreciation

Contents

List of figures List of tables List of contributors Acknowledgements 1 What’s unique about immigration in Europe?       .                 .  2 Europe’s immigration challenge in demographic perspective         3 Migration into OECD countries 1990–2000     .        ,             ,      4 Divergent patterns in immigrant earnings across European destinations            `        .         5 Economic consequences of immigration in Europe              ,       .      ,       .      6 Occupational status of immigrants in cross-national perspective: A multilevel analysis of seventeen Western societies               7 Immigrants, unemployment, and Europe’s varying welfare regimes  

page ix xi xvi xix 1

30 43

85 111

147

172

vii

viii

Contents

8 How different are immigrants? A cross-country and cross-survey analysis of educational achievement    9 Immigration, education, and the Turkish second generation in five European nations: A comparative study                           10 Managing transnational Islam: Muslims and the state in Western Europe               11 Migration mobility in European diasporic space             12 The new migratory Europe: Towards a proactive immigration policy?             13 European immigration in the people’s court                     14 The politics of immigration in France, Britain, and the United States: A transatlantic comparison       .       15 “Useful” Gastarbeiter, burdensome asylum seekers, and the second wave of welfare retrenchment: Exploring the nexus between migration and the welfare state         16 The European Union dimension: Supranational integration, free movement of persons, and immigration politics          17 The effectiveness of governments’ attempts to control unwanted migration     .           Index

200

235

251 274

298 327

362

393

419

442

473

Figures

1.1. Estimated and projected populations of the world, major areas: 1950, 2000, and 2050. page 12 3.1. Total volume of gross immigration inflows to 17 OECD countries, 1990–2000, by source-country continent. 50 3.2. Gross immigration inflows from EU- and Non-EU countries as a percentage of total inflows from Europe into 17 OECD countries, 1990–2000. 52 3.3. Stock of foreign population as a percentage of total population in 1990 and 2000 in selected OECD countries. 53 3.4. Proportion of total immigration stock in OECD countries by continents of origin, 1990 and 2000. 54 5.1. Share of the 20–29 age group in selected European regions. 139 7.1. Percentage of persons under the 50 median income poverty line by immigrant status. 179 7.2. Unemployment regimes and expected poverty outcomes. 182 7.3. Poverty rates (50% median) for different earners’ groups amongst immigrants. 189 7.4. Poverty rates (50% median) for different earners’ groups amongst non-immigrants. 190 7.5. Poverty reduction effectiveness by benefit type for workless immigrant households. 192 7.6. Poverty reduction effectiveness by benefit type for workless non-immigrant households. 193 7.A.1. Luxembourg Income Study data. 199 8.1. Percentage of natives and immigrants not able to solve basic math tasks in TIMSS. 206 8.2. Ratio of achievement scores of immigrants to natives by percentile for PISA reading. 210 ix

x

List of figures

8.3. Differences in achievement and socioeconomic background between natives and immigrants in PISA reading. 13.1. Estimated vs actual percent foreign-born in country. 17.1. Asylum applications in Europe and North America, 1980–99. 17.2. Asylum policy deterrence index in the EU (1985–2000). 17.3. Relative asylum burdens, 1985–99.

215 340 445 458 460

Tables

1.1. 1.2. 1.3. 1.4. 1.5. 1.6. 1.7.

2.1.

2.2.

2.3.

3.1. 3.2.

3.3.

Population figures for Europe in 2003 Rough estimates of Muslim population, by country Foreign population by country, 2002 Fertility Labor force participation by older people (aged 55–64), 2003 Indicators of population structure: Elderly dependency, 2004, 2025, 2050 Immigration required to avoid population decline and maintain working-age population and support ratios, European Union, 2000–2050 Population (millions) of the eight most populous countries (as of 2005) of the EU-25 and of the Broader Middle East, 1950 and 2005, and projected population in 2050 Population change (percent) in the eight most populous countries of the EU-25 and of the Broader Middle East, 1950–2005, and projected population change, 2005–2050 and 1950–2050 Assumed net international migration: most populous countries of the EU-25 and of the Broader Middle East, 2005–2050 Proportion of immigration stock in particular OECD countries by continents of origin, 1990 and 2000 Estimation of migration flows from 129 source countries (i) to 26 (OECD) destination countries (j), 1990–2000 Estimation of migration flows from 129 source countries (i) to 26 (OECD) destination countries (j), random effects 1990–2000

page 3 8 10 12 13 14

16

34

36

38 56

63

67

xi

xii

List of tables

3.A.1. Descriptive statistics of basic variables for OECD destination countries (means, standard deviations, and number of years observed in the data set) 3.A.2. Descriptive statistics of basic variables for source countries (mean, standard deviation, and number of years with information) 4.1. Number of immigrants in sample by continent of origin for each destination country 4.2. Means of variables for men 4.3. Means of variables for women 4.4. Predicted percent differences in earnings of immigrants at arrival and natives, with the same characteristics, by gender 4.5. Predicted percent differences in earnings of immigrants at arrival in different destination countries as compared to those arriving in Germany, by gender 4.6. Selected partial effect of variables from a regression analysis of earnings for the native and foreign-born by gender(a) 4.7. Partial effect of region of birth on earnings compared to native-born in that country, by gender 4.A.1. Share of immigrants in sample by origin and destinations 4.A.2. Number of individuals and of foreign-born in the sample by destination 4.A.3. Predicted natural logarithm of earnings at arrival for women: Natives, foreigners, and non-EU foreigners 4.A.4. Predicted natural logarithm of earnings for men: Natives, foreigners and non-EU foreigners 4.A.5. Proportion of immigrants with group language match 5.1. Review of empirical studies: The wage impact of international migration 5.2. Review of empirical studies: The impact of international migration on employment 5.3. Population living in private households with prime-aged heads in selected EU countries, 1994–98, by household immigrant status 5.4. Pre-government income positions and redistribution effects for households with prime-aged heads in selected EU countries, 1994–98

73

77 87 92 94

97

99

101 104 107 108 109 109 110 122 124

127

129

List of tables

5.5. Odds ratios predicting the concerns of German adults about immigration in 2000 5.6. Odds ratios predicting the concerns of parents about immigration 6.1. Occupational status (ISEI) of male immigrants by country of origin, country of destination, and setting 6.2. Occupational status (ISEI) of female immigrants by country of origin, country of destination, and setting 6.3. Cross-classified multilevel regression of socioeconomic status (ISEI) in seventeen Western countries, 1980–2002, immigrants between 25 and 54 years old 7.1. Unemployment rates amongst immigrants and non-immigrants, 2001–02 average 7.2. Poverty ratio 7.3. Poverty incidence amongst households in receipt of unemployment benefits 7.4. Relative poverty reduction effectiveness scores for unemployment benefits 7.5. Distribution of no-earner, one-earner, and two-earner households amongst immigrants and non-immigrants 7.6. Recipience of means-tested benefits amongst workless households 8.1. Educational achievement surveys and sample sizes of immigration countries covered 8.2. Differences in average scores between native and immigrant pupils by country and measure 8.3. Selection of OLS regression results showing differences in achievement between migrants and natives unconditional (Model 1) and conditional on parental background (Model 2) and on parental background and school segregation (Model 3) for PISA math, TIMSS math (8th graders) and PIRLS; dependent variable pupils’ achievement score 8.A.1. Correlation coefficients of differences in achievement scores between natives and immigrants between surveys 8.A.2 Variables and coding for regression analysis 8.A.3. PISA summary statistics 8.A.4. TIMSS summary statistics 8.A.5. PIRLS summary statistics 8.A.6. OLS regression results for PISA math (Model 2) 8.A.7. OLS regression results for TIMSS math (Model 2)

xiii

133 134 157 158

160 175 180 185 186 187 191 203 208

214

227 228 229 230 231 232 233

xiv

List of tables

8.A.8. OLS regression results for PIRLS reading (Model 2) 9.1. Population of Turkish descent in Germany, The Netherlands, France, Austria, and Belgium 9.A.1. Highest completed level of education of Turkish second-generation youngsters in the age category 16–25, according to gender (in percent) in Germany, 1995 9.A.2. Highest completed level of education of Turkish second-generation youngsters in The Netherlands aged 15–35 who had ended their educational careers by 1998 (percentages rounded) 9.A.3. Highest completed level of education of Turkish second-generation youngsters aged 18–40, according to sex, France, 1999 9.A.4. Highest completed level of education of Turkish second-generation men aged 18 years and older in Belgium, 1996 9.A.5. Highest completed level of education of Austrian residents born in Turkey or having Turkish citizenship according to the age category 15–35 (N=590), 2001 9.A.6. Highest completed level of education of Turkish second-generation youth in the age category 16–25, except pupils in the general education system or in night schools (in percent) in Germany, 1999 11.1. Ghanaian nationals by citizenship in selected EU countries (1998) 11.2. Somali nationals by citizenship in selected EU countries (1998) 11.3. Nigerian nationals by citizenship in selected EU countries (1998) 11.4. Ethiopian nationals by citizenship in selected EU countries (1998) 11.5. Senegalese nationals by citizenship in selected EU countries (1998) 11.6. Democratic Republic of Congo nationals by citizenship in selected EU countries (1998) 11.7. Mobility patterns of selected interviewees 11.8. Asylum applications from Ghanaian nationals (1980–99) 13.1. European beliefs about immigration and immigrants, by country

234 237

249

249

249

250

250

250 278 278 278 279 279 279 284 285 333

List of tables

13.2. Relationships between demographic and economic context and immigration opinions 13.3. Predicting opposition to immigration 13.4. Statistical significance of selected variables in predicting preferred level of immigration in twenty European countries 13.A. Country-level data 13.B. Individual-level questions and measures 14.1. Percentage of respondents who claim there are “too many Arabs in France” 14.2. The motivations of voters: 1984–97 14.3. Circumscriptions in France (m´etro) with 10%+ immigrant population (1999) by the political party of the d´eput´e 14.4. MPs’ attitudes on immigration and repatriation in 1969 and 1982 14.5. British local authorities with 10% or more non-white population, by political representation 14.6. Opposition to immigration in the United States 14.7. US Congressional Districts 10%+ born outside of the United States (2000) after Congressional elections of 1998 15.1. Increase in total number of migrants in percent from 1970 to 2000, arranged by welfare state family 17.1. Average number of asylum applications per year, 1985–99 (per thousand of population) 17.2. The main asylum seekers producting countries (top 5) 17.3. Expected relationship between variables 17.4. Determinants for the relative number of asylum applications 17.5. Impact of individual deterrence measures 17.6. Questioning the effect of restrictive policy measures: The case of Germany 17.7. Questioning the effect of restrictive policy measures: The case of the UK

xv

336 347

351 357 358 377 378

379 382 383 385

387 402 446 455 459 464 465 467 467

Contributors

         ` , Assistant Professor of Economics; NICHD Postdoctoral Fellow, Population Research Center, University of Chicago, IL, USA; Research Fellow, IZA Institute for the Study of Labor, Bonn, Germany.             , Senior Lecturer in Italian Studies, Department of European Studies, University of Bath, United Kingdom.              , Head, Department of European Integration and Comparative Analysis, Institute for Employment Research (IAB), Nuremberg, Germany; Research Fellow, IZA Institute for the Study of Labor, Bonn, Germany.     .         , UIC Distinguished Professor; Research Professor and Head, Department of Economics; Director, UIC Center for Economics Education, University of Illinois at Chicago, IL, USA; Program Director, Migration Studies, IZA Institute for the Study of Labor, Bonn, Germany.         , Professor of Political Science; Associate Director, Institute of Governmental Studies, University of California at Berkeley, CA, USA.           , Senior Researcher, Institute for Migration and Ethnic Studies (IMES), University of Amsterdam, The Netherlands.         , Distinguished Scholar, Population Council, New York, NY, USA.       .      , Deputy Head, German Socio-Economic Panel Study (SOEP); Senior Research Associate, German Institute for Economic Research (DIW Berlin), Germany; Research Fellow, IZA Institute for the Study of Labor, Bonn, Germany.

xvi

List of contributors

xvii

              , Assistant Professor of Political Science, Boston College, MA, USA; Visiting Fellow, Center on the US and Europe, The Brookings Institution, Washington, DC, USA.          , PhD candidate, Department of Political Science, University of Washington, Seattle, WA, USA.            , Research Director, Belgian National Fund for Scientific Research (FNRS) and Director, Centre d’Etudes de l’Ethnicit´e et des Migrations; Institut des Sciences Humaines et Sociales, Universit´e de Liege, Belgium.         , Lecturer in Political Economy, Goldsmiths College, London, United Kingdom.             , PhD candidate, Department of Social Sciences, Roskilde University, Denmark.       .       , Assistant Professor of Political Science, University of Oregon, Eugene, OR, USA.     .        , Professor of Economics, University of Aarhus and The Danish National Institute of Social Research, Denmark; Research Fellow, IZA Institute for the Study of Labor, Bonn, Germany.             , PhD candidate, Department of Economics, Aarhus School of Business, Denmark; Member, Centre for Integration and Marginalization (CIM), Denmark.       .       , Professor of Politics; Director, Center for European Studies, New York University, NY, USA.                  , Research Officer, Southampton Statistical Sciences Research Institute, University of Southampton, United Kingdom; Research Affiliate, IZA Institute for the Study of Labor, Bonn, Germany.          , Assistant Professor of Political Washington University, Washington, DC, USA.

Science,

George

       .         , Maxwell Professor of Public Policy; Director, Center for Policy Research, Syracuse University, Syracuse, NY, USA; Director, Luxembourg Income Study.          , Professor of Economics, Aarhus School of Business, Denmark; Research Director, Centre for Integration and Marginalization (CIM), Denmark.

xviii

List of contributors

    .           , Lecturer in European Politics and Policy, Department of Government/European Institute, London School of Economics and Political Science, United Kingdom.               , Assistant Professor of Sociology, Department of Sociology, Interuniversity Center for Social Science Theory and Methodology (ICS), Utrecht University, The Netherlands; Research Fellow, European Research Centre on Migration and Ethnic Relations (Ercomer), Utrecht, The Netherlands.              , Emeritus Professor, University of Amsterdam, The Netherlands.    .      , Professor of Economics, Berlin University of Technology (TUB); Head, German Socio-Economic Panel Study (SOEP); Research Director, German Institute for Economic Research (DIW Berlin), Germany.

Acknowledgements

The editors want to thank the many people who made this volume possible. Funding for the project came from the European Commission, through its generous grant to the Maxwell European Union Center; from the Moynihan Institute for Global Affairs of the Maxwell School, through its matching contributions to the European Union Center that operates under its umbrella; from the Luxembourg Income Study (LIS); from the Luxembourg national science foundation, Le Fonds National de la Recherche (FNR); from the Luxembourg Ministry of the Economy and Foreign Trade; and from the Ford Foundation. On a more personal level, we are grateful especially to Kati Foley and Caroline de Tombeur of LIS; Maxwell European Union Center interns Jennifer Reynolds and Stella Samillan Aguilar; Center for Policy Research staff Martha Bonney, Kim Desmond, Karen Cimilluca and Kelly Bogart; Peg Hermann and Juanita Horan of the Moynihan Institute; Chris Harrison at Cambridge University Press; our authors; and also the many other contributors to our conference in Luxembourg in 2004, all of whom enriched the chapters that were selected for the volume.

xix

1

What’s unique about immigration in Europe? Craig A. Parsons and Timothy M. Smeeding

To begin A new kind of historic transformation is underway in Europe at the outset of the twenty-first century. Twentieth-century Europeans were no strangers to social, economic, and political change, but their major challenges focused mainly on the intra-European construction of stable, prosperous, capitalist democracies. While the extra-European world obviously affected the continent in many ways, the biggest problems turned on compromises within or between European societies (and with the most influential offshoot of European society, the United States). In many ways, the creation of a single currency for the European Union in 1999 marked a fitting conclusion to Europe’s inwardly-focused twentieth century. Today, by contrast, most Europeans perceive their main challenges as related to flows across their borders – flows of Europeans from other European Union nations (including the ten new partners from eastern and southern Europe), but particularly inflows of non-European people. Immigration and minority integration consistently occupy the headlines and loom over the political agenda, even playing some role in the French and Dutch rejections of the European Constitution in Spring 2005. Moreover, the issues that rival immigration for immediate political salience – unemployment, crime, terrorism – are often presented by politicians as its negative secondary effects. Immigration is also intimately tied to serious global economic pressures, the challenges of population ageing, and welfare-state reform. Both academic observers and the European public are increasingly convinced that Europe’s future will turn to a substantial degree on how they incorporate and integrate non-Europeans into European culture, customs and institutions. European’s new transition is not an isolated continental phenomenon, of course. Similar questions about immigration confront all industrialized societies. Partly this is because they are victims of their own success. Their populations are living longer and favoring individual pursuits over large families. They enjoy a wide range of welfare-state benefits, especially 1

2

Immigration and the Transformation of Europe

in old age. This combination is unsustainable whether in Europe, North America, Australia, or Japan: replenishment of the working-age population is now too slow to fund promised benefits for the swelling ranks of retirees, or even to maintain economic growth in the longer run. Most of the “new world” Anglo-Saxon nations (the United States, Australia, and Canada) have coped relatively well with immigration and its social and economic consequences (Antecol et al. 2003; Chiswick 1979, 1988). Nonetheless, the result is a discussion emerging across the industrialized world about the compatibility of increased immigration with other goals and values. Governments are caught between broad public-policy incentives to immigration, commitments to openness (at least in principle) in liberal constitutions and courts, and the considerable hesitation or fear that the prospect of large-scale immigration inspires (Joppke 2000). Europe’s version of this transition is unique, however, in several respects. The common challenges to industrialized societies vary in the severity and immediacy of demographic and pension-funding problems, in the size and integration of already-established immigrant minorities, and in the availability of political myths and institutional openings to legitimate immigration. On each score European difficulties are particularly acute. Some European societies are ageing as quickly as Japan (and much faster than the United States, Canada, or Australia), and their much more generous welfare states make the economic challenge especially pressing. Unlike Japan, European countries have already taken in substantial numbers of immigrants, so they face difficult problems of integration simultaneously with a debate over the need for more inflows. Indeed, Table 1.1 suggests that net population migration was already the dominant source of European population growth at the end of the last century. By 2003, European population (within the European Union twenty-five) was growing at a rate of only 3.8 persons per 1,000 inhabitants, or .38 percent. Of these 3.8 persons, 3.3 were due to net immigration and only .4 percent due to “natural” national population growth (the difference between births of 10.4 per 1,000 and deaths of 10.0 per 1,000). Thus on net, almost 90 percent of European population growth in these nations is due to immigration. In nine countries (including Germany and Italy) natural population growth was negative early in the new century, with immigration being the only gross and net source of population increase. In nineteen of twenty-five nations, immigration was the largest source of population growth. Only three countries in Europe show population growth of greater than 1 percent per year (10 per 1000): Cyprus (21.7), Spain (15.6), and Ireland (15.4). It is no great secret that these are among the strongest economies in the European Union. All three nations show not only large natural population growth, but also the largest

What’s unique about immigration in Europe?

3

Table 1.1 Population figures for Europe in 2003 (per 1,000 inhabitants)

Country

Births

Deaths

Natural population growth

Net migration

Total increase

EU 25 Austria Belgium Denmark Finland France Germany Greece Ireland Italy Luxembourg Netherlands Portugal Spain Sweden United Kingdom

10.4 9.5 10.8 12.0 10.9 12.7 8.7 9.5 15.5 9.5 11.8 12.4 10.8 10.5 11.1 11.7

10.0 9.6 10.3 10.7 9.4 9.2 10.4 9.5 7.3 10.0 9.1 8.7 10.4 9.2 10.4 10.3

0.4 0.0 0.5 1.3 1.5 3.5 −1.7 0.0 8.3 −0.5 2.9 3.7 0.4 1.3 0.7 1.4

3.3 4.0 3.4 1.3 1.1 0.9 1.8 3.2 7.1 8.9 4.7 0.2 6.1 14.3 3.2 1.7

3.8 4.0 3.9 2.6 2.6 4.5 0.0 3.2 15.4 8.4 7.4 3.9 6.5 15.6 3.9 3.1

Cyprus Czech Republic Estonia Hungary Latvia Lithuania Malta Poland Slovakia Slovenia

11.3 9.2 9.7 9.3 9.0 8.8 10.1 9.2 9.6 8.7

7.7 10.9 13.4 13.4 13.9 11.8 8.1 9.6 9.7 9.7

3.6 −1.7 −3.7 −4.1 −4.9 −3.0 2.3 −0.4 −0.1 −1.0

18.0 2.5 −0.3 1.5 −0.3 −1.8 4.3 −0.4 0.3 1.7

21.7 0.8 −4.0 −2.5 −5.3 −4.8 6.5 −0.7 0.2 0.7

Source: Eurostat (2004).

positive net migration flows. And since 2003, we expect that immigration has increased in importance in all of these nations. Were it not for immigration, very few European nations would exhibit positive population growth. If immigration has already been very important on the ground in Europe for some time, in a broader perspective immigration is all “new” to the “old” Europe. Unlike the United States, Canada, or Australia, European populations have never thought of themselves as “countries of immigration.” For many Europeans, to bolster social programs and economic growth through immigration could be to lose their nation and Europe as they know it. A further complication unique to Europe lies in its unprecedented “institutional growth project” at the regional level, as

4

Immigration and the Transformation of Europe

European states find their freedom to rework national compromises and policies limited by the cumbersome, complex, quasi-federal framework of the European Union (EU). This collection of papers aims to help students and scholars understand the nature of the immigration challenge in Europe, and how Europeans are beginning to grapple with the issues that arise in this process. Our goal is to offer an accessible and comprehensive insertion point into the vast literature on this subject. Like most scholarship, work on migration is largely balkanized into internal exchanges between sociologists, political scientists, demographers, economists, or historians. There are good reasons for this: the complex processes and consequences of migration cannot be studied in their full depth in all areas of inquiry at one time. And the various disciplines use different tools and terms to capture different elements of the subject matter. But such a division of labor has the often-lamented cost of fragmenting information and analysis. The cost is particularly heavy for students and for more practical, policy-oriented readers, who lack the time or expertise to trace and relate the many strands of academic thinking. Several recent books help to reduce this fragmentation in various ways. They offer interdisciplinary surveys of migration theories (Brettell and Hollifield 2000), broad overviews of migration by region (Massey et al. 1998), interdisciplinary perspectives on particular facets of migration like border-control policies or education (Guiraudon and Joppke 2001; Luchtenberg 2004), or exhaustive surveys of all migration-related research on a single country (Su´arez-Orozco et al. 2001). Our strategy, by contrast, is to assemble an empirical picture of migration-related trends in Europe by offering a series of focused disciplinary papers covering most of the relevant topics in the European arena. In order to accomplish this goal, we asked experts across the disciplines to choose the appropriate mix of pan-European data or crossnational comparisons to best display what their research agenda can tell us about immigration and integration in Europe. The disadvantage of this approach is that we may reproduce conceptual disciplinary divides, though several joint meetings helped prepare and focus more widely on the larger topics at hand. The advantages are that this one volume brings a variety of theoretical and conceptual tools to bear on Europe’s complex transition, and that it provides a point of access from which readers can enter more deeply into the many strands of the literature in classes or further reading. This introductory chapter outlines just how unique today’s European transition is in both a comparative and an historical perspective. At a broad level we chart quantitatively and qualitatively how recent

What’s unique about immigration in Europe?

5

immigration and minority populations in Europe compare to those in other industrialized nations and in Europe’s own past experience. Then we offer an overview of how the chapters provide more detailed cuts into immigration and public policy in the contemporary European landscape. What is special about recent European immigration? International migration is at an all-time high in absolute terms, with more than 200 million people living outside their country of nationality (United Nations 2002). Nearly one in ten residents of advanced industrialized states is an immigrant. Perhaps the most widely-read book on migration proclaims “a transnational revolution that is reshaping societies and politics around the globe” (Castles and Miller 2003). As might be expected, Europe is overrepresented as a host within this evolving transnational pattern, with fifteen million migrants among the 370 million inhabitants of the fifteen western European members of the European Union – roughly 8 percent of world migrants in an area with 6 percent of world population. Even classic sources of emigration like Ireland, Italy, or Spain are now major receiving countries; again, Spain, Italy, and Ireland have the three highest rates of inflow per 1,000 persons at the end of the twentieth century (excluding tiny Cyprus, see Table 1.1). From these figures it is a small step to the common wisdom that a rising wave of immigration into long-stable European societies has inaugurated a particularly difficult era of change. But we must begin by qualifying the notion that the challenges of immigration in Europe today flow from a simple quantitative rise in migration. The world migrant population in 1990 – a year of unusually high migration in Europe – was no larger as a percentage of global population than were migrants in 1965 (Zlotnick 1999). The proportion of foreign-born in most European countries is not very different from the eve of World War I, and is much lower than at many points in the nineteenth century (Zolberg and Long 1999). Coherent data is difficult to assemble for stocks and flows of migrants, since European countries categorize immigrants, foreign-born, and citizens in different ways (Lemaitre 2005). To the extent that migration into Europe did jump in recent decades, it did so in a relatively brief burst in the late 1980s and early 1990s that has since moderated, rather than as an inexorable upward trend (OECD 2004). Net inflow into European Union countries increased by a third from 1988 to 1996, but in recent years the overall stock of foreign-born has changed very little (Salt and Clark 2000; see also Figures 3.1 and 3.2 in chapter 3). Between 1989 and 1992, the Federal Republic of

6

Immigration and the Transformation of Europe

Germany (FRG) took in almost 3 million people – more than half of what the United States absorbed in the 1920s – but by the mid1990s, these numbers had fallen steeply and in 2004, less than 100,000 immigrated legally to the FRG. While we have seen a more continuous but modest rise in European immigration from poorer countries (Pederson et al. 2004), and southern European countries (Spain and Italy) are now the largest recipients of immigrants, the popular impression of hordes of prospective immigrants from the South and East is considerably exaggerated. Many people from poorer countries do want to get into Europe, but not as a migratory movement that is terribly striking in an historical perspective (Sassen 1999). If anti-immigration policies are not as effective as their proponents would like – with illegal immigrants into western Europe estimated at around 350,000 per year, as opposed to approximately 480,000 that enter the United States (Passel 2005) – they still stand as very substantial obstacles to an open continent. The salience of immigration in Europe today is not, then, a simple story of crisis-level inflows from a swelling sea of international migration. Instead, the sense of profound change reflects the novel ways in which recent immigration relates to the make-up of European societies and intersects with other major (but often quite distinct) trends. Perhaps the most obvious way in which the immigration of recent (and likely future) decades stands out from previous European experience is that it includes many more non-European, non-white, non-Christian people than ever before (notwithstanding traditional colonial migration into countries such as Britain, France, and the Netherlands). Integration of Africans and Asians is commonly perceived as even more difficult than the earlier acceptance of migrants from southern or eastern Europe. Yet experts often point out that earlier waves of Poles, Jews, Italians, or Portuguese confronted broadly similar perceptions of cultural difference, and arguably provoked similar levels of anxiety and conflict (Zolberg and Long 1999). Within two or three generations these immigrants typically came to be seen, and to see themselves, as well integrated. Economic studies suggest that after two or three generations, the children and grandchildren of most foreign-born assume the labor force and earnings patterns of natives (Chiswick and Hatton 2002). Still, it is certainly debatable just how much heterogeneity can be accommodated by liberal European societies, and academics disagree sharply on how much they see integrative policies as successful so far, vis-`avis extra-European immigrants (Brubaker 2001; Alba and Nee 2003; Joppke and Morawska 2003; Mitchell 2004). But it is plausible that if sheer cultural difference were the sole challenge, the integrative record

What’s unique about immigration in Europe?

7

of European states would support optimistic expectations for the newest arrivals. Another distinctive facet of postwar immigration points to greater pessimism. Recent waves of non-European immigration arguably represent more clearly unintended – and often explicitly undesired – processes than did earlier inflows. At no point did substantial groups or policy-makers in Europe intend for most of these postwar immigrants to settle permanently. This is not to say that earlier immigration in Europe or into other industrialized nations has often flowed from intentional, explicit calls for migrant settlement, but much recent immigration into Europe stands out as very much the reverse. Significant migration into postwar Europe began with labor migration in the 1950s, with male workers brought in on temporary contracts. These Gastarbeiter were seen not as prospective citizens but as filling a passing need in labor markets. By the 1970s, however, large numbers of workers had not returned home, and they increasingly brought in families to establish full-fledged communities in European cities. Far from reflecting pro-settlement public policies, it was non-majoritarian institutions – constitutional guarantees of human rights and courts – that protected this movement against restrictions by elected officials (Joppke 2000; Guiraudon 2000). Together with similar dynamics in the politically-salient realm of asylum-seekers, and with the gradual rise in illegal immigration, this background strengthens the grounds for democratic (or, more negatively, “populist”) challenges to integration and further immigration. Whatever their legal commitments to tolerance and social capacity for long-term integration, European people generally did not actively choose to let recent migrants in. In addition to this contested history, many recent and future European immigrants confront a particularly unhelpful global context since September 11, 2001 (and the Madrid bombings of March 11, 2004). In terms of their internal principles and institutions, European societies might be able to incorporate non-European minorities as they have earlier migrants – even given occasional populist outbursts to the contrary. How well they can do so with large numbers of Muslim immigrants in a global context of a Muslim-focused “war on terror” is another question altogether. The Muslim population in the European Union (Table 1.2) is now almost 15 million, with 6 million in France (roughly 10 percent of the population), 3 million in Germany (3.7 percent) and 1.5 million in the United Kingdom (2.5 percent). Their degree of integration varies widely, and these numbers include many third- or fourth-generation Muslims who feel more at home in Europe than anywhere else. But the context of conflict around Islamic

8

Immigration and the Transformation of Europe

Table 1.2 Rough estimates of Muslim population, by country

Country

Total population in 2003 (millions)

Muslim population in 2003 (millions)

Percent of population that is Muslim in 2003

EU-15 Austria Belgium Denmark Finland France Germany Greece Ireland Italy Luxembourg Netherlands Portugal Spain Sweden United Kingdom

380.40 8.20 10.40 5.40 5.20 59.80 82.60 11.00 4.00 57.20 0.50 16.20 10.40 41.30 9.00 59.20

14.49 0.18 0.37 0.16 0.01 5.98 3.06 0.17 0.00 1.37 0.01 0.87 0.05 0.50 0.28 1.48

3.81 2.23 3.60 3.02 0.18 10.00 3.70 1.50 0.01 2.40 1.10 5.40 0.50 1.20 3.10 2.50

Source: Islamicpopulation.com (2005), collected from a variety of open government and other sources.

fundamentalism means that many Europeans are less inclined to be welcoming to Muslims overall, and at least some Muslims may be less inclined to want to integrate (Joppke and Morawska 2003). Rising sentiment against Islam in the long-tolerant Netherlands, for example, where Muslims are now more than 5 percent of the population, suggests that security concerns involving only a tiny fraction of migrants can alter the entire atmosphere for integration. The general economic malaise of continental Europe adds another special challenge for recent and future immigrants. It is a historical coincidence that in the 1970s, just as postwar non-European immigrant communities were becoming visibly established, the postwar economic boom ended and most western European countries began to suffer a seemingly ineluctable rise of unemployment. In the 1980s, this coincidence translated into a core theme of rising anti-immigrant extremism, perhaps most famously displayed in Jean Marie Le Pen’s formula in France, “Three million immigrants equals three million unemployed.” When they are not stealing native jobs, say such critics, immigrants are draining welfare-state resources and contributing disproportionately to rising levels of crime. In fact, these claims find little support in economic research: there is no substantial “welfare magnet” effect in the distribution of migrants (Pederson et al. 2004). Immigrant crime rates mostly reflect their

What’s unique about immigration in Europe?

9

overrepresentation among the socio-economically disadvantaged (Tonry 1997). Yet such arguments remain politically resonant in a broad climate of economic insecurity. Europeans increasingly fear that their “European social model” is too expensive and uncompetitive in a “globalizing” world, and see non-European immigration as part of a set of external threats that ostensibly undercut the hallowed class compromises at the core of their societies, rather than as a benefit to the mounting welfare state costs of an ageing society. Relative to earlier European migrants, recent and future arrivals confront issues like low native-born birthrates, less supportive economic and geopolitical contexts, not to mention claims of greater cultural distinctiveness and explicitly undesired routes of arrival. All these difficulties are exacerbated by an important political difference between European societies and the other rich democracies that have received large inflows. The percentage of foreign-born in most European countries has risen toward the levels of the United States, Canada, or Australia. But each of the latter countries has long defined itself as a country of immigration. While such broad myths in no way rule out anti-immigrant public opinion, they make it possible to argue that further immigration – even on a large scale – need not unravel the national identity. Even in the face of fears about immigrants that are very similar to Europeans’ objections, these myths sustain rhetoric celebrating immigrants as a basic source of dynamism and entrepreneurialism in these societies. Such rhetorical space hardly exists in Europe. While European societies may not have as closed (or “ethnic”) national identities as Japan, they tend to stand much closer to that pole than to the open (or “civic”) identities of the New World and the Antipodes. German citizenship laws provide the most famous example. Until hotly contested changes in 2000, German citizenship operated by jus sanguinis (“law of blood,” or inheritance-based) principles rather than jus solis (“law of soil,” or residency-based) rules. Naturalization was basically limited to immigrants of German ancestry. Thus, Russified Germans of the Volga region – speaking no German but claiming German ancestry – could easily obtain citizenship after decades of living in the former Soviet Union, whereas third-generation Turkish immigrants remained foreigners. While the citizenship regime in France, Britain, and The Netherlands has long been more liberal – more civic than ethnic in a legal sense, with residency-based routes to naturalization – this is considerably less true of widespread perceptions of “Frenchness,” “Britishness,” or even “Dutchness.” Some progressive European politicians may argue that immigrants are beneficial, or that they should be welcomed in the name of human rights and decency, but they do not argue that new immigrants are the lifeblood of their national identities.

10

Immigration and the Transformation of Europe

Table 1.3 Foreign population by country, 2002

Country

Foreign Population1 (% of total population)

Austria Belgium Denmark Finland Francea Germany Greeceb Ireland Italy Luxembourg Netherlands Portugal Spain Sweden

8.8 8.2 4.9 2.0 5.6 8.9 7.0 5.6 2.6 38.1 4.3 4.0 3.1 5.3

United Kingdom Australia Canadab Japan United States

4.5 27.1 18.2 1.5 11.5

Source: OECD (2004). Notes: 1 Data for Australia and the US relate to the proporation of foreign-born persons in total population. a 1999. b 2001.

The points we have touched on so far dominate current public discussion of immigration in Europe. The perceived challenges are constructed around the number of immigrants, the ostensible difficulties of integrating non-European minorities in ethnically-based and religiously differentiated cultures, their relationship to unemployment, and their connection to terrorism and crime. In general scholarly work (including this volume) tries to move away from the hyperbole which tends to accompany each of these challenges, often driven by the self-serving alarmism of relatively extreme politicians. Rather than adding to this, we instead stress that these challenges are not the end of the story. It is the intersection of recent and prospective immigration with even less directly-related trends that place it so squarely at the nexus of European futures.

What’s unique about immigration in Europe?

11

The demographic and socio-economic challenge As our opening paragraphs stressed, the most striking background trends are demographic. By the late 1990s, European fertility rates had fallen to the point that in many countries, net immigration was the sole source of population growth (see Table 1.1). The implications are still far from widely recognized. Not so long ago Europeans were preoccupied with the prospect of global overpopulation, and it was not until the first years of the new millennium that the imminent ageing and absolute decline of their populations began to receive much attention. But by then the demographic curves for the first several decades of the millennium were set: to put it simply, the people necessary to keep the population steady or rising in coming decades have already not been born. Fertility rates vary somewhat across Europe in ways that display some sensitivity to public policies, for example, there are higher rates in France, where maternity and child care receives greater public support (Table 1.4). They also respond to economic growth, as displayed in the most rapidly growing economies and populations of Ireland and Luxembourg. But even the highest rates in Europe are below the “replacement” fertility rate of 2.1 children per woman, even when the higher birth rates of new immigrant families raise these fertility rates above what they would be in their absence (Bledsoe 2004). And even the most optimistic experts agree that public policy can only affect fertility rates marginally and then only in the very long term (Demeny 2004). Europe has to confront the reality that it has already chosen a shrinking future, even at current or higher rates of net population inflow. Of course there are some reasons to celebrate this development. Europe is very densely populated already, and its cities may become less compacted. Soaring housing prices may moderate in the long term, and labor markets should tighten. But this breathing room will come with very significant costs. Continued growth on every other continent means that Europe’s raw weight in the world will diminish (Figure 1.1). Europe (broadly defined) now has roughly the same population as Africa, twice the population of North America, and one-fifth of the population of Asia. Even the most generous projections see Europe losing 95 million people by 2050 (and some see a loss of more than 150 million). North America will stay at roughly 5 percent of the world population, while Europe will shrink to 7.1 percent of the total, about one-third of the population of Africa and a little more than one-ninth that of Asia (Figure 1.1).1 While a great many factors are more important than raw population in 1

World Population Prospects: The 2002 Revision. New York, United Nations.

12

Immigration and the Transformation of Europe

Table 1.4 Fertility

Country

Total fertility rate, 2000–2005∗

Austria Belgium Denmark Finland France Germany Greece Ireland Italy Luxembourg Netherlands Portugal Spain Sweden United Kingdom

1.39 1.66 1.75 1.72 1.87 1.32 1.25 1.94 1.28 1.73 1.72 1.47 1.27 1.64 1.66

Australia Canada Japan United States

1.75 1.51 1.33 2.04

Source: United Nations (2005). Note: ∗ Births per 1,000 population.

1950 (total population = 2,519 billion)

2000 (total population = 6,057 billion) 0.5 5.2 8.6

2050 (total population = 8,919 billion) 0.5

13.1

5.0

6.8 0.5 8.8 6.6

8.6

12.0

20.2

7.1 21.8 60.6 55.5

Africa Asia Europe Latin America and the Caribbean North America Oceania

58.5

Figure 1.1. Estimated and projected populations of the world, major areas: 1950, 2000, and 2050. Source: United Nations (2003). Note: The 2050 data are based upon medium fertility variants.

What’s unique about immigration in Europe?

13

Table 1.5 Labor force participation by older people (aged 55–64), 2003

Country

Older workers participation rates

Austria Belgium Denmark Finland France Germany Greece Ireland Italy Luxembourg Netherlands Portugal Spain Sweden United Kingdom EU-15 Average

31.7 28.5 63.1 54.1 39.5 43.1 43.2 50.5 31.5 27.9 44.9 53.4 62.9 72.5 57.5 47.0

Australia Canada Japan United States

52.2 56.6 65.8 62.4

Source: OECD (2005).

determining how peoples gain security and project power in the world, declining population can hardly help a continent that already complains of powerlessness. In the most pessimistic vision, falling population is the apotheosis of the relative decline of European power through the twentieth century (for example, Samuelson 2005). It is unsurprising then, that some Europeans perceive immigration vaguely as the leading edge of the takeover of an older, declining world by a younger, growing one (Bledsoe 2004). If massive relative and substantial absolute drops in population seem dramatic for Europe’s future, the same demographic curves provoke much more immediate political problems due to the age structure of European societies. Besides having fewer children, Europeans are living longer than ever. They also tend to retire earlier than people in other industrialized countries. Thus, the ratio of working-age people as to those that have retired has already declined dramatically in recent decades, and its fall is accelerating (see Tables 1.5 and 1.6). The massive political problem is that today’s European societies are built more than any

14

Immigration and the Transformation of Europe

Table 1.6 Indicators of population structure: Elderly dependency, 2004, 2025, 2050 Elderly Dependency Ratio (%)1 Country

2004

2025

2050

EU-15 Austria Belgium Denmark Finland France2 Germany Greece Ireland Italy Luxembourg Netherlands Portugal Spain Sweden United Kingdom

25.5 22.8 26.1 22.5 23.3 25.2 26.8 26.4 16.4 28.9 21.0 20.5 24.9 24.6 26.4 24.3

36.3 34.5 35.6 33.8 41.4 36.9 39.3 35.5 25.2 39.7 27.7 32.5 34.7 33.6 36.5 33.2

53.2 53.2 48.1 40.0 46.7 47.9 55.8 58.8 45.3 66.0 36.1 38.6 58.1 67.5 40.9 45.3

Source: Eurostat (2004). Notes: 1 Population aged 65 and more as a percentage of population aged between 15 and 64. 2 Data for France refer to metropolitan France.

others – historically or elsewhere in the world – on transfers between these two groups. Though they vary considerably in their provisions and modes of financing, European welfare states are the most generous in the world, and most are financed by direct generational transfers (so-called “pay-as-you-go” systems, as opposed to individual-account systems in which people set aside money for their own retirement). Such transfers effectively serve as a multiplier of the impact of societal ageing; the more generous the pension system, the earlier and more dramatic the fiscal and political consequences of ageing. Europe’s welfare programs were originally premised on worker-retiree ratios well above 5:1; today’s EU average ratio is roughly 4:1, and is projected to fall to 2:1 by 2040. Though welfare-state and particularly pension reform and rising health care costs have been major issues in most European countries for several years, awareness of this basic demographic challenge seems low. Relatively modest attempts at reform in France, Germany, and Italy have met with substantial resistance. German retirees, for example, are currently

What’s unique about immigration in Europe?

15

seeing their benefits partly de-indexed from cost-of-living changes and their pensions more heavily taxed as well – but no one perceives the reform process as close to finished (Feldstein 2005). The opposition tends to cast reform as an agenda imposed by “globalization” or “Americanization” from outside (or sometimes as “Europeanization” imposed by the European Union), in which the fundamental challenge to the European social model is a race-to-the-bottom competition on labor costs and social regulation in open markets. Supporters of reform tend to contribute to this view by invoking similar pressures – arguing first and foremost that welfare-state programs and social regulation must be altered for European firms and economies to remain competitive. As some of the more successful European cases suggest, however (like low-unemployment Denmark or The Netherlands), the incompatibility of generous welfare states with high economic openness is very debatable (Rodrik 1998; Garrett 2001; Hay 2000). Their incompatibility with a rapidly ageing population, on the other hand, is not debatable at all. The most direct challenge to the European model is not an economic race with other societies, but an unavoidable need to rework their basic internal social retirement pension and health care provisions amongst themselves. Immigration hovers on the edge of debates over welfare-state reform, since it seems to be the one obvious policy response, over and above some version of entitlement cuts and tax hikes. If the key problem is a declining worker-retiree ratio, Europe can let in more of the working-age people who are eager to immigrate. Immigrants tend to be young and to have higher fertility rates than native populations. Already most European countries have moved to render some immigration easier, though generally only for limited quotas of highly-skilled individuals in sectors with labor shortages. Unfortunately, the magnitude of the demographic shift is such that no reasonable immigration scenario could restore European age balances. Even to maintain its working age population – which would still mean a falling age ratio given swelling ranks of retirees – Germany would need to bring in 35 million immigrants by 2050 (see Table 1.7). Moreover, even importing such huge numbers would not entirely sidestep the welfare-state challenge. As Steve Camarota has pointed out in the United States context (and as Paul Demeny notes in his chapter), while immigrants’ relative youth and fertility rates mean that very large flows could help age balances, their low average socio-economic status means that they offer less of a solution for redistributive policies. Pension-system contributions are progressive in all European societies, so poor workers contribute the least and benefit disproportionately (Camarota 2005). In other words, immigrants soon become part of the very systems that they could be brought in to “save” (Boeri and Tabasso 2004). What Europe

270,000 237,000 7,000 204,000 6,000 20,000 33,000 760,000

EU-15 Big 4 EU France Germany Italy United Kingdom Other EU countries United States

4 3 4 2 42 3 8 0

1,588,000 1,093,000 109,000 487,000 372,000 125,000 495,000 359,000

6 5 16 2 62 6 15 0

Multiple of 1995 immigration

Source: Martin (2003). Note: ∗ Migrants necessary to maintain 1995 population ratio of people ages 15–64 to those ages 65 or older.

Immigration in 1995 (thousands)

Region/Country

To maintain 1995 working-age Multiple of 1995 population immigration (thousands)

13,480,000 8,884,000 1,792,000 3,630,000 2,268,000 1,194,000 4,596,000 11,851,000

To maintain population Support ratio∗ (thousands)

Average annual number of immigrants required

50 37 256 18 378 60 139 16

Multiple of 1995 immigration

Table 1.7 Immigration required to avoid population decline and maintain working-age population and support ratios, European Union, 2000–2050

What’s unique about immigration in Europe?

17

really needs is to bring in a vast number of rich, fertile people – and these are not on offer. Still, the argument that immigration cannot solve Europe’s demographic challenges actually just makes it all the more pressing. Immigration is no panacea, but it is one of the few obvious partial responses to large problems. The coming crisis is so severe that any policy that even partly mitigates problems of ageing and decline will become more attractive. This does not necessarily mean that Europeans will see it as sufficiently compelling to revolutionize the cultural make-up of their societies by letting in tens of millions of foreigners. If Germany did use immigration to maintain its working-age population, the jus sanguinis “Germans” would be a minority before 2050. We can confidently predict, however, that the debate over how much to open the doors will become more and more salient. Political disarray and constraints Any discussion of European responses to immigration must begin with the policy challenges discussed so far: managing immigration and integration in a context of cultural anxiety, economic malaise, global tension, and demographic shifts. But policy challenges rarely dictate public policies or attitudes directly. The arrangement of political institutions and organizations like political parties deeply shape how immigration and integration are perceived, the patterns of mobilization around them, and the availability of policy options. Moreover, this political obstacle course is not just a static filter for issues like immigration and integration, but has its own dynamics. Two broad trends in the European political landscape are particularly important to our subject, though neither bears any direct relationship to immigration per se. The first is the much-remarked dislocation of classic Right–Left political party competition in Europe since the seventies. Scholars disagree on why exactly national party systems across Europe have generally moved away from relatively clear battles of Right versus Left, with older mainstream parties being challenged by new movements of Greens, the far Right, or subnational regional identities. Some emphasize a change in the economic terrain itself, with a shift to “postindustrial” production and services breaking up established notions of class (Priore and Sable 1984). Others go further to portray a shift to “postmodern” values beyond the economic terrain of class coalitions (Inglehart 1990; Kitschelt 1994). Still others stress the increasing external constraints on economic policies in a context of economic interdependence, with real or

18

Immigration and the Transformation of Europe

perceived narrowing of policy options leaving little room for Right–Left debate (Frieden 1991; Friedman 2000). Another view privileges the mediatization and personalization of electoral competition, which has weakened classic party organizations and opened the door to new charismatic competitors (Swanson and Mancini 1996; Mazzoleni and Schultz 1999). Whoever is most right – and it is prudent to recognize some insight in all these arguments – they collectively tell a story in which European voters have increasingly separated from familiar political currents, become more disillusioned with politics, and created more complex, volatile electoral arenas. Attitudes, mobilization, and policy agendas on immigration depend heavily on who seizes the resultant opportunities in electoral competition. The strength of anti-immigration, anti-integration attitudes is commonly perceived as measured by the electoral strength of the far Right, which has increasingly championed such views across Europe since the 1980s. Yet sophisticated analyses of far Right parties emphasize that the simple prevalence of anti-immigrant attitudes does little to explain variation in their support (Mudde 1999). Their emergence requires organizational “space” in electoral competition that opens or closes with change in mainstream party support and strategies. A telling anecdotal example is the French presidential election of 2002, where the xenophobic Le Pen shocked Europe by coming in second in the first round of balloting. Observers generally agree that the most important factors that gave Le Pen 17 percent of the vote – and more importantly, that made this the second-highest score – were the dismal approval ratings of incumbent President Jacques Chirac and his Socialist challenger, Prime Minister Lionel Jospin, as well as the fragmentation of the Left vote across a number of minor candidates (Lewis-Beck 2004). Where far Right candidates do well, or where anti-immigrant positions achieve some impact on more mainstream parties, reflects an extremely complex conjuncture of immigrant presence, the “political geography” of constituencies, electoral rules, and party organization. Though the causal relations are very difficult to tease out, anti-immigration sentiment may thus be as much an effect of party system change as a cause. Policy responses to integration and future immigration will depend, then, on how they emerge from political landscapes that are themselves evolving. Such evolutions are presumably occurring across industrialized societies – which all share post-industrial economies, apparent attitudinal change, perceptions of increasing external constraints, and mediatization – but anti-immigrant mobilization has been a larger and more consistent factor in European politics than elsewhere. Arguably

What’s unique about immigration in Europe?

19

this is because party system change is interacting with other European specificities we have already highlighted: only in Europe have “nonimmigration” countries taken in large numbers of immigrants during times of economic weakness. A few other factors may help explain why anti-immigrant sentiment has surged more aggressively into the openings of party system change in Europe than elsewhere. Europe’s mostly proportional-representation electoral rules facilitate the rise of new parties (or, in France’s unique two-round system, encourage “protest votes” for extreme parties in the first round) (Lewis-Beck 2004). Continental Europe’s uniquely weak record of job creation since the 1970s has helped maintain immigrant scapegoating. More abstractly, postwar Europe’s uniquely progressive discourse on human rights and solidarity – enshrined in a variety of conventions and foreign-aid practices, and contrasting fairly sharply to the less “postmodern” discourses of the United States or Japan – sits uneasily with its mostly-ethnic identities. Extremist challengers find particularly fertile ground to charge mainstream parties with hypocrisy. The other crucial political backdrop to discussions of immigration is more obviously uniquely European: the rise of the European Union (EU). The quasi-federal European Union constrains national options in ways that industrialized countries elsewhere simply cannot quite imagine. Immigration was long kept outside the EU framework, since national governments saw it as too politically sensitive for extensive delegation of powers (or even for much intergovernmental coordination). By the early 1990s, however, the EU’s Single Market and Schengen accord on open borders made more extensive coordination of border policies unavoidable. The goal of abolition of internal controls on movement effectively created a common frontier. The result has been increasing acceptance of EU responsibilities in this area, but ongoing contestation of the technically necessary and politically desirable degree of “supranationality” (or centralization). The “big bang” Eastern enlargement of 2004 added new urgency to this debate, giving western EU members an interest in building up migrant-control capacities before transitional barriers expire and the common frontier reaches the Ukraine and the Black Sea. If the failure of the EU Constitution represents a major setback for advocates of centralization – it would have supranationalized immigration policy broadly, with majority voting and full roles for the European Commission, European Parliament, and European Court of Justice – the Member States agreed in late 2004 on a back-up plan that will supranationalize policies on illegal immigration and political asylum. For the moment, policies on legal immigration remain the preserve of national governments, but the debate continues.

20

Immigration and the Transformation of Europe

European Union authority in immigration is thus significant but remains mostly a factor for the future. Beyond its direct powers, however, many observers remark that the EU has already indirectly strengthened the focus of national-level policies on the exclusionary, security-themed aspects of immigration policies, further distancing them from debate on socio-economic and demographic trade-offs (Guiraudon 2005). Reasonably enough, EU-level discussions on migration tend to focus on each country’s concerns that other Member States will let in undesirables. An interesting story of bureaucratic politics has further encouraged this focus. Some of the early advocates of moving some immigration authority to the EU level were national interior ministries, who sought to side-step more pro-immigration competitors in national-level arenas by shifting border-control coordination to Brussels (where, for these issues, interior ministries would represent their governments). The EU’s supranational Commission and Parliament have tended to take more positive, less securitized views of immigration, but again, the governments have attempted to minimize their role so far. The overall result neatly displays what we mean by the effect of indirectly-related trends on European immigration. Growing EU influence in immigration has not been caused by antiimmigration demands. Instead it is an unintended by-product of European progress in economic and regulatory integration. Yet it has tended to increase the influence of anti-immigration actors in policy-making and the “securitization” of immigration discourse. In sum, recent European immigration stands out historically and comparatively above all as part of a conjuncture of change. The basic issues of substantial flows of people into Europe and of learning to live with new migrants are the least novel aspects of this conjuncture. They differ from earlier European experience to debatable degrees in numbers and cultural distinctiveness, and are substantial, but hardly unheard-of, in broader global patterns of migration. But when these substantial flows and stocks of migrants are set in the context of weak job creation and slow growth, dramatic demographic shifts, and increasing worldwide tensions with the largest immigrant religious group, Europe’s contemporary challenges of immigration and integration begin to look relatively unique in historical and comparative perspective. When we add that any successful response will involve at least some change to both core welfare-state compromises and the ethnic quality of national identities, these challenges look all the more unprecedented. Policy-makers will also have to construct such solutions in the context of fragmenting national party systems and evolving EU constraints. To arrive at democratic, stable solutions that preserve the prosperity and openness of European societies will require political leadership and innovation of the first order.

What’s unique about immigration in Europe?

21

Breaking down the European transformation: our contribution The previous section set the background for our authors’ contributions. The breadth of issues mentioned underscores that no single volume can develop all the facets of European immigration in concrete detail. Where other volumes have mapped this multidimensional object mainly by shifting to theoretical discussions or to very general empirics, we have chosen to offer a selection of more focused disciplinary cross-sections. Our goal is to provide a set of representative inroads for novices to develop expertise on the subject, and a set of overlapping foundations on which more familiar readers may build toward more synthetic views. Four sections organize the chapters. The first three focus mainly on charting Europe’s immigration-related challenges in several dimensions, focusing on demography, economics, and social integration. The last and largest section considers responses to immigration in European public opinion, party politics, and policies. Demography The first section draws a concrete baseline for the new European transition. The distinguished demographer Paul Demeny discusses Europe’s immigration challenges in light of fifty-year population projections. He finds that even with optimistic immigration rates, Europe’s population will decline by 2050 unless birth rates increase drastically. His conclusion – slightly different from the thrust of this introduction – is that conflict over immigration is more a distraction from Europe’s real challenges than it is a crucial part of the solution. Next, Peder Pedersen, Mariola Pytlikova, and Nina Smith chart recent patterns of immigration into twenty-seven OECD (Organization for Economic Cooperation and Development) countries, showing that they reflect “network” effects linking immigrant streams and cultural or linguistic ties more than straightforward economic selection of workers with desired skills or the attraction of generous welfare regimes. This first part of the book captures one of the most basic tensions in the current transition, between Europe’s “mathematical” demand for people and the hard-to-control, historically and culturally-shaped nature of the patterns which people follow on the supply side.2

2

A later qualitative chapter by Jacquie Andall traces the micro movements of a particular Ghanaian migrant group to lend micro credence to the macro data in chapter 3.

22

Immigration and the Transformation of Europe

Economics The paper by Alicia Adser`a and Barry Chiswick is a magisterial overview of recent European immigrants and their earnings. They use the 1994– 2000 waves of the European Community Household Panel to conduct a systematic analysis of the earnings of immigrants as compared to native workers on a EU country-specific basis. In particular, they test whether there is any systematic variation in the labor market performance of immigrants across gender with relation to duration in the destination, schooling, age at immigration, country of origin, or country of destination. And they find significant differences across EU destination nations. This overview is a natural starting place for understanding the rewards and penalties for immigrants of various types. Herbert Brucker, ¨ Joachim Frick, and Gert Wagner then provide more macro-oriented empirical evidence on who gains (and who gains the most) from immigration, including fiscal implications, and conclude with a discussion of the economically optimal immigration policies for contemporary Europe. The picture they draw is quite positive: while migrants get most of the benefits of immigration, there are potentially significant gains for host-country economies overall. Moreover, while theoretical models suggest that immigration may reduce the wages of native workers, empirical studies consistently show these effects to be minor or non-existent. Frank van Tubergen and Ann Morissens then qualify this positive image somewhat by focusing in on immigrants’ welfare in more detail. van Tubergen analyzes immigrants’ occupational status across seventeen countries, underscoring how aspects of their country of origin and destination interact to create varying dynamics of skill selection and occupational discrimination. Morissens probes the patterns and consequences of immigrant unemployment, examining how varying national welfare regimes and labor market conditions affect immigrants’ search for jobs and poverty rates. Migrants may capture most of the economic gains of immigration – improving in most cases on their pre-migration conditions – but they tend not to attain the same economic footing as native populations. This segues very directly to the next section. Even if both immigrants and host countries gain on the economic dimension from migration, enduring economic inequality combines with other dimensions of interaction to create serious challenges in social integration. Social integration Our forays into social integration begin with two chapters on education, which is widely seen as the critical terrain where integration ultimately

What’s unique about immigration in Europe?

23

succeeds or fails. Sylke Schnepf draws on an unusually wide range of performance measures to analyze just how well immigrant children do in school, and how much this varies across European countries. As native birthrates fall and as immigrant children and the children of first generation immigrants continue to populate elementary and secondary schools, these performance measures will be critical gauges of how well integration is working. The OECD (2005) argues that the period between 1998–2003 witnessed a large increase in foreigners coming to study in European institutions of higher learning. Maurice Crul and Hans Vermeulen report the conclusions of a comparative study of second-generation Turkish and Moroccan immigrants’ education and employment in cross-national context. They find a trade-off that varies across Europe: in countries with well-developed apprentice systems and vocational education, more immigrants are channeled into this system, leading to relatively good employment outcomes, but more limited educational achievement. The reverse is true where the institutional channels of apprenticeship are less developed. Next we turn to the critical issue of the integration of Islam into European societies. Jonathan Lawrence traces state attempts to manage the religious practices of growing Islamic minorities. Until the late 1980s, he notes, European governments saw geopolitical reasons to allow foreign states to dominate Islamic organizations on their territory. Then a collective realization of the size and permanence of their Muslim populations led to a series of parallel (if somewhat different) national attempts to extend pre-existing domestic religious settlements to Islam – with only partial success. For the moment, European Islam remains organized in a powerful set of transnational communities that national frameworks are struggling to reframe. On the one hand, the historical success of these same states in the earlier challenge of “domesticating” Christianity and Judaism may hold some hope for long-term incorporation. On the other hand, the drawn-out conflicts around earlier incorporations suggest that European states and their Muslims have a long, difficult road to travel. In a similar vein, Jacqueline Andall uses a case study of the Ghanaian diaspora to emphasize the transnational, often itinerant, nature of many migrant populations in Europe. The social integration of many immigrants is not simply a question of assimilating or compromising with an established permanent enclave in each European country. A substantial proportion of today’s European immigrants do not just migrate to one preferred destination and settle. Instead they often find themselves shunted away from their first-choice destination (or have no clear first choice), crossing borders several times as they follow diasporas,

24

Immigration and the Transformation of Europe

immigration-policy changes, and economic opportunities to a series of temporary locations. Andall’s image of repeated transnational movement around Europe adds an additional challenge to nationally-based programs for social integration, and implies that more successful integration may require deeper EU-level policies. Political and policy responses The book concludes with a selection of six chapters on how Europeans have responded to immigration in terms of political attitudes, mobilization, and policy. Marco Martiniello sets the historical stage with a rapid but comprehensive survey of European immigration and immigration policies since 1945. Again, the core story is one of early labor migration that led unexpectedly to the establishment of permanent settlement, and later to full-blown minority communities through reunification of families. European governments failed to appreciate the human dimension of migration, initially treating migrants as manipulatable cogs in an economic machine, and have been uneasily grappling with the political, social, and security-related consequences since the 1970s. This is followed by two chapters that stress the social and political complexity of opposition to immigration. The first addresses public opinion. Jack Citrin and John Sides offer a sophisticated statistical analysis of the location and sources of anti-immigration attitudes. They document that national demographic and economic conditions do not directly generate such attitudes, and even individual income, working class status, and unemployment have little direct effect. Instead Europeans became critical of immigration due to membership in particular groups, networks that communicate a negative outlook on immigrants, and psychological dispositions that led them to perceive outsiders as threats to national identity and economic welfare. Martin Schain then turns our attention to political party action on immigration. He uses a comparison of immigration politics in the United States, Britain, and France to highlight how the “political geography” of mobilization against immigration has played out differently on opposite sides of the Atlantic. In the United States, he stresses, a background of multiracial discourse limited the racialization of immigration debates. Just as important, the relatively large number of nationally important electoral districts with high immigrant populations facilitated a shift in the 1990s in which both major parties began treating immigrants as potential voters rather than as threats to voters. Both Britain and France (and other European countries) display the opposite dynamics, with some national variation: discourse on national identity calls attention to immigrants’

What’s unique about immigration in Europe?

25

race and culture, and electoral geography has provided incentives for more anti-immigration mobilization. Adam Luedtke tells a more institutional story of how the supranational institutions of the EU have haltingly gained influence over immigration issues. The salience and conflictive nature of immigration debates in Europe since the 1980s has only exacerbated national government reluctance to delegate powers in this area to Brussels. In the Maastricht Treaty of 1991, the Member States explicitly set immigration apart from the more supranationalized EU procedures. Over the course of the 1990s, the European Commission and European Court of Justice gradually weakened this firewall around immigration policy, even in areas with very little national support like the granting of rights to third-country nationals. Commission powers of agenda setting and European Court of Justice legal review seem likely to deliver still more supranational influence in the future, if they can overcome the hostility of the infamous “Polish plumber” who contributed to the French defeat of the EU constitution in May 2005. The chapter underscores that we cannot really understand the complexity of immigration to Europe without studying the frictions – sometimes healthy, sometimes unhealthy – between European and national law. No doubt these frictions will continue for generations to come. Georg Menz poses a question that has attracted much politicized speculation but little careful analysis: does immigration undermine the “European model” of social policies, and which of Europe’s different social regimes are best or worst able to handle immigration? He stresses that there is no simple answer: the impact of immigration depends on getting into institutional details that vary considerably from country to country. Some of the most generous welfare regimes are partly buffered against immigrant costs because full benefits depend on prior formal participation in the national labor market – and so they are partly protected by the high levels of immigrant unemployment discussed in the economic section. Immigration’s consequences for social policy also depends heavily on how policies interact with demographic trends – these vary substantially across national contexts. Eiko Thielemann, lastly, investigates just how much ability European governments have – individually or collectively at EU level – to actually implement the tough measures they have advocated to deter unwanted immigrants. He looks mainly at provisions on asylum, where EU Member States reacted to a massive wave of asylum seekers in the early 1990s by clamping down and attempting to cooperate to distribute cases more evenly amongst themselves. In a conclusion that ties back to the chapters by Andall, and Pedersen, Pytlikova, Smith, Thielemann

26

Immigration and the Transformation of Europe

finds that these policies have been fairly ineffective because the asylumseekers’ choices of host country reflect historical, reputational, and economic factors over which governments have little control. Conclusion Overall, these chapters further elaborate the challenge of Europe’s emerging transformation more than they showcase solutions. Yet solutions are hard to find at this juncture in European history. The political and policy responses to European immigration to date display distinct local, national, and European actions, with little sign of clear and broadly accepted choices on how Europeans mean to face their future. The permanent establishment of minorities through postwar labor immigration was basically unintended. Native populations have reacted to these in highly localized and socially-contextualized ways. Responses from national politicians have varied with their institutions and political geography. Their policies have generally been long on rhetoric and short on effective implementation. Meanwhile, EU institutional actors have capitalized on these transnational flows to attempt to broaden their responsibilities. Given that all European countries are still largely at a loss about how to incorporate their current minorities, it is unsurprising that they are slow to make forward-looking choices between the imminent demographic and welfare state demand for immigrants and the political, social, psychological, religious, and security-related difficulties of integrating even more extraEuropean people. Yet if Europeans do not confront these choices, seriously engaging the option of greater immigration and clarifying the decision-making responsibilities between national and European levels, the choices will effectively be made for them. Today it is far from clear what most Europeans prefer on their most pressing trade-off: the eventual evisceration of the core of the “European model” of society – welfare-state programs – versus large new inflows of immigrants. Positions on this dilemma are difficult to fashion without extensive public debate, since the trade-off is such a complicated one. Even large-scale immigration will only lessen, not entirely resolve, the mathematical need for welfare-state reform. More importantly, as we have seen repeatedly in the chapters above, immigration brings with it much more than a younger workforce. The current challenges of minority integration already rock the foundations of European identities, and greater inflows will hugely magnify these issues. Unfortunately, the complexity of this decision does not make it any less immediate. Europe’s low fertility rates and generous welfare states pose these questions more sharply than anywhere else in the industrialized world. Its

What’s unique about immigration in Europe?

27

combination of “nativist” national identities, established minorities, and quasi-federal government make political solutions particularly difficult to elaborate. But Europeans do not have the option of avoiding a historic transformation of their societies. The best they may be able to do, with luck, is to choose how to shape that transformation in a conscious, open, democratic way.  Alba, Richard D. and Victor Nee 2003 Remaking the American Mainstream: Assimilation and Contemporary Immigration. Cambridge, MA: Harvard University Press. Antecol, Heather, Robin Cohen and Stephen J. Trejo 2003 “Immigration Policy and the Skills of Immigrants to Australia, Canada, and the United States.” Journal of Human Resources 38(1) (Winter): 192–218. Bledsoe, Caroline H. 2004 “Reproduction at the Margins: Migration and Legitimacy in the New Europe.” Special Collection 3, Article 4. Rostock, Germany: Max Planck Institute for Demographic Research. April 17. www.demographic-research.org/special/3/4/s3-4.pdf. Boeri, T., F. Fasani and D. Tabasso 2004 “Immigrazione e assistenza sociale,” in M. C. Guerra and A. Zanardi (eds) La finanza pubblica italiana – Rapporto 2004. Il Mulino, Bologna. Brettell, Caroline and James F. Hollifield 2000 Migration Theory: Talking across Disciplines. New York, London: Routledge. Brubaker, Rogers 2001 “The Return of Assimilation? Changing Perspectives on Immigration and Its Sequels in France, Germany, and the United States.” Ethnic and Racial Studies 24(4): 531–48. Camarota, Steven A. 2005 “Immigration in an Aging Society: Workers, Birth Rates, and Social Security.” Backgrounder. Washington, DC: Center for Immigration Studies. April. www.cis.org/articles/2005/back505.pdf. Castles, Stephen and Mark J. Miller 2003 The Age of Migration: International Population Movements in the Modern World. 3rd edn. New York: Guilford Press. Chiswick, Barry R. 1979 “The Economic Progress of Immigrants: Some Apparently Universal Patterns” in Contemporary Economic Problems, 1979, William John Fellner (ed). Washington, DC: American Enterprise Institute, 357–99. Chiswick, Barry R. 1988 “Immigration Policy, Source Countries and Immigrant Skills: Australia, Canada and the United States,” in The Economics of Immigration, Lyle Baker and Paul W. Miller (eds). Canberra: Australian Government Printing Service, 163–206. Chiswick, Barry R. and Timothy J. Hatton 2002 “International Migration and the Integration of Labor Markets.” IZA Discussion Paper No 559. Bonn: Institute for the Study of Labor. August. ftp://ftp.iza.org/dps/dp559.pdf. Demeny, Paul 2004 “Closing Remarks” presented at the conference Immigration in a Cross-National Context: What Are the Implications for Europe? Bourglinster, Luxembourg, June 21–22. Eurostat 2004 “European Demography in 2003: EU25 Population up by 0.4% to reach 456 million.” STAT/04/105. Luxembourg: Eurostat, August 31.

28

Immigration and the Transformation of Europe

Feldstein, Martin 2005 “Rethinking Social Insurance.” NBER Working Paper No. 11250. Cambridge, MA: National Bureau of Economic Research, March. Frieden, Jeffry A. 1991 “Invested Interests: The Politics of National Economic Policies in a World of Global Finance.” International Organization 45 (Autumn): 425–51. Friedman, Thomas L. 2000 The Lexus and the Olive Tree: Understanding Globalization. New York: Anchor Books. Garrett, Geoffrey 2001 “Globalization and Government Spending Around the World.” Studies in Comparative International Development 35(4): 3–29. Guiraudon, Virginie 2000 Les politiques d’immigration en Europe. Paris: L’Harmattan. Guiraudon, Virginie 2005 “Drawing the EU’s Borders: Immigration Policy” in The State of the European Union: with US or against US?, Vol 7, Craig Parsons and Nicolas Jabko (eds). Oxford: Oxford University Press, 325–44. Guiraudon, Virginie and Christian Joppke (eds) 2001 Controlling a New Migration World. New York: Routledge. Hay, Colin 2000 “Contradictions of Capitalism: Contemporary Capitalism, Globalization, Regionalization and the Persistence of National Variation.” Review of International Studies 26(4): 609–32. Inglehart, R. 1990 Culture Shift in Advanced Industrial Society. Princeton, NJ: Princeton University Press. Islamicpopulation.com 2005. “European Muslim Population.” Accessed June 27, 2005. Available from www.islamicpopulation.com/europe general.html. Joppke, Christian 2000 Immigration and the Nation-State: The United States, Germany, and Great Britain. Oxford: Oxford University Press. Joppke, Christian and Ewa T. Morawska (eds) 2003 Toward Assimilation and Citizenship: Immigrants in Liberal Nation-States. New York: Palgrave Macmillan. Kitschelt, Herbert 1994 The Transformation of European Social Democracy. New York: Cambridge University Press. Lemaitre, Georges 2005 “The Comparability of International Migration Statistics.” Statistics Brief No 9. Paris: Organisation for Economic Co-operation and Development. July. www.oecd.org/dataoecd/4/41/35082073.pdf. Lewis-Beck, Michael S. (ed) 2004 The French Voter: Before and after the 2002 Elections. New York: Palgrave Macmillan. Luchtenberg Sigrid (ed) 2004 Migration, Education and Change. New York: Routledge. Martin, Philip L. 2003 Sustainable Migration Policies in a Globalizing World. Geneva: International Labour Organization (International Institute for Labour Studies). www.ilo.org/public/english/bureau/inst/download/ migration.pdf. Massey, Douglas S., and Committee on South–North Migration of the International Union for the Scientific Study of Population 1998 Worlds in Motion: Understanding International Migration at the End of the Millennium. New York: Oxford University Press. Mazzoleni, Gianpietro, and Winfried Schultz 1999 “‘Mediatization’ of Politics: A Challenge for Democracy.” Political Communication 16(247): 261.

What’s unique about immigration in Europe?

29

Mitchell, Katharyne 2004 “Geographies of Identity: Multiculturalism Unplugged.” Progress in Human Geography 28(5) (October): 641–51. Mudde, Cas 1999 “The Single-Issue Party Thesis: Extreme Right Parties and the Immigration Issue.” West European Politics 22(3): 182–97. OECD 2004 OECD in Figures: 2004 Edition. 012004071E1. Paris: OECD. OECD 2005 Labour Force Statistics 1983–2003. Paris: OECD. Passel, Jeffrey S. 2005 Unauthorized Migrants: Numbers and Characteristics. Washington, DC: Pew Hispanic Center. http://pewhispanic.org/files/reports/ 46.pdf. Pedersen, Peder J., Mariola Pytlikova and Nina Smith 2004 “Selection or Network Effects? Migration Flows into 27 OECD Countries, 1990–2000.” IZA Discussion Paper No 1104. Bonn: Institute for the Study of Labor. April. ftp://ftp.iza.org/dps/dp1104.pdf. Priore, Michael J. and Charles F. Sable 1984 The Second Industrial Divide: Possibilities for Prosperity. New York: Basic Books. Rodrik, Dani 1998 “Why do more open economies have Bigger Governments?” Journal of Political Economy 106(5) (October) 997. Salt, J. and J. Clarke 2000 “International Migration in the UNECE Region: Patterns, Trends, Policies.” International Social Science Journal 165 (September): 313–28. Samuelson, Robert J. 2005 “The End of Europe.” The Washington Post, June 15, A25. Sassen, Saskia 1999 Guests and Aliens. New York: New Press. SOPEMI: Systeme d’Observation Permanente sur les Migrations 2005 Trends in International Migration: 2004 edition. Paris: OECD. Su´arez-Orozco, Marcelo M., Carola Su´arez-Orozco and Desiree Qin-Hilliard 2001 Interdisciplinary Perspectives on the New Immigration. New York: Routledge. Swanson, David L. and Paolo Mancini 1996 Politics, Media and Modern Democracy: An International Study of Innovations in Electoral Campaigning and their Consequences. Westport, CT: Praeger. Tonry, Michael H. 1997 Ethnicity, Crime and Immigration: Comparative and Crossnational Perspectives. Chicago, IL: University of Chicago Press. United Nations 2002 “International Migration 2002.” New York: Population Division, Department of Economic and Social Affairs, United Nations. United Nations 2003 World Population Prospects: The 2002 Revision. New York: United Nations. United Nations 2005 “Social Indicators: Indicators on Child-Bearing.” New York. Accessed June 30, 2005. Available from http://unstats.un.org/unsd/ demographic/products/socind/childbr.htm. Zlotnik, Hania 1999 “Trends of International Migration since 1965: What Existing Data Reveal.” International Migration 37(1): 21–61. Zolberg, Aristide R. and Long Litt Woon 1999 “Why Islam Is Like Spanish: Cultural Incorporation in Europe and the United States.” Politics and Society 27(1) (March): 5–38.

2

Europe’s immigration challenge in demographic perspective Paul Demeny

On January 29, 2004, United Nations Secretary-General Kofi Annan went to the European Parliament in Brussels to accept an award – the Andrei Sakharov Prize for Freedom of Thought. In his acceptance speech he chose to address a topic that his September 2002 report to the United Nations General Assembly identified as a priority issue for the international community: migration. The topic was in the news. A month earlier, with the prompting of the United Nations, a Global Commission on International Migration was established, one of the multiplying recent initiatives that signal an intent to qualify the sovereign right of each nation to make unilateral decisions about immigration. A shorter version of Annan’s speech to the EU Parliament received wide distribution; identical texts appeared in leading newspapers in virtually all European countries. Annan’s opening sentences set out the thesis somewhat ambiguously: “One of the biggest tests for the enlarged European Union, in the years and decades to come, will be how it manages the challenge of immigration. If European societies rise to this challenge, immigration will enrich and strengthen them. If they fail to do so, the result may be declining living standards and social division.” But the Secretary-General went on to explain that the challenge is to adopt policies that accommodate greater immigration: “There can be no doubt that European societies need immigrants. Europeans are living longer and having fewer children. Without immigration . . . jobs would go unfilled and services undelivered, as economies shrink and societies stagnate. Immigration alone will not solve these problems, but it is an essential part of any solution.” And the concluding paragraph of the address wraps up the gist of the argument: “In this twenty-first century, migrants need Europe. But Europe also need migrants. A closed Europe would be a meaner, poorer, weaker, older Europe. An open Europe will be a fairer, richer, stronger, younger Europe – provided Europe manages immigration well” (Annan 2004). 30

Europe’s immigration challenge

31

This speech echoed propositions that in the past decade became common currency in European discussions on the demographic destiny of the old continent. Indeed, they reflect views dominant among European political and economic elites. But the assertion “there can be no doubt” is unwarranted. Much can be said in support of the claims as articulated by Annan and, before him, by many Europeans. Equally, there are counterarguments that weaken those claims or even contradict them outright. Issues of international migration are complicated because the benefits and costs – economic, social, and political – that immigration imparts on a receiving population are multifarious and are unevenly distributed. In what follows, I expand on the brief mention of these issues in the editors’ introduction by examining some demographic patterns in a bit more detail. Dynamic but prudent Europe Europe’s role as the vanguard in modern economic development and that continent’s demographic uniqueness underlie the complexity inherent in the issue of immigration in Europe. By virtue of its social institutions combined with auspicious geography, Europe led the way toward industrialization and higher material standards of living. It was also the pioneer in demographic change – in the transition from a quasi-equilibrium in which high death rates balanced high birth rates, to a prospective equilibrium of low death rates and low birth rates. In that process, death rates invariably fall ahead of birth rates, hence population growth accelerates. Many contemporary examples show how extraordinarily rapid such transitional population growth can be. The Philippines had a population of 7 million people in 1900. In 2005, despite significant out-migration, the population count is 83 million – a twelve-fold increase within the span of a single century. The demographic growth pattern of industrializing Europe was, by the standards of its earlier history, dynamic but it never came close to exhibiting expansion of a speed commonly found in the contemporary less-developed world. The continent’s population was 180 million in 1800 and some 390 million in 1900 (McEvedy and Jones 1978). In 2005 it is estimated as 728 million.1 Thus, over the course of two centuries, Europe’s population has doubled and doubled again. Europe became a populous and relatively densely settled land. 1

For population estimates post-1950 and for projections up to 2050, I draw on data prepared by the United Nations Population Division (United Nations 2005).

32

Immigration and the Transformation of Europe

Demographic expansion was no doubt a stimulus for the continent’s economic development. Its relatively moderate pace – well below 1 percent per year on average over the stretch of two centuries – has, however, prevented the build-up of numbers comparable to the world’s demographic giants, China and India. This demographic restraint, spontaneously achieved, is an important factor that helps to explain contemporary Europe’s unprecedented material prosperity. For that prosperity, today’s Europeans can thank the wisdom of their forbears in keeping birth rates prudently restrained. Europe’s demographic marginalization Prudence does not bring benefits only. A consequence of Europe leading the way toward demographic stability has been a marked drop of that continent’s relative demographic weight within the world’s total, within world regions, and, with few exceptions, country by country. Europe’s share of the global population at the middle of the twentieth century was 22 percent: 547 million out of 2.52 billion. This was already a few percentage points below Europe’s all-time peak in the global population share, reached a few decades earlier. Population growth in Europe between 1950 and 2005 was substantial, amounting to some 180 million. This growth resulted mainly from “natural increase” – the demographers’ term for the surplus of the number of births over the number of deaths – but was also enhanced by immigration. Population increase in the rest of the world, and in particular in the less-developed regions was, however, much faster. Thus, by 2005, Europe’s population share in the global total was halved. Its 728 million population was only 11 percent of a world population of 6.46 billion. Inevitably, Europe’s demographic marginalization will continue in the decades to come. This is the consequence of the time lag between the decline of fertility in Europe and that decline in the rest of the world. Fertility decline is the main driving force in the demographic transition. The stylized model of the transition envisages a convergence to a stationary population as fertility is lowered to a total fertility rate (TFR) of approximately 2.1 children per woman. In combination with low mortality, such a TFR produces zero population growth. Populations experiencing early fertility decline in comparison to other populations will have an earlier deceleration of their growth rate, hence will see their size diminish relative to other populations. In the European experience, an experience that may foreshadow a generalized global tendency, this effect is accentuated by the fact that the drop in fertility did not stop at replacement level: it overshot that threshold (Demeny 1997). In Europe at large the TFR was

Europe’s immigration challenge

33

already below 2 in the 1970s and declined steadily in the 1980s and early 1990s. Its estimated average value between 1995 and 2005 was 1.4. This is a level without precedent in a large population, with enduring effects on population dynamics and population age structure. What will be Europe’s demographic future in the coming decades? The UN “medium” population projections envisage, rather optimistically, a recovery of the average European fertility in the next 45 years. They assume, specifically, that by the middle of the twenty-first century the TFR will reach 1.85 – still below replacement level but substantially above its present average value. The projections also assume a sizable, although, in comparison with recent flows, somewhat diminished, net immigration to Europe from lands outside that continent. The estimated level of net immigration for the period 1995–2005 is 11 million; for the period 2005– 2050 the assumed net inflow is 32 million. Expected future changes in mortality also have a positive effect on population size as expectation of life at birth rises from 74 years (the current average level) to about 81 years by 2050. The combined effect of these three factors – fertility, immigration, and mortality – would yield a 2050 population of 653 million, that is, some 75 million smaller than its current 2005 level. The world as a whole, in contrast, is bound to experience substantial further population growth in the first half of the present century. Even though average world TFR by 2050 is assumed to sink to replacement level from its current value of 2.6, the momentum of growth would bring the global total to 9.1 billion. Thus, by mid-century, Europe’s share would be approximately 7 percent – one-third of what it was 100 years earlier. Other simple comparisons can also usefully illustrate this rapid relative shift in population magnitudes. Kofi Annan’s strictures, cited above, were primarily addressed to the European Union. The 1950 population of the 25 countries that today make up the EU was 350 million. India’s population in the same year was roughly the same: 358 million. In the next 55 years, India added 745 million people to its population while the EU-25 added just one-seventh of that number, 105 million. In the same 55-year period, China’s population grew by 811 million – from 505 million in 1950 to 1,316 million in 2005. The UN’s medium projections for 2050 are 431 million for the EU-25, 1,593 million for India, and 1,392 million for China. In absolute terms, the projected 45-year change represents a decrease by 24 million in the EU-25 and a gain of 76 million for China and 490 million for India. A comparison between the population dynamics of the EU-25 and of sub-Saharan Africa is even more striking. In the course of a single century – a period long in individual human terms, but short by historical standards – relative population sizes diverge rapidly. In 1950, the EU-25’s

34

Immigration and the Transformation of Europe

Table 2.1 Population (millions) of the eight most populous countries (as of 2005) of the EU-25 and of the Broader Middle East, 1950 and 2005, and projected population in 2050 1950 Germany France United Kingdom Italy Spain Poland Netherlands Greece Total top eight 2005 Germany France United Kingdom Italy Spain Poland Netherlands Greece Total top eight 2050 Germany France United Kingdom Italy Spain Poland Netherlands Greece Total top eight

68.4 41.8 49.8 47.1 28.0 24.8 10.1 7.6 277.6 82.7 60.5 59.7 58.1 43.1 38.5 16.3 11.1 371.0 79 63 67 51 43 32 17 11 363

Egypt Turkey Iran Algeria Morocco Iraq Saudi Arabia Yemen

21.8 21.5 16.9 8.8 9.0 5.3 3.2 4.3

Total top eight

90.8

Egypt Turkey Iran Algeria Morocco Iraq Saudi Arabia Yemen

74.0 73.2 69.5 32.9 31.5 28.8 24.6 21.0

Total top eight

333.5

Egypt Turkey Iran Algeria Morocco Iraq Saudi Arabia Yemen

126 101 102 50 46 64 50 60

Total top eight

599

Source: United Nations 2005. Projected figures from the “medium” variant.

350 million was nearly double sub-Saharan Africa’s 180 million. By 2005 the relative magnitudes show almost the reverse: 455 million for the EU and 751 million for sub-Saharan Africa. And by 2050 the proportions approximate 1:4 – 431 million in the EU and 1.7 billion in sub-Saharan Africa. Broad regional aggregates tend to obscure differences at the country level. As a final illustration of contrasting patterns of population growth, Table 2.1 presents data for individual EU countries and compares them

Europe’s immigration challenge

35

with countries in Europe’s southern hinterland: countries in North Africa and in south-west Asia – an area which for brevity’s sake has been labeled the Broader Middle East (BME). To keep the comparisons relatively simple, the table includes only the eight largest countries in each of the two groups, ranking within each group being determined by population size in 2005. In 1950, six of the EU’s top eight countries exceeded in population size the largest country in the BME, Egypt. Germany was more than three times the size of Egypt; France, the UK, and Italy each about twice the size. The combined population of the top eight EU countries was 278 million, or more than triple the corresponding group total in the BME. By 2005 the relative magnitudes are sharply different. Three of the BME countries – Egypt, Turkey, and Iran – are more populous than any of the EU countries with the exception of Germany. The group totals still show an edge for the top eight EU countries: 371 million versus 334 million. As to the future, the UN projections, as was implied above, assume recovery of EU fertility to a level of a TFR of 1.85 by 2050. For the BME countries the governing assumption is decline of fertility to replacement level by mid-century. The results are shown in the bottom panel of Table 2.1. In 2050, the three most populous countries of the BME surpass Germany – the EU’s largest country – by a wide margin. Algeria’s population was less than one-fifth of Italy’s in 1950; in 2050 the two countries are projected to be of equal size. And in that year the projected population totals of the top eight BME countries exceed the corresponding EU number by some 236 million – more than the population size of Germany, France, and the United Kingdom combined. Table 2.2 summarizes these shifts in terms of percentage changes during the two periods: 1950–2005 and 2005–2050, and also shows the percentage changes for the 100-year span 1950–2050. From 1950 to 2005 the percentage change in the EU countries ranges from 20 percent (for the UK) to 61 percent (for The Netherlands). The corresponding range in the BME is from 239 percent (Egypt) and 669 percent (Saudi Arabia). In the next 45 years, five of the top eight EU countries are projected to lose population, ranging from –4 percent in Greece to –17 percent in Poland. The group as a whole also exhibits negative growth: −2 percent between 2005 and 2050. In the BME countries, the slowest growth is projected for Turkey: 38 percent over the 45-year span. Yemen, at the other end of the ranking, would register an increase of 183 percent. The combined total growth is much slower than in the second half of the twentieth century yet it is still substantial: 71 percent in the next 45 years. The 100-year EU-BME contrast is correspondingly spectacular. The change ranges from 8 percent (Italy) to 70 percent (Netherlands) in the EU group and between 371 percent (Turkey) to 1,445 percent (Saudi

36

Immigration and the Transformation of Europe

Table 2.2 Population change (percent) in the eight most populous countries of the EU-25 and of the Broader Middle East, 1950–2005, and projected population change, 2005–2050 and 1950–2050 1950–2005 Germany France United Kingdom Italy Spain Poland Netherlands Greece Total top eight 2005–2050 Germany France United Kingdom Italy Spain Poland Netherlands Greece

21 45 20 26 54 55 61 46

Egypt Turkey Iran Algeria Morocco Iraq Saudi Arabia Yemen

239 241 311 274 250 443 669 388

34

Total top eight

286

Egypt Turkey Iran Algeria Morocco Iraq Saudi Arabia Yemen

70 38 47 51 47 121 101 183

−5 4 12 −12 −1 −17 5 −4

Total top eight 1950–2050

−2

Total top eight

Germany France United Kingdom Italy Spain Poland Netherlands Greece

15 51 35 8 52 29 70 42

Egypt Turkey Iran Algeria Morocco Iraq Saudi Arabia Yemen

Total top eight

31

Total top eight

71 478 371 503 466 418 1093 1445 1278 560

Source: United Nations 2005. Projected figures from the “medium” variant.

Arabia). In combination the eight EU countries grow by 31 percent over the 100-year span. The population of the BME countries grows nearly seven-fold. Ageing Europe Total population figures, it should be noted, conceal economically and socially significant differences in the age structure of the population. Population ageing, that may be characterized by such indicators as the

Europe’s immigration challenge

37

support ratio – the ratio of the number of persons in some meaningfully demarcated segment of the population considered as “old,” hence economically inactive and dependent, to the number of persons considered as potentially economically active. While modern industrial societies should have broad latitude in successfully adjusting to distortions in the age structure of the population, notably to population ageing or to declines in the number of persons of the labor force age, such adjustments, if the distortions are large, are apt to be costly. Comparison of the EU population to the population of the BME with respect to age structure would indicate substantial economic advantages inherent in the relatively youthful population still prevailing in the countries of the BME. These differences amplify the contrasts reflected in aggregate population size alone. For example, those who will be between the ages of 15 and 24 in 2020 are already alive, and hence can be projected with high accuracy. The size of that age group is of special interest for a variety of reasons – economic, military, political, and reproductive. The relative size of this age group shows greater contrast between EU countries and BME countries than do total population figures. Thus, in 2020, Egypt will have 16.7 million persons in the age group 15–24, in contrast to Germany’s 8.3 million. The corresponding figures for Turkey and France are 14.3 and 7.5 million, respectively. Immigrant-receiving Europe In assessing the contrasting projected population dynamics in the EU and in the BME countries, described in Tables 2.1 and 2.2, it should be kept in mind that both groups of countries are affected by international migration. In the last half-century, the EU grew faster because of an influx of migrants. The BME countries’ growth was, in turn, somewhat moderated by out-migration. As to the future, the projections assume fairly substantial net immigration in each of the EU countries. The sole exception is Poland, which is expected to experience a modest level of net out-migration. In contrast, seven of the BME countries are assumed to be net senders of international migrants. The sole exception is Saudi Arabia, which is a country of net immigration. The relevant figures are shown in Table 2.3. In the absence of these assumed migratory flows, the EU–BME contrast in the projected population sizes for 2050 and in the projected population changes between 2005 and 2050 would, of course, be wider. In particular, the EU countries, again with the exception of Poland, would have, ceteris paribus, smaller populations and the BME countries, again with the exception of Saudi Arabia, would have larger populations. Thus, for example, Germany is assumed to take in 9.1 million migrants, net, during

38

Immigration and the Transformation of Europe

Table 2.3 Assumed net international migration: most populous countries of the EU-25 and of the Broader Middle East, 2005–2050 Germany France United Kingdom Italy Spain Poland Netherlands Greece

9.1 2.7 5.9 5.4 3.0 −0.7 1.4 1.6

Egypt Turkey Iran Algeria Morocco Iraq Saudi Arabia Yemen

Total top eight

28.4

Total top eight

−3.6 −2.1 −1.4 −1.1 −2.8 −0.1 1.4 −0.9 −10.6

Source: United Nations 2005.

the next 45 years and the UK some 5.9 million. Even The Netherlands, one of the most densely populated countries in the world, is assumed to take in 1.4 million migrants. The assumed migrant flows, in addition to the sheer number of migrants, have further repercussions on population size. The migrants tend to be relatively young and, after their arrival, generate births additional to those accruing to the native population. Although the projections somewhat unrealistically assume that immigrants instantaneously conform to the fertility behavior of the native population, the amplification of the demographic impact of immigration through second-generation births is appreciable. Thus, for example, Germany’s projected population in 2050 – assuming 9.1 million net immigration during the next 45 years – is 78.8 million, while a zero net migration assumption would yield a 2050 population of 65.6 million. The difference between these two projected figures is 13.2 million, substantially larger than the 9.1 million figure for net migration. There is a multiplier effect: the 9.1 million migrants increase the 2050 population by 1.45 times 9.1 million. In similar fashion, the calculations show that the 5.85 million net migrants into the UK would increase the 2050 UK population by 9.8 million, or 1.67 times the actual number of net immigrants. And for The Netherlands, the assumed 1.35 million net immigrants between 2005 and 2050 would account for a 2050 population that is 1.9 million larger than would be the case in the absence of immigration. Lessons from Europe’s demographic shrinkage What are the lessons intimated by this brief account of shifting relative population size between Europe and the rest of the world, or, in particular,

Europe’s immigration challenge

39

between the EU and its Asian and African neighborhood? Two major implications may be suggested. First, Europe’s continuing demographic marginalization is inevitable. If that is seen to be a problem, immigration is not a solution for it. Europe cannot, and arguably ought not, engage in a demographic race with India, China, and the rest of Asia, or with the Broader Middle East, let alone with sub-Saharan Africa. No plausible internal demographic revival could appreciably narrow the size differences that now exist between Europe and its neighbors, whether close or distant, or the size and structural differences that will inevitably continue to widen during the coming decades. The same is true for immigration. Given the large differences in average income levels (and in other social and political amenities inadequately measured by income) between the EU on the one hand, and much of the rest of the world on the other, a potentially massive inflow of migrants into the EU is a theoretical possibility. Indeed, it is virtually certain that in the absence of the current barriers to immigration maintained by the EU, future inflows would greatly exceed the numbers now envisaged in the UN projections. But as the projections summarized above clearly suggest, and as calculations presented in a recent United Nations study (United Nations 2000) strongly imply, preventing population decline in the EU, and especially counteracting population ageing through immigration, would require a volume of migratory flow that would radically change the social and economic characteristics of the population in the receiving countries. From a demographic point of view whether an “open Europe” would indeed be a “fairer, richer, stronger, younger Europe” is, at the very least, highly questionable. “Younger” may seem a straightforward proposition, but it is not. The rejuvenating effect is temporary, as immigrants also age and become dependent. To sustain such an effect, would require a continuously maintained influx of youthful immigrants, a pattern that, in combination with below-replacement fertility of the native population, would eventually fully replace the original natives with immigrants and their descendants. “Fairer, richer, stronger” are outcomes that can be linked only tenuously with mass immigration. Fairness in a polity first and foremost should be evaluated by criteria that measure fairness with reference to the native population. To formulate immigration policy in response to the demographic behavior of other countries on the ground that “migrants need Europe” is unlikely to meet that standard. As deductive reasoning as well as much empirical evidence suggests, the kind of immigration that is demographically significant is likely to be detrimental to the

40

Immigration and the Transformation of Europe

material welfare and social well-being of the poorer segments of the receiving population by depressing wage levels at the lower end of the scale and creating friction in schools, housing markets, and claims for environmental amenities (though see the longer discussion and counterarguments in chapter 5). Mass immigration into an already populous country is likely to make that country “richer” only as measured by aggregate income rather than on a per capita basis. For a population that is already sizable, only the latter criterion is of real interest. And even if per capita gains are achieved through admitting large numbers of immigrants, the distribution of gains is likely to be lopsided, leaving many less well-off. Compensation for losses, even if theoretically possible, is in fact never effected. Thus, while the immigrants themselves are clear beneficiaries of the move they voluntarily elected to make, the welfare gain for the receiving population is far from assured. Finally “strength,” in an already populous polity, whether measured in terms of social cohesion, political stability, or military might, is unlikely to be enhanced by the large-scale immigration implicit in the “open Europe” concept, especially if the migrants are distant from the receiving population in their ethnic and cultural characteristics and if they are reluctant to adopt and conform to the social and political ethos of the receiving society. In this regard, recent European experience is not encouraging. It is then mass immigration itself that may bring “declining living standards and social division” rather than borders that are closed for permanent migrants, save for the select few whose admittance serves the interests of the population of the receiving country. Would, in the absence of immigration, “jobs go unfilled and services undelivered”? Would economies “shrink and societies stagnate”? In other words, is immigration the long-sought-for recipe for countering the failings of the European social model, with its existing high unemployment, low labor force participation rates, early retirement age, and inflexible labor markets? The remedy for these problems must be administered through domestic institutional reforms that command consensus of the native population. Immigration in all likelihood would provide an excuse for delay in tackling that agenda, thus making the timely adoption of the needed reforms less likely. Is a Europe that is less than “open” to immigration necessarily a “meaner Europe”? That charge addressed to the Parliament of the then EU-15 seems to have been singularly ill-timed, made as it was just three months before the enlargement of the EU through the admission of ten distinctly poorer new member countries. That enlargement created an area within which 450 million people will have the right freely to move,

Europe’s immigration challenge

41

seek employment, and settle. Successful economic and social integration of the enlarged EU, with its immense human diversity, is a difficult enough task for decades to come. There is no compelling reason to risk the success of that extraordinary project by endless further enlargement into an “open Europe.” But does not Europe need migrants because “Europeans are living longer and having fewer children”? Living longer would of course be seen by most Europeans as a blessing, hence something they should welcome and accommodate, rather than fear. Adjustment to a non-growing or slowly declining and much older population should be now at the forefront of the European social agenda. Albeit with a time lag, inevitably it will also be on the agenda of countries everywhere, including the lessdeveloped countries. There should be no attempt to unload the costs of solving Europe’s ageing problem onto the rest of the world. A domestic solution for the ageing problem is eminently feasible in rich industrial societies as long as fertility is at, or not far below, replacement level. In any case, solution through immigration is only a temporary remedy that would leave bigger problems in its wake. Most importantly, recourse to such a solution – substitution of immigrants for home-grown births – would provide a continuing excuse for Europe not to confront the problem of its fertility deficit. Societies that want to survive must learn how to reproduce themselves. The second lesson of the preceding demographic outline is simply stated, as it is the corollary of the first. Europe’s neighborhood – in Africa and in south-west Asia – has an inevitable demographic agenda of its own: to speed-up and complete the demographic transition without compromising the longer term economic viability of its societies. This problem cannot be solved by exporting people to Europe. Europe has a large population in its own terms, but not in comparison with the rest of the world. Even if the entire EU-25 disappeared today, that cataclysmic loss would represent barely six years of global population growth. The solution for unsustainable demographic expansion is not out-migration: it has to be domestic adjustment to social and economic realities.  Annan, Kofi 2004 “Why Europe needs an immigration strategy” www.un.org/ News/ossg/sg/stories/sg-29jan2004.htm. Demeny, Paul 1997 “Replacement-level fertility: The implausible endpoint of the demographic transition,” in Gavin W. Jones, Robert M. Douglas, John C. Caldwell and Rennie M. D’Souza (eds) The Continuing Demographic Transition. Oxford: Clarendon Press, 94–110.

42

Immigration and the Transformation of Europe

McEvedy, Colin and Richard Jones 1978 Atlas of World Population History. New York: Penguin Books. United Nations, Population Division 2000 Replacement Migration: Is It a Solution to Declining and Ageing Populations? New York: United Nations. United Nations, Population Division 2005 World Population Prospects: The 2004 Revision. Population Database. http://esa.un.org/unpp/

3

Migration into OECD countries 1990–2000

∗

Peder J. Pedersen, Mariola Pytlikova, and Nina Smith

Introduction As the previous chapter showed, many OECD countries expect to face the problem of declining and ageing populations in coming decades. This will impose increasing pressure on their welfare systems. Immigration of young people to these ageing OECD countries is often cited as one of the possible solutions to this problem. However, the opponents of this solution point to the findings of recent studies on immigrants’ economic performance in a number of European countries. They show that immigrants as a group actually tend to be more welfare-dependent than natives (see Riphahn 1999; Hammarstedt 2000; Storesletten 2003; and Wadensjo¨ and Orrje 2002). Thus, increasing the immigration flows may not be a solution to the problem of population ageing, but might instead impose a higher fiscal burden for the receiving economies. This suspicion is strengthened by change in the composition of immigrants to the OECD countries in recent decades. While labor migration flows dominated until the 1970s, refugee immigrants and family reunion migration from non-Western or less developed countries are now growing sources of net immigration in many OECD countries (see Chiswick and Hatton 2002). These immigrants show lower rates of social mobility, skills transferability, and skills acquisition, implying that they have difficulties entering the labor market (see Borjas 1994; and Chiswick 1986, 2000). Why has the composition of immigrants changed compared to a few decades ago? What are the driving forces behind recent immigration? The classical explanation is that relative real wages and employment opportunities are some of the main determinants of international migration. Other ∗

We are grateful to Anna Kossowska for her very helpful research assistance. We would also like to thank Helena Skyt Nielsen, Tor Eriksson, Antonio Rodriguez, Michael Rosholm, participants at the EALE Conference, Conference on Ethnic Minorities, Integration and Marginalisation, DGPE workshop, and the Seminar on Welfare Research for several helpful comments.

43

44

Immigration and the Transformation of Europe

more recent explanations focus on the effects of welfare-state regimes. Generous social services and benefit levels and a high tax pressure are characteristics of many OECD countries nowadays. According to Borjas (1987, 1999a, 1999b), the generosity of the welfare state may play an important role in migrants’ decisions when choosing a country of destination – the so-called “welfare magnet effect.” On the other hand, a number of non-economic factors are also considered highly important regarding the migration decision. Beside classic factors as “love and wars,” these include random events, environment, climate, language and aspects of “cultural distance.” Regarding the last factor, it is a standard result that the more “foreign” or distant the new culture is and the larger the language barrier, the less likely an individual is to migrate. However, changes and improvements in communication, continued globalization of the economy and declining costs of transportation may imply that the effect of “distance” has been reduced during the latest decades. Further, network effects may also counteract “distance.” If the concerned ethnic group is already present in the destination country, this may induce further immigration from the ethnic group concerned. Thus, an interesting question is: how much do the “pure” economic factors like relative wages or incomes, employment opportunities, tax pressure, and social expenditure level explain migration behavior, and how much is explained by other factors like immigration policies, social networks, cultural and linguistic distance, threat to own freedom and safety, random events or love? Despite a wide body of theoretical and empirical studies on the determinants of migration existing in the literature, the evidence from a multicountry perspective has been rather scant. Due largely to data limitations, most studies have only focused on the migration flows into one country. In this chapter, we add to the empirical evidence by analyzing the determinants of gross migration flows into a large number of OECD countries. We estimate a number of annual regressions using panel data econometric techniques on the flow of immigrants from 129 countries to 26 OECD countries for the period 1990–2000. Our results indicate that gross migration flows respond to economic determinants, but the traditional factors such as cultural and linguistic distance are important as well. We further find the network effect to be a significant driving force in immigration development. The rest of the chapter is organized as follows. The second section surveys earlier research in the area. The third section describes the database collected for this study, and the fourth section describes immigration development and trends in the OECD countries. The fifth section presents a model of international migration. Results from the

Migration into OECD countries 1990–2000

45

econometric analyses are given in the sixth section. Finally, the seventh section offers some concluding remarks. Review of earlier research The classical economic theories on migration have focused on differences in income opportunities as the main determinant of international migration (see Hicks 1963; and Sjaastad 1962). However, in reality, the incentives to migrate measured only by differentials in expected earnings have failed to explain why so few people move given huge differences in wages across the world. Some modifications within the neo-classical framework have been introduced, for example, the probability of being employed or unemployed (Harris and Todaro 1970; Jackman and Savouri 1991). Further, the decision to migrate has been seen as a family or household decision. A move takes place only if the net gain accruing to some members exceeds the others’ net loss (Mincer 1978; Holmlund 1984). A further step is made by the new economics of labor migration, which sees labor migration as a risk-sharing behavior in families. In contrast to individuals, households may diversify their resources, such as labor, in order to minimize risks to the family income (Stark 1991). Another theory is based on migration networks. Immigrants do not have all the information on the alternatives for potential immigration targets and often they perform only limited research. One possible way to reach relatively good and safe decisions in the face of uncertainty is to decide on the basis of the migration network’s information. Massey et al. (1993) define migration networks as “. . . sets of interpersonal ties that connect migrants, former migrants, and non-migrants in origin and destination areas through ties of kinship, friendship, and shared community origin.” The models of migration networks have been based on the network externalities theory. Positive externalities exist if the immigrant utility (utility of newly coming immigrants and previous immigrants) grows in response to an increase in the number of newcomers. The network externalities theory distinguishes between so-called “community effects,” which increase the utility of a community (i.e. inflow of people from the same nation helps create subcultures), and “family effects,” which only increase the utility of friends and relatives (Carrington et al. 1996). However, there might also be a negative externality stemming from continuously increasing immigration population. The growing number of immigrants increases competition in the market and may reduce wages, so that accelerated migration could put a strain on immigrants’ well-being.

46

Immigration and the Transformation of Europe

Nevertheless, immigration flows may not stop even if the immigration creates negative externalities (see Epstein 2002; Bauer et al. 2002; and Heitmueller 2003). An important question in most of the recent literature is the importance of selection processes in the migration decision (see Borjas [1999c] for an overview). One of the first contributions in this area is found in Borjas (1987). Within the framework of the Roy model (1951), Borjas looked at the skill differentials between immigrants and natives in relation to the variance in the wage distribution. The composition of the migration flows by skill is determined by the individuals’ position in the home-country wage distribution and the cross-country variance differential. Above-average performers in the home labor market are potential emigrants to a country with a high wage dispersion. On the other side, below-average performers are potential migrants to a country with a low wage dispersion. Thus the model predicts that a country with a low wage dispersion will have an over-representation among the below-average performing immigrants. The more positively selected migrants are, the more successful their adjustment in the new country, and the more beneficial their impact on the destination economy and society. The selection theory was tested on data for immigration flows to the United States during the period 1951–80. Borjas found that the lower the source-country income level (per capita) and the higher the source-country inequality, the larger is the inflow rate. Borjas (1999b) focuses on the level of welfare programs as a pull factor for potential immigrants, that is, a “welfare magnet” effect. The theories of self-selection are combined with the fact that potential emigrants must take into account the probability of being unemployed in the destination country. The consequences of this risk may be lowered by the existence of welfare benefits in the destination country. Such welfare income is basically a substitute for earnings during the period of searching for a job. Borjas (1999b) investigates whether immigrants’ location choices after arrival to the United States are influenced by the dispersion in the welfare benefits. He argues that immigrant welfare recipients will be clustered in the states that offer the highest welfare benefits – while the native welfare recipients will be much more dispersed across the states. His empirical work indicates a negative selection of immigrants into California – a state with a relatively generous system compared to other United States states. The selection theories and the Borjas studies have gained considerable attention, some supportive and some critical (Jasso and Rosenzweig 1990; Chiswick 2000). For example, one of the important assumptions of the Borjas model is the non-existence of fixed out-of-pocket money costs, which in reality are quite high (e.g., transportation costs,

Migration into OECD countries 1990–2000

47

housing), and which are considered very important in human capital migration models (Chiswick 2000). These migration costs constitute huge barriers to migration especially for low-skilled people from poor countries characterized by an unequal income distribution. Therefore, there could very well be a positive selection from countries with an unequal income distribution. Such considerations seem to be reflected in results from empirical studies, which fail to give clear support to the Borjas selection theory. Zavodny (1997) finds, based on studies of immigration to the United States, that immigrants do not respond to interstate differentials in welfare generosity, but rather to differences in the sizes of the foreign-born populations. By using aggregate data on immigration to the United States from eighteen countries of origin in 1982 and 1992, her results indicate that new immigrants are attracted to areas with large immigrant populations, indicating that network effects dominate. Because earlier immigrants have been disproportionately located in high-welfare states, it may appear that high welfare benefits attract immigrants (though of course these two parallel factors are difficult to disentangle). Like Zavodny, Urrutia (2001) found no evidence that United States immigrant settlement was determined by high levels of welfare benefits. Urrutia (2001) finds that the relative costs of migration present the main explanation of the observed migration pattern. Countries with relatively low (high) fixed costs, for example, due to geographical distance, are more likely to send immigrants from the bottom (top) of the distribution of abilities. Likewise the results in Chiquiar and Hanson (2002), using Mexico and United States census data, fail to support the selection hypothesis. In a study by Hatton and Williamson (2002), based on time series on migration flows to the United States the results are more mixed. They find significant and quantitatively important effects of source country per capita income and education and they also confirm the Borjas-Roy selection model as they find that larger source-country inequality increases emigration to the United States. On the other hand, a number of other factors are also found to be important, like distance, language, and the stock of former immigrants, indicating that network effects or herding behavior also play a major role in international migration. Some empirical research on this issue has been conducted for European countries as well (see Hatton and Williamson [2002] for the United Kingdom and the survey on studies of migration into Germany by Fertig and Schmidt [2000]). By using data from the European Community Household Panel, Boeri et al. (2002) examine whether welfare dependency is larger in countries with more generous benefit systems. Their findings are consistent with the view that welfare benefits distort the

48

Immigration and the Transformation of Europe

composition of immigrants, both in terms of observable and unobservable characteristics. They argue that although the effects are quantitatively moderate, some of the most generous countries seem to act as welfare magnets. As regards studying determinants of international migration in a multicountry perspective, Mayda (2004) has carried out a recent study on the migration flows into fourteen OECD countries. Her results show that income prospects in destination countries and distance are major determinants of migration flows. In this study, we are able to further extend the number of countries included in Mayda’s analysis, and we also have access to information on a large number of characteristics for both source and destination countries. The database It is not an easy task to collect data on international migration flows because a number of problems arise with respect to availability, variations of definitions of immigrants and migrations flows, and difficulties in getting comparable data from many countries on variables which may combine to explain migration flows. In order to have more precise data on immigration, we have contacted the statistical bureaus in the twenty-six selected destination OECD countries and asked them for detailed information on immigration flows and stocks in their respective country during the period 1989–2000. This information is supplemented by published OECD statistics from “Trends in International Migration” publications.1 Besides flow and stock information, we have collected a number of other time-series variables, which are used in the estimation of migration behavior. These variables are collected from different sources: OECD, World Bank, UN (United Nations), ILO (International Labour Organization), and IMF (International Monetary Fund) publications. The Appendix contains a list of all the variables used in estimated models, including definitions and data sources for each variable. In total, the data set contains information on immigration flows and immigration stocks in twenty-six OECD countries from 129 countries of origin. Although our data set presents substantial progress over those used in earlier research, a number of problems remain. First of all, the data set is unbalanced, with missing observations in the panel. For the majority of destination countries, we have information on migration flows and the 1

Unfortunately, we are not able to distinguish whether the immigrants are job- or studyrelated people, tied movers in relation to family reunions or refugees and asylum seekers.

Migration into OECD countries 1990–2000

49

stocks of immigrants for most of the years, but with different numbers of observation for each destination country (see Appendix, Table 3.A1, for means, standard deviations and number of observations for each destination country on gross migration flows, stock and other variables we have used in our analysis). There are missing observations regarding explanatory variables for some countries of origin as well (see Appendix, Table 3.A2). Second, as noted in almost all the chapters in this volume, different countries use different definitions of an “immigrant”2 and different sources for their migration statistics.3 Further, information on our data set is given in Pedersen et al. (2004). Description of migration trends During the 1980s and the beginning of the 1990s, immigration inflows increased in almost all OECD countries. Figure 3.1 shows the development of total volume of gross immigration inflows into seventeen OECD countries (see note 1 in Figure 3.1) during the period 1990– 2000. According to Figure 3.1 the immigration flows peaked in 1991, reaching more than 3.7 million that year. As regards the composition of gross immigration flows by sourcecountry continent, we observe in Figure 3.1 that on average, Europe constitutes a source for almost 50 percent of all immigration flows. The 2

3

In definitions of immigration flows, some countries like Australia, Canada, The Netherlands, New Zealand, Poland, the Slovak Republic, and the US define an “immigrant” by country of origin or country of birth. Austria, the Czech Republic, Denmark, Finland, Greece, Iceland, Italy, Norway, and Sweden define an immigrant by citizenship. Some countries like Belgium, France, Hungary, Germany, Japan, Luxembourg, Portugal, Spain, Switzerland, and the UK define an immigrant by nationality. For immigration stock, the definition of immigrant population differs among countries as well: the majority of countries, especially Australia, Austria, Canada, Denmark, Iceland, The Netherlands, New Zealand, Poland, the Slovak Republic, Sweden, the UK, and the US define immigrant populations by country of origin or country of birth. Some, like the Czech Republic, Finland, Greece, Italy, and Norway define immigrant populations by citizenship and others like Belgium, France, Hungary, Germany, Japan, Luxembourg, Portugal, Spain, and Switzerland define immigrant populations by nationality. The differences in definition of immigrant populations in the case of immigration stock are relatively important. The first one, by country of origin/birth takes into account foreign-born population, i.e. the first generation immigrants, and thus it contains also immigrants that have obtained citizenship. The second and third definitions, by citizenship and nationality, include second and higher generation foreigners, but do not cover naturalized citizens. Thus, the nature of legislation on citizenship and naturalization plays a role. For example, Belgium, Germany, Luxembourg, The Netherlands, Switzerland, and the Scandinavian countries use data based on population registers, while the majority of southern and eastern European countries use data based on issuing residence permits. Australia, Canada, New Zealand, and Poland use data from censuses: some countries like Greece, the UK, and the US use labor force surveys while others, like France and Japan, use information from various sources, including employer-reported social security data.

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

1990

1991

Europe

1992

1994

North America and Oceania

1993

1996

1997

South and Central America

Year

1995

Africa

1998

Asia

1999

2000

Figure 3.1. Total volume of gross immigration inflows to 17 OECD countries, 1990–2000, by source-country continent.1 Note 1: The included destination countries are: Australia, Belgium, Canada, Denmark, Finland, France, Germany, Hungary, Japan, Luxembourg, New Zealand, Norway, Poland, Spain, Sweden, Switzerland, and the United States. These countries are selected because we have annual data for all years, i.e. no missing observations on flows, for these countries. The following countries have been excluded due to missing observations on flows for some years: Austria, Czech Republic, Greece, Iceland, Italy, Netherlands, Portugal, Slovak Republic and the United Kingdom. Source: Own calculations.

Millions

Migration into OECD countries 1990–2000

51

observed dynamic development of immigration flows from Europe was mostly due to the removal of the Iron Curtain in 1989 and the Yugoslavian civil war, which gave rise to a large increase of migration within Europe in the early 1990s. This can be seen from Figure 3.2, in which we show immigration flows from the European Union4 and the rest of Europe5 as a percentage of total European flows. The percentage of immigrants coming from countries other than Europe reached 80 percent of all immigration flows from Europe in 1992 (see Figure 3.2). In the most recent years (legal) migration flows from these countries seem to have stabilized, mainly due to immigration restrictions. The immigration flows stemming from EU member countries have been very stable over time, fluctuating around 320,000 annually. The distribution of OECD immigration flows by source-country continents has been relatively stable since the early 1990s (see Figure 3.1). We observe a slight increase in overall immigration flows at the end of the decade, especially from South American, African and Asian countries. It should be noted that Figures 3.1 and 3.2 describe gross migration flows, not net flows. If there are large differences with respect to out-migration behavior for the different immigrant groups, the net migration flows may be very different from the gross flows. Non-Western immigrants tend to have much lower return and out-migration rates than Western immigrants in many countries, and thus the stocks of OECD immigrants from different regions may still be changing despite the apparently quite stable development in Figure 3.1 Figure 3.3 shows the stock of foreign population as a percentage of the total population in twenty seven OECD countries in the years 1990 and 2000. The stocks of immigrants in OECD countries vary considerably, ranging from 37 percent in Luxembourg in 2000, to less than 1 percent in the Slovak Republic. It is also apparent from Figure 3.3, that migration flows have changed in the sense that some countries, for example, Australia and Canada, have experienced a much smaller growth in their immigrant population during the latest decade compared to relatively new immigration countries like Austria, Denmark, Iceland, Luxembourg, and some of the southern European countries. The decomposition of immigrants’ stock by continents of origin in the years 1990 and 2000 is shown in Figure 3.4. The highest proportion of 4 5

Here we consider the “old” EU made up of 15 Member States. We mean the following countries: Albania, Bulgaria, Cyprus, Czech Republic, Hungary, Poland, Romania, Slovakia, Turkey, all states of the former USSR (Union of Soviet Socialist Republics) and the former Yugoslavia.

1990

1991

1992

1993

EU countries

1994

1995

1996

Non-EU countries

Year

1997

1998

1999

2000

Figure 3.2. Gross immigration inflows from EU- and non-EU countries as a percentage of total inflows from Europe into 17 OECD countries, 1990–2000.1 Note: 1 See Figure 3.1. Source: Own calculations.

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

0

2

4

6

8

10

12

14

16

18

20

22

24

26

28

30

32

34

36

38

Denmark

Iceland 1990

France 2000

Hungary Greece

Japan

Finland Czech Rep. Portugal

Poland Spain

Italy

Norway

Ireland

UK

Belgium

Germany Austria

Netherlands

Sweden

US

N. Zealand

Switzerland Canada

Luxembourg

Australia

Figure 3.3. Stock of foreign population as a percentage of total population in 1990 and 2000 in selected OECD countries. Note: 1 Due to data availability the figure shows information on: 1991 instead of 1990 for Austria, Iceland, Italy, and Spain; 1991 and 2001 instead of 1990 and 2000, respectively, for Canada, Luxembourg, and New Zealand; 1999 instead 2000 for France; 1997 instead of 2000 for Greece; 1994 instead of 1990 for the Czech Republic; 1994 and 1999 instead of 1990 and 2000, respectively, for Hungary; 1995 instead of 1990 for the Slovak Republic and 1992 instead of 1990 for the United Kingdom. Source: Own calculations.

%

40

Slovak Rep.

54

Immigration and the Transformation of Europe Year 1990

Unknown 9% Other European countries 18%

North America and Oceania 5%

Year 2000

South and Central America 5% Africa 9%

Unknown 7%

Africa 10%

Other European countries 24%

Asia 16%

EU 38%

North America and Oceania South and Central 4% America 6%

Asia 18%

EU 31%

Figure 3.4. Proportion of total immigration stock in OECD countries by continents of origin, 1990 and 2000. Note: 1 Due to data availability the figure shows information on: 1991 instead of 1990 for Austria, Iceland, Italy, and Spain; 1991 and 2001 instead of 1990 and 2000, respectively, for Canada, Luxembourg, and New Zealand; 1999 instead 2000 for France; 1997 instead of 2000 for Greece; 1994 instead of 1990 for the Czech Republic; 1995 instead of 1990 for the Slovak Republic, and 1992 instead of 1990 for the United Kingdom. Hungary and Poland have been excluded due to non-detailed information on countries of origin and missing year 1990 in the case of Poland. Source: Own calculations.

immigrants residing in OECD countries originates from Europe and the majority of them come from EU member countries. Nevertheless, the proportion has changed over time, and has dropped from 38 percent in 1990, to 31 percent in 2000. At the same time, the share of immigrants from countries other than the EU has increased sharply by 6 percentage points, from 18 percent of total stock in 1990 to 24 percent in 2000. Further, we observe a slight growth in the share of immigrants from Asia, Africa, and South and Central America, with an increase of 2, 1 and 1 percentage point/s respectively. There are large variations in the composition of immigrant stocks across individual OECD countries and the composition is changing over time (see Table 3.1). One common trend observed in almost all OECD countries is a falling share of foreign population from the EU, North America, and Oceania. Thus, although there has been a relatively stable

Migration into OECD countries 1990–2000

55

development of immigration flows, the share of immigrants from these countries of origin is in fact falling due to a high propensity to return migration.6 In some countries, like Luxembourg, Belgium, and Switzerland, the large stock of immigrants stems mainly from other OECD countries (working in EU institutions and the financial sector) while in other countries, to some extent in new immigration countries, the proportion of immigrants who stem from poor source countries is large. Typical examples are southern European countries, where the proportion of immigration stock from Africa, Asia, and South and Central America is relatively high mostly due to geographical and linguistic7 proximity. Scandinavian countries – relatively new immigration countries – experienced a sharp increase in the numbers of immigrants of African and Asian origin which, almost doubled over the decade studied.8 The traditional immigration countries like Australia, Canada, and New Zealand have all encountered a significant increase in population of Asian origin. The United Kingdom has historically had a large proportion of people coming from the Asian continent (mainly former colonies). Nevertheless, the increase in their share has been rather moderate. The largest share of immigrant population in the United States originates, not surprisingly, from South and Central America mostly due to the well-represented Mexican immigrant population. The share has further grown by 5 percentage points, reaching 41.5 percent of the entire immigrant population in the United States. Nearly all OECD countries have experienced an increase in foreign population from European countries outside the EU (which, for our period, includes those countries that joined in 2004). But the proportion is especially high in countries like Germany, Austria, and Switzerland. The Nordic countries experienced a fairly large increase in the stock from non-EU origins, with the highest increase in Finland, where the share of non-EU immigrants reached nearly 52 percent of the total foreign population. This is mostly due to a large movement from Estonia that shares common Nordic history and language proximity.9 The non-EU nationals are also well represented among foreigners in the Central European OECD Member States – Czech and Slovak

6 7 8 9

Mostly highly educated, professionals, students, etc. Language plays a role for immigrants coming from South and Central America. Except for Norway. The Estonian language belongs to the same Finno-Ugric family of languages as Finnish and Hungarian.

1990

48.9 18.7 62.2 41.0 12.5 39.5 46.7 36.6 31.5 31.3 61.7 11.6 2.2 92.7 20.8 52.2

Australia Austria Belgium Canada Czech Rep. Denmark Finland France Germany Greece Iceland Italy Japan Luxembourg Netherlands New Zealand

37.0 15.9 65.9 29.8 6.9 22.0 19.6 37.6 26.3 27.5 47.5 11.4 2.1 85.5 18.7 37.8

2000

EU (15) countries (in %)

OECD

Share of immigration stock by continents of origin:

6.1 37.5 10.6 8.8 66.3 22.8 22.2 8.5 51.8 22.9 8.2 11.7 0.1 2.8 19.2 1.0

1990 8.6 61.2 10.0 11.1 70.6 27.4 51.8 10.0 52.7 33.7 20.3 26.4 0.3 7.9 19.7 1.4

2000

Other European countries (in %)

7.6 1.6 1.5 6.8 3.9 3.7 8.0 0.9 2.0 12.5 14.2 6.4 4.5 1.3 1.7 15.3

1990 11.2 1.0 1.7 5.5 1.9 2.8 3.5 1.1 1.8 10.7 10.6 1.7 4.0 0.9 1.9 14.8

2000

North America and Oceania (in %)

0.0 0.4 0.3 6.8 0.7 1.9 0.6 0.5 0.8 2.8 0.5 0.0 6.5 0.4 15.6 0.2

1990 0.0 0.6 0.6 7.1 0.3 1.8 1.0 1.6 1.1 2.0 1.2 7.7 18.1 2.0 14.4 0.3

2000

South and Central America (in %)

1.4 1.7 19.1 3.5 1.6 4.5 3.7 42.8 3.4 9.8 1.2 30.5 0.1 1.4 16.8 1.6

1990 2.6 1.6 16.0 4.4 1.0 8.7 8.0 40.8 3.4 7.3 1.7 30.1 0.3 1.6 17.0 4.2

2000

Africa (in %)

12.2 4.4 1.8 24.7 14.3 25.5 9.8 3.8 8.8 19.8 7.4 13.7 85.5 1.4 20.5 10.7

1990

19.8 1.9 2.8 35.1 16.1 28.5 15.9 7.3 10.2 16.0 13.5 18.5 72.8 1.6 21.8 19.5

2000

Asia (in %)

23.8 35.8 4.6 8.6 0.6 2.1 9.0 6.8 1.7 0.8 6.7 26.1 1.0 0.1 5.3 19.1

1990

20.8 17.8 3.0 7.0 3.2 8.8 0.2 1.6 4.4 2.8 5.3 4.3 2.3 0.6 6.4 22.1

2000

Unknown (in %)

Table 3.1 Proportion of immigration stock in particular OECD countries by continents of origin, 1990 and 2000

100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100

1990

100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100

2000

Total (in %)

41.0 – 17.0 8.8 48.3 51.4 71.6 30.0 15.3

45.5 20.3 27.3 8.0 35.5 37.7 56.8 27.7 10.6

9.4 – 0.0 70.5 1.6 18.4 19.7 6.1 6.4

17.0 73.7 0.0 72.3 5.4 26.0 27.6 7.0 3.7

7.8 – 8.3 3.6 4.0 2.1 1.4 7.9 0.7

5.4 1.4 5.1 2.6 1.8 1.9 1.5 8.2 1.4

4.5 – 15.4 0.6 17.0 5.7 1.3 3.6 36.6

2.0 0.0 12.5 0.5 22.1 5.5 2.1 4.3 41.5

5.4 – 40.2 2.4 15.3 1.9 1.7 12.2 1.3

5.7 0.1 45.1 1.4 26.1 4.4 2.4 14.8 2.0

29.0 – 0.0 9.2 6.8 15.1 3.3 28.2 23.3

19.6 1.0 2.6 10.8 6.1 21.5 4.9 28.6 25.5

2.8 – 19.1 4.8 7.0 5.5 1.1 12.1 16.3

4.9 3.6 7.5 4.4 3.0 3.0 4.8 9.4 15.3

100 – 100 100 100 100 100 100 100

100 100 100 100 100 100 100 100 100

Note 2: Due to data availability the figure shows information on: 1991 instead of 1990 for Austria, Iceland, Italy, and Spain; 1991 and 2001 instead of 1990 and 2000, respectively, for Canada, Luxembourg, and New Zealand; 1999 instead 2000 for France; 1997 instead of 2000 for Greece; 1994 instead of 1990 for the Czech Republic; 1995 instead of 1990 for the Slovak Republic and 1992 instead of 1990 for the United Kingdom. Ireland and Hungary have been excluded due to non-detailed information on countries of origin. Source: Own calculations.

Norway Poland Portugal Slovak Rep. Spain Sweden Switzerland UK US

58

Immigration and the Transformation of Europe

Republics, Hungary, and Poland. It is apparent from the development in Figure 3.3 and Table 3.1 that the new EU Member States became emigration and immigration countries all at the same time. Although the proportion of foreigners per population is still very low in the Czech and Slovak Republics and Poland, it has risen significantly over time (see Figure 3.3). The major influx of foreigners comes from non-EU countries (see Table 3.1), and in particular from the former USSR (Union of Soviet Socialist Republics) countries, Romania, Bulgaria, and the former Yugoslavia. There may be many factors explaining the changing composition of immigration observed above. Besides differences in return migration behavior, there might be different factors driving migration flows. Moreover, the determinants of migration may vary in time and across various countries of origin. The next section describes in detail some particular determinants of migration and we present a formal model of migration flows. A model of international migration Standard neoclassical theory assumes that potential migrants have utilitymaximizing behavior, that they compare alternative potential destination countries and choose the country which provides the best opportunities, all else being equal. The immigrants’ decision to choose a specific destination country depends on many factors, which relate to the characteristics of the individual, the individual’s country of origin, and all potential countries of destination. Following Zavodny (1997) we consider individual k’s expected utility in country j at time t given that the individual lived in the country i at time t−1 Ui j kt = U(Si j kt , Di j , Xi kt , Xj kt )

(1)

where Si j kt is a vector of characteristics that affects an individual’s utility of living in country j at time t, given that the individual lived in country i at time t−1. For example, an individual may want to move to a country where his friends or family members are. Di j reflects time-independent, fixed-out-of-pocket, and psychological/social costs of moving from country i to country j. Xi kt and Xj kt are vectors of push and pull factors that vary across time and affect individual k’s choice where i denotes source country and j denotes destination country, (i = 1, . . . , 129, and j = 1, . . . , 26); t is time period (t = 1, . . . , 11). We assume that the utility of an individual has a linear form: Ui j kt = α1 Si j kt + α2 Di j + α3 Xi kt + α4 Xj kt + εi j kt

(2)

Migration into OECD countries 1990–2000

59

where εi j kt represents an idiosyncratic error term and α1 , α2 , α3 and α4 are vectors of parameters of interest to be estimated. A potential immigrant maximizing his utility chooses the country with the highest utility at time t conditional on living in country i at time t−1. Thus, we can write the conditional probability of individual k choosing country j from 26 possible choices as: Pr( jkt /i kt−1 ) = Pr[Ui j kt = max(Uki 1t , Uki 2t , . . . , Uki 26t )]

(3)

Model (3) might be used for estimation of the determinants of the individual’s locational choice.10 However, as we use macro data, we aggregate up to population level by summing over k individuals. The number of individuals migrating to country j, that is, whose utility is maximized in that country, is given by:
Mi j t = Pr[Ui j kt = max(Uki 1t , Uki 2t , . . . , Uki 26t )] (4) k

where Mi j t is the number of immigrants moving to country j from country i at time t. This migration measure represents an ‘ex ante’ measure of the migration flows. The resulting and observed ‘ex post’ flow may of course also be affected by migration policy, illegal immigration, etc. We assume a linear form of the variables that influence the locational choice of immigrants. Hence we have: Mi j t = β1 Si j t + β2 Di j + β3 Xi t + β4 Xj t + µi j t ,

(5)

where µi j t is an error term assumed to be iid with zero mean and constant variance. We normalize the immigration flows by population size in the destination country, that is, we use the immigration rate, mi j t , instead of immigration flow in absolute numbers as the dependent variable. mi j t is defined as immigration flow to country j from country i divided by population size in country j in the period t. All time-varying explanatory variables are lagged by one year in order to account for information on which the potential immigrants base their decision to move. Further, we include the normalized lagged stock of immigrants, s i j t−1 , that is, the stock of immigrants from source country i, divided by population in destination country j. The (normalized) stock of immigrants s i j t−1 is expected to catch the existence of “networks” – links between sending 10

The model does not take into account potential out-migration or return migration. Since the stock of immigrants is the net result of in- and outflow mechanisms, and since outmigration is non-negligible for many immigrant groups, this topic is also very important when explaining the composition of immigrant groups in different countries. However, in this study we only focus on gross immigration.

60

Immigration and the Transformation of Europe

and receiving countries. Through the “networks” the potential migrants receive information about the immigration country – about the possibility of getting a job, about economic and social systems, immigration policy, people, and culture. It facilitates easier immigration and further easier adaptation of newly coming immigrants into the new environment. We have further included destination countries fixed or random effects, cj, in order to capture unobserved time-constant factors influencing immigration flows,11 for example, differences in national immigration policy (see Fertig and Schmidt [2000] for the importance of the homogeneity assumptions). Thus, the model to be estimated is: mi j t = β1 s i j t−1 + β2 Di j + β3 Xi t−1 + β4 Xj t−1 + c j + µi j t

(6)

Di j contains variables reflecting costs of moving to a foreign country. First, we include a variable describing cultural similarity denoted neighbouring country. It is a dummy variable assuming the value of 1 if the two countries are neighbors, 0 otherwise. The variable colony is a dummy variable assuming the value of 1 for countries ever in colonial relationship, 0 otherwise. This variable is included because the past colonial ties might have some influence on cultural distance: provide better information and knowledge of the potential destination country and thus lower migration costs, which could encourage migration flows between these countries. Further, we include a variable linguistic distance, which is a dummy variable equal to 1 for common language in two countries, 0 otherwise. In order to control for the direct costs (transportation costs) of migration, we use the measure of the distance in kilometres between the capital areas in the sending and receiving countries. We also include a variable trade volume, which is defined as the total trade values (both imports and exports) for all country pairs.12 We expect that the business ties represented by the volume of trade could have (positive) effects on international migration. Moreover, this variable is often considered as an indicator of globalization. The explanatory variables included in Xit−1 and Xjt−1 cover a number of push and pull factors such as the economic development measured by GDP (Gross Domestic Product) per capita in destination and source 11

12

We have also estimated a model with both destination and source countries fixed effects, but it does not result differently. Moreover, we found source countries fixed effect hard to interpret bearing in mind the large range of source countries. Import and export values from Direction of Trade Statistics are expressed in nominal US dollar prices. The values in constant prices would be more suitable for our analysis. However, we decided to use the nominal ones as it is a quite complex task getting suitable export and import deflators.

Migration into OECD countries 1990–2000

61

countries (which are supposed to catch relative income opportunities in the two countries), employment opportunities in the sending and receiving countries, measured by unemployment rates, and demographic and political factors. The hypothesis is that a higher (lower) level of economic development in the destination country will lead to higher immigration rates because potential immigrants expect to experience better (worse) income opportunities. The effect of GDP per capita in the source country may be more mixed. Earlier studies have found an inverted “U” relationship between the source country GDP and emigration; see Hatton and Williamson (2002). At very low levels of GDP, emigration is low because people are too poor to pay the migration costs. At higher income levels, migration increases, and when GDP levels increase further, migration may again decrease because the economic incentives to migrate to other countries decline. The GDP variable is supplemented by a variable reflecting the educational level of the source country, measured by adult illiteracy rate. It is expected, following Harris and Todaro (1970), that a low (high) unemployment rate in the destination (source) country will cause higher immigration flows. We also include a variable capturing population pressure, for example, population in the source country i divided by population in destination country j. The higher the relative population in the source countries, the larger migration pressure is expected. A more appropriate measure, which we are not able to include because of data limitations, would be the proportion of the population in the younger adult age groups, because a large proportion of migration flows is driven by these age groups, see, for example, Fertig and Schmidt (2000). The political pressure in the source country may also influence migration. Therefore, we include the variable freedom house index which intends to measure the degree of freedom, political rights and civil liberties in the countries. The variable is taking on values from one to seven, with one representing the highest degree of freedom and seven the lowest. Violated political rights and civil liberties are expected to increase migration flows. We include some variables which are assumed to capture potential pull factors relating to the “welfare magnet” theories, as presented by Borjas (1987, 1999a, 1999b). We have experimented with two variables: public social expenditure and tax revenue, both expressed as a percentage of GDP in the potential destination countries. Since the variables are highly correlated, we only include one of them at a time. In the estimations presented, only the relative tax level is included. We have also experimented with distributional indicators by including Gini coefficients as measures of inequality. However, we have had difficulties in getting comparable and reliable information for the majority of countries on this variable, and at the moment we are not able to include this factor in our study. Moreover,

62

Immigration and the Transformation of Europe

since we use aggregated macro data, we are not able to test directly for selectivity effects saying that there is a negative or positive selection from a given source country into a given destination, that is, that immigrants from the poor countries being at the lower part of income distribution may be more likely to move to the countries with more comprehensive welfare programs while immigrants from the upper end of the skill distribution in the poor countries may prefer destination countries with low tax pressure and lower social standards. All variables used in the estimations, except dummy variables, are in logs, that is, the estimated coefficients represent impact elasticities. The model given by (6) has been estimated by pooled OLS (Ordinary Least Squares) as well as by panel data estimators, that is, fixed effects and random effects estimators. Results The results from estimating a model of the log gross flows between the 129 source countries (i) and the 26 OECD destination countries (j) on annual unbalanced panel data for the period 1990–2000 are presented in Table 3.2. Column 1 shows the estimates using OLS and excluding the lagged stock of immigrants from country i in country j, while column 2 includes the stock variable. Comparing the two columns indicates that the existing stock of immigrants of a given ethnic origin is an important factor explaining future migration flows, exactly as it is found in other studies (Zavodny 1997; Hatton and Williamson 2002). The explanatory power (R2 ) of the model increases from 51 to 72 percent when including the stock variable,13 thus, this variable is included in all subsequent models. The highly significant coefficient to the stock variable indicates the existence of strong network effects. This could consist of a number of possible mechanisms, that is, as a background for family reunification or as indicators of faster access to the labour market in the new country, the more people already there from your own ethnic group. When comparing the pooled OLS results with the panel models treating the destination country in columns 3–4 as fixed or random effects, the overall impression is that the results regarding sign and statistical significance are quite robust across the different specifications. However, as 13

In order to see whether this result is not driven by the drop in observations when including the stock variable as regressor, we have estimated the model in column (1) without the stock variable and including exactly the same observations as in columns (2)–(5), i.e. 7,268 observations. The explanatory power increased in similar fashion, from 53 to 72 percent.

1.422 [0.056]∗∗∗ −0.097 [0.025]∗∗∗ −0.203 [0.074]∗∗∗ 0.191 [0.007]∗∗∗ −0.143 [0.031]∗∗∗ −0.088 [0.019]∗∗∗ −0.198 [0.016]∗∗∗ 0.137 [0.037]∗∗∗

2.131 [0.062]∗∗∗ −0.286 [0.030]∗∗∗ −0.682 [0.089]∗∗∗ 0.399 [0.008]∗∗∗ −0.003 [0.037] 0.100 [0.022]∗∗∗ −0.018 [0.018] 0.055 [0.043] No Yes 9689 0.510

Xjt−1 GDP per cap PPP, j Unemployment rate, j Tax revenue in j/GDP, j

Xit-1 Population (i)/Population (j) GDP per cap PPP, i Unemployment rate, i Illiteracy rate, i Freedom House Index, i

Fixed/Random effects of destination, cj Constant term included No of observations Adjusted R-squared

[0.008]∗∗∗

Yes Yes 7268 0.795

0.039 [0.010]∗∗∗ −0.266 [0.029]∗∗∗ −0.005 [0.017] −0.205 [0.013]∗∗∗ 0.200 [0.032]∗∗∗

0.788 [0.226]∗∗∗ −0.267 [0.029]∗∗∗ −0.093 [0.310]

– – – – 0.129 [0.012]∗∗∗

0.636

FE (cj )

Notes: 10, 5 and 1% levels of confidence are indicated by ∗ , ∗∗ and ∗∗∗ , respectively. Standard errors are in parentheses.

No Yes 7268 0.722

0.098 [0.053]∗ 0.347 [0.058]∗∗∗ −0.061 [0.077] −0.212 [0.016]∗∗∗ −0.006 [0.008]

0.615

[0.009]∗∗∗

0.394 [0.064]∗∗∗ 1.274 [0.060]∗∗∗ 0.237 [0.087]∗∗∗ −0.395 [0.018]∗∗∗ 0.280 [0.008]∗∗∗

–

Sijt−1 Stock of Foreigners/Pop.(j)

OLS

Dijt−1 Neighboring country (0/1) Linguistic distance (0/1) Colony (0/1) Distance in kilometers Trade volume

OLS

Dependent variable: mijt = Gross Flows per 1,000 inhabitants Independent variables:

Yes Yes 7268 0.706

0.107 [0.012]∗∗∗ −0.192 [0.030]∗∗∗ −0.022 [0.017] −0.194 [0.014]∗∗∗ 0.170 [0.032]∗∗∗

0.892 [0.167]∗∗∗ −0.266 [0.029]∗∗∗ 0.115 [0.255]

−0.004 [0.046] 0.388 [0.051]∗∗∗ 0.385 [0.070]∗∗∗ −0.089 [0.016]∗∗∗ 0.074 [0.013]∗∗∗

0.614 [0.008]∗∗∗

RE (cj )

Table 3.2 Estimation of migration flows from 129 source countries (i) to 26 (OECD) destination countries (j), 1990–2000

64

Immigration and the Transformation of Europe

expected, the absolute size of the coefficients is generally larger when applying OLS on the pooled samples of countries while the panel data estimators which control for country-specific fixed or random effects are generally smaller in numerical magnitude. To be able to know which model, fixed or random effects, fits our panel data context best, one should first establish whether there is a correlation between the unobserved factors influencing immigration flows and the explanatory variables. If the unobserved factors cj correlate with the explanatory variables, then the fixed effects panel data model has priority. If they are unrelated, then the random effects model is preferred. From the econometric point of view, a standard procedure is to test for this correlation by using a Hausman specification test. In our panel data context, the Hausman test actually confirms the random effects assumption of zero correlation between explanatory variables and country-specific effects.14 Concentrating on the results from the random effects estimation (RE) in column 4, the elasticity of the flow of immigrants from country i with respect to the stock of immigrants in country j is estimated to be about 0.6, implying that on average an increase in the stock of immigrants of 10 percent from a given source country induces an increase in annual gross flow of about 6 percent of new immigrants from this source country. Since we control for other country-specific factors, this result is mainly explained by the existence of network effects which seem to be both statistically significant and quantitatively of a considerable size. Similar results are found in Zavodny (1997) and Hatton and Williamson (2002). In regressions estimated by RE panel data technique, the dummy variable for source and destination countries being neighbors is found to be insignificant. The other distance-related dummy variables, linguistic distance and a dummy for the source country having in the past been a colony to the destination country, are consistently found to have the expected positive impact on migration flows with most coefficients being significant. Finally, in this group of variables, the distance between countries measured in kilometres and the pair wise trade volume between source and destination countries are both significant with expected signs. Increasing distance and smaller trade volume imply lower migration flows and vice versa. The next block of variables in Table 3.2 contains the pull factors in the destination countries. GDP per capita as a pure measure of gross income comes out with significant positive coefficients. In the same way, we find consistently that higher unemployment in destination countries has a significantly dampening impact on migration. The welfare state 14

It gives chi-squared (10) =8.48.

Migration into OECD countries 1990–2000

65

attractor among the pull factors is measured by the tax pressure needed to finance the welfare state. The effect is negative, but the tax level is only significant in the OLS estimations where we do not control for other country-specific factors.15 Thus, it seems that the tax level variable as a welfare state measure does not act as an attractor in migration flows. Zavodny (1997) also found that controlling for country-specific factors and network effects resulted in welfare state variables becoming insignificant regarding immigration to the USA. However, in our multi-destination countries case we get a negative coefficient to the welfare state variable in OLS regressions while Zavodny (1997) gets a positive coefficient when not controlling for stock and fixed effects. Next, we come to a block of source-country push factors. The first of these is a simple pair wise population ratio between source and destination country populations. Not surprisingly, the coefficient is significantly positive in all specifications. Further, we enter GDP per capita in source countries finding significantly negative coefficients, that is, higher income in source countries has a dampening impact on emigration from these countries. We find a negative impact on migration flows from unemployment in the source countries. In a regional context inside a country this would be a counterintuitive result as higher unemployment is expected to push people to other regions. Here, however, we deal with international mobility which is expected to be much more costly in both financial and other terms. Higher unemployment in a low-income country could simply indicate a situation making it more difficult due to financial restrictions to finance migration to another, eventually distant, country. The negative coefficient of the illiteracy rate indicates the same tendency. Migration to the rich OECD countries increases when the educational level in source countries increases. Overall, ‘poverty’ effects seem to be among the important determinants for migration flows. Higher economic growth in source countries is thus expected to create counteracting impacts on out-migration incentives. Unemployment will go down and educational standards will go up, acting to reduce the barriers to migration. But, at the same time, income goes up with a counteracting effect and the net effect becomes indeterminate.

15

It might be argued that controlling for country-specific factors partly “kills” the welfare effect because the characteristics of different welfare regimes are quite stable in most cases over an eleven-year period as used in our estimations. Further, we have tried several specifications with social expenditure as a proportion of GDP. This variable was insignificant as well.

66

Immigration and the Transformation of Europe

Finally, we have included the Freedom House Index among the sourcecountry push factors. The effect is positive and significant, indicating that lower degrees of freedom create out-migration incentives, part of it being in the form of refugees. Next, we show in Table 3.3 the separate results from the RE estimation disaggregated by continents of origin. Comparing with the results in Table 3.2, the importance of networks seems to be universal across continents of origin. The variables linguistic distance and colony seem to be significantly negative with respect to Africa as the continent of origin, in contrast to the results in Table 3.2. Further, we see that the insignificant coefficient to relative tax revenues in the RE estimation in Table 3.2 is the net outcome of insignificant coefficients regarding the flows from European and overseas Anglo-Saxon countries while the coefficient is found to be significantly negative for Latin America and Asia as continents of origin. A possible interpretation of this is that relative tax revenues function as a proxy variable for the tightness of immigration policies towards people from low income countries.

Conclusion In this chapter, we present results from empirical work on the migration flows from 129 countries into 26 OECD countries during the years 1990–2000. The estimations are made using both pooled OLS and panel data models and a very comprehensive database of potentially important background factors. The background factors include variables measuring the “distance” in different ways as well as linguistic and historical ties between the countries. Further, a number of economic variables are used, including indicators of the extent of national welfare state programs which could be among the attractors in international migration flows. This allows us to examine the economic or non-economic determinants of international migration flows. A very robust key result of our econometric analysis is that the network effects measured as the coefficient to the stock of immigrants of own national background already resident in a country have a large positive effect on immigration flows, and, networks play an important role in explaining current immigration flows. Further, linguistic closeness, former colonial and current business ties are important factors, although the magnitude of the impact on migration flows varies for different groups of destination countries. Geographic distance, on the other hand, has a negative impact on migration flows. This suggests that the costs of migration play an important role.

0.549 [0.512] −0.095 [0.088] 0.534 [0.642] 0.231 [0.076]∗∗∗ −0.269 [0.645] −0.206 [0.192] −1.018 [0.649] 1.429 [0.891]

1.410 [0.246]∗∗∗ −0.179 [0.042]∗∗∗ 0.433 [0.378] 0.070 [0.019]∗∗∗ −0.461 [0.059]∗∗∗ −0.119 [0.027]∗∗∗ −0.184 [0.022]∗∗∗ 0.239 [0.056]∗∗∗ Yes Yes 3646 0.638

Xjt−1 GDP per cap PPP, j Unemployment rate, j Tax revenue in j/GDP, j

Xit-1 Population (i)/Population (j) GDP per cap PPP, i Unemployment rate, i Illiteracy rate, i Freedom House Index, i

Fixed/Random effects of destination, cj Constant term included No of observations Adjusted R-squared

Yes Yes 1151 0.774

0.330 [0.033]∗∗∗ −0.654 [0.108]∗∗∗ 0.196 [0.050]∗∗∗ −0.062 [0.080] 0.135 [0.070]∗

0.716 [0.364]∗∗ −0.273 [0.053]∗∗∗ −0.920 [0.469]∗∗

0.421 [0.300] 0.486 [0.208]∗∗ −0.134 [0.208] −0.700 [0.161]∗∗∗ 0.037 [0.028]

0.669 [0.020]∗∗∗

South and Central America

Yes Yes 561 0.786

0.111 [0.045]∗∗ −0.218 [0.084]∗∗∗ −0.075 [0.059] −0.100 [0.153] −0.395 [0.120]∗∗∗

0.194 [0.383] −0.327 [0.079]∗∗∗ −0.673 [0.515]

−0.022 [0.285] −0.308 [0.146]∗∗ −0.729 [0.258]∗∗∗ −0.250 [0.125]∗∗ 0.149 [0.035]∗∗∗

0.737 [0.035]∗∗∗

Africa

Notes: 10, 5 and 1% levels of confidence are indicated by ∗ , ∗∗ and ∗∗∗ , respectively. Standard errors are in parentheses.

Yes Yes 553 0.747

0.711 0.165 [0.142] −0.524 [0.726] 0.071 [0.112] 0.106 [0.058]∗

[0.379]∗

0.471 [0.041]∗∗∗

North America and Oceania

0.022 [0.055] 0.346 [0.081]∗∗∗ 0.274 [0.098]∗∗∗ −0.030 [0.040] 0.154 [0.023]∗∗∗

0.570 [0.012]∗∗∗

Europe

Dijt−1 Neighboring country (0/1) Linguistic distance (0/1) Colony (0/1) Distance in kilometers Trade volume

Dependent variable: mijt = Gross Flows per 1,000 inhabitants Independent variables:

Yes Yes 1357 0.742

0.125 [0.025]∗∗∗ −0.184 [0.059]∗∗∗ −0.117 [0.032]∗∗∗ −0.208 [0.029]∗∗∗ 0.052 [0.062]

0.497 [0.277]∗ −0.403 [0.062]∗∗∗ −0.833 [0.413]∗∗

−1.545 [0.448]∗∗∗ 0.355 [0.140]∗∗ −0.057 [0.173] −0.208 [0.072]∗∗∗ 0.003 [0.026]

0.657 [0.017]∗∗∗

Asia

Table 3.3 Estimation of migration flows from 129 source countries (i) to 26 (OECD) destination countries (j), random effects (RE) panel data model, disaggregating by continents of origin, 1990–2000

68

Immigration and the Transformation of Europe

The impact from economic factors is measured by entering GDP per capita (PPP – Purchasing power parity – adjusted) and unemployment rates in both destination and source countries and tax pressure. Migration flows tend to react positively to higher income gaps and to react negatively to depressed labor markets in destination countries. In contrast to the simple welfare magnet hypothesis, it turns out that across our full set of cases, the coefficient to the tax pressure as a welfare state indicator becomes negative or insignificant. When we look at the results disaggregated on continents of origin there seems to be a clear pattern, that is, the tax variable is insignificant for immigration flows from high income countries and significantly negative for the flows from Latin America and Asia. This pattern could reflect the impact from restrictive welfare state immigration policies dominating the welfare magnet mechanism. Due to data availability, migration flows in the present approach are based on aggregate measures – no distinction can be made between the three main flows of migrants, being job- or study-related people (mostly intra-OECD), tied movers in relation to family reunions and finally refugees. In the long term, welfare magnet mechanisms might influence these flows in the direction pointed out in Borjas (1999b). In the short to intermediate term, however, job movers are only in incomplete ways entitled to social benefits in source countries, the flows of tied movers are by nature strongly influenced by the stock of immigrants in a destination country, that is, the network effect, and finally the flow of refugees consists of convention refugees, where entry depends on political decisions, and spontaneous individual asylum seekers, where the conditions for granting a residence permit depend on national immigration policies. To sum up, the evidence from the analysis of gross migration flows in twenty-seven OECD countries presented in this chapter shows that migration flows respond to economic differences across the countries and that many other non-economic measures like linguistic closeness, cultural distance, and costs of migration are important as well. We find strong network effects in driving international migration flows.

Appendix Description, definitions and sources of the Basic variables Migration flows (ij): Gross flow of migrants from country i to country j. Source: National statistical offices and “Trends in International Migration” SOPEMI 2000 OECD.

Migration into OECD countries 1990–2000

69

Stock of immigrants (ij): Stock of foreigners of country i – origin residing in country j. Source: National statistical offices and “Trends in International Migration” SOPEMI 2000 OECD. Population (i), Population (j): Total population is based on the de facto definition of population, which counts all residents regardless of legal status or citizenship – except for refugees not permanently settled in the country of asylum, who are generally considered part of the population of their country of origin. Source: World Bank. GDP (i), GDP (j): GDP per capita (constant 1995 international $) based on purchasing power parity (PPP). PPP GDP is gross domestic product converted to international dollars using purchasing power parity rates. An international dollar has the same purchasing power over GDP as the US dollar has in the United States. GDP at purchaser’s prices is the sum of gross value added by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the products. It is calculated without making deductions for depreciation of fabricated assets or for depletion and degradation of natural resources. Data are in constant 1995 international dollars. Source: World Bank, International Comparison Programme database. Unemployment rate (i), Unemployment rate (j): Unemployment, total (percent of total labor force). Unemployment refers to the share of the labor force that is without work but available for, and seeking employment. Definitions of labor force and unemployment differ by country. Source: World Bank: International Labor Organization, Key Indicators of the Labor Market database. Illiteracy rate (i): Adult illiteracy rate is the percentage of people ages 15 and above who cannot, with understanding, read and write a short, simple statement on their everyday life. Source: World Bank (United Nations Educational, Scientific, and Cultural Organization.) Public social Expenditure/GDP (j): Social expenditure is the provision by public institutions of benefits to, and financial contributions targeted at, households and individuals in order to provide support during circumstances which adversely affect their welfare, provided that the provision of the benefits and financial contributions constitutes neither a direct payment for a particular good or service nor an individual contract or transfer. Such benefits can be cash transfers, or can be the direct

70

Immigration and the Transformation of Europe

(“in-kind”) provision of goods and services. Public social expenditure is shown as a percentage of GDP (SNA93). Source: OECD Social Expenditure Database (SOCX). Tax Revenue/GDP (j): Tax revenue comprises compulsory transfers to the central government for public purposes. Compulsory transfers such as fines, penalties, and most social security contributions are excluded. Refunds and corrections of erroneously collected tax revenue are treated as negative revenue. Data are shown for central government only and as a percentage of GDP. Source: World Bank: International Monetary Fund, Government Finance Statistics Yearbook and data files, and World Bank and OECD GDP estimates. Distance between countries (ij): distance between capitals in km. Source: MapInfo, own calculations. Freedom house index (j): It represents scores of political rights, civil liberties, and freedom. These are measured on a scale of one to seven, with one representing the highest degree of freedom and seven the lowest. Source: Annual Freedom in the World Country Scores 1972–73 to 2001–02. Linguistic distance (ij): A dummy for a common language in two countries. This dummy has a value 1 for a common language in two countries and 0 for no common language. Source: Ethnologue: Languages of the World, 14th edn. www.ethnologue. com/web.asp Colony (ij): A dummy for countries ever in colonial relationship. This dummy has a value 1 for a common historical past of two countries, 1 and 0 otherwise. Source: The dataset freely available at the webpage of Andrew K. Rose and used for the paper: Rose, A. (2002): “Do We Really Know that the WTO Increases Trade?” NBER Working Paper No 9273. Neighbors (ij): A dummy for neighboring countries. This dummy has a value 1 for a common border between two countries 1 and 0 otherwise. Source: MapInfo, own calculations. Trade volume (ij): Trade volume represents bilateral trade flows that are based on IMF Direction of Trade data; the IMF data lists total trade values (both imports and exports) for all country pairs for all years. Source: IMF Direction of Trade Statistics, Yearbooks 1989–2001.

Migration into OECD countries 1990–2000

71

 Bauer, Thomas, Gil S. Epstein and Ira N. Gang 2002 “Herd Effects or Migration Networks? The Location Choice of Mexican Immigrants in the U.S.” IZA Discussion Paper No 551. Bonn: Institute for the Study of Labor, August ftp://ftp.iza.org/dps/dp551.pdf. Boeri, Tito, Gordon H. Hanson and Barry McCormick 2002 Immigration Policy and the Welfare System: A Report for the Fondazione Rodolfo Debenedetti. Oxford: Oxford University Press. Borjas, George J. 1987 “Self-Selection and the Earnings of Immigrants.” American Economic Review 77(4) (September): 531–53. Borjas, George J. 1994 “Immigration and Welfare, 1970–1990.” NBER Working Paper No 4872. Cambridge, MA: National Bureau of Economic Research, September. Borjas, George J. 1999a “Immigration and Welfare Magnets.” Journal of Labor Economics 17(4, Part 1) (October): 607–37. Borjas, George J. 1999b Heaven’s Door: Immigration Policy and the American Economy. Princeton, NJ: Princeton University Press. Borjas, George J. 1999c “The Economic Analysis of Immigration.” Chapter 28 in Handbook of Labor Economics, Vol. 3A, Orley Ashenfelter and David Card (eds). Amsterdam: Elsevier Science, North-Holland, 1697– 1760. Carrington, William J., Enrica Detragiache and Tara Vishwanath 1996 “Migration with Endogenous Moving Costs.” American Economic Review 86(4): 909–30. Chiquiar, Daniel and Gordon H. Hanson 2002 “International Migration, SelfSelection, and the Distribution of Wages: Evidence from Mexico and the United States.” 9242. NBER. Chiswick, Barry R. 1986 “Is the New Immigration Less Skilled Than the Old?” Journal of Labor Economics 4(2): 168–92. Chiswick, Barry R. 2000 “Are Immigrants Favorably Self-Selected? An Economic Analysis,” in Migration Theory: Talking Across Disciplines, Caroline Brettell and James Frank Hollifield (eds) New York: Routledge, 61–76. Chiswick, Barry R., and Timothy J. Hatton 2002 “International Migration and the Integration of Labor Markets.” IZA Discussion Paper No 559. Bonn: Institute for the Study of Labor. August. ftp://ftp.iza.org/dps/dp559.pdf. Epstein, Gil S. 2002 “Information Cascades and Decision to Migrate.” IZA Discussion Paper No 445. Bonn: Institute for the Study of Labor. March. ftp://ftp.iza.org/dps/dp445.pdf. Fertig, Michael and Christoph M. Schmidt 2000 “Aggregate-Level Migration Studies as a Tool for Forecasting Future Migration Streams.” IZA Discussion Paper No 183. Bonn: Institute for the Study of Labor. August. ftp://ftp.iza.org/dps/dp183.pdf. Hammarstedt, Mats 2000 “The Receipt of Transfer Payments by Immigrants in Sweden.” International Migration 38(2): 239–68. Harris, John R. and Michael P. Todaro 1970 “Migration, Unemployment and Development: A Two-Sector Analysis.” American Economic Review 60(1): 126–42.

72

Immigration and the Transformation of Europe

Hatton, Timothy J. and Jeffrey G. Williamson 2002 “What Fundamentals Drive World Migration?” 9159 National Bureau of Economic Research. Heitmueller, Axel 2003 “Coordination Failures in Network Migration.” IZA Discussion Paper No 770. Bonn: Institute for the Study of Labor. April. ftp://ftp.iza.org/dps/dp770.pdf. Hicks, Sir John R. 1963 The Theory of Wages. 2nd edn. London: Macmillan. Holmlund, Bertil 1984 Labor Mobility: Studies of Labor Turnover and Migration in the Swedish Labor Market. Stockholm: Almqvist & Wiksell International. Jackman, Richard and Sawas Savouri 1991 Regional Migration in Britain: An Analysis of Gross Flows Using NHS Central Register Data. London: Centre for Economic Performance, 27. Jasso, Guillermina and Mark R. Rosenzweig 1990 “Self-Selection and the Earnings of Immigrants: Comments.” American Economic Review 80(1): 298–304. Massey, Douglas S, Joaquin Arango, Graeme Hugo, Ali Kouaouci, Adela Pellegrino and J. Edward Taylor 1993 “Theories of International Migration: Review and Appraisal.” Population and Development Review 19(3): 431–66. Mayda, Anna Maria 2004 “International Migration: A Panel Data Analysis of Economic and Non-Economic Determinants.” Discussion Paper Series No 1590 Institute for the Study of Labor (12A). Germany: Bonn May. Mincer, Jacob 1978 “Family Migration Decisions.” The Journal of Political Economy 86(5): 749–73. Pedersen, Peder J., Mariola Pytlikova and Nina Smith 2004 “Selection or Network Effects? Migration Flows into 27 OECD Countries, 1990–2000.” IZA Discussion Paper No 1104. Bonn: Institute for the Study of Labor. April. ftp://ftp.iza.org/dps/dp1104.pdf. Riphahn, Regina T. 1999 “Immigrant Participation in Social Assistance Programs: Evidence from German Guestworkers.” CEPR Discussion Series No 2318 (December). London: Center for Economic Policy Research. Roy, Andrew D. 1951 “Some Thoughts on the Distribution of Earnings.” Oxford Economic Papers, New Series, 3(2) (June 1951): 135–46. Sjaastad, Larry A. 1962 “The Costs and Returns of Human Migration.” The Journal of Political Economy 70(5): 80–93. Stark, Oded 1991 The Migration of Labor. Oxford: Blackwell. Storesletten, Kjetil 2003 “Fiscal Implications of Immigration–Net Present Value Calculation.” Scandinavian Journal of Economics, 105(3): 487–506. Urrutia, C. 2001 “On the Self-Selection of Immigrants.” Manuscript, Universidad Carlos III de Madrid. Wadensjo, ¨ Eskil and Helena Orrje 2002 Immigration and the Public Sector in Denmark. Aarhus: Aarhus University Press. Zavodny, Madeline 1997 “Welfare and the Locational Choices of New Immigrants.” Economics Review (Federal Reserve Bank of Dallas), Second Quarter: 2–10.

Population (in thousands)

18 000 (725) 1 548 89–00 (12)

1 340 486 120 581 861 306 (1 575 (219 624) (16 602) 127)

Sum of immigration stock∗ ∗ ∗ ∗

7 959 (153) 1 548 89–00 (12)

10 100 (99) 1 548 89–00 (12)

6 508 701 (775 077) 78 400 (7 058) 1 548 89–00 (12)

548 757 (1 222 297) 57 700 (7387) 1 548 89–00 (12)

58 697 (23093) 5 088 (68) 1 548 89–00 (12)

225 023 (38 322) 5 227 (69) 1 548 89–00 (12)

117902 (89 299) 10 300 (18) 1 548 89–00 (12)

1 157 930 (1 620 004) 29 100 (1 089) 1 548 89–00 (12)

57 261 (204 320) 1 364 89–00 (12)

41 416 (113 211) 159 90, 99 (2)

649 (1 758) 1 085 89–00 (12)

1.776 (3 918) 1 520 89–00 (12)

1 582 (6 389) 894 93–00 (8)

50 787 (1 595 564) 377 91,96,01 (3)

15 755 (38755) 656 89–00 (12)

5 811 (19 172) 249 91,96– 99,00 (6)

148 943 (236 324) 108 90,95-96, 99-00 (5)

Immigration stock ∗ ∗ ∗

725 797 (192 338)

78 465 (20 171)

15 951 (31 208) 546 89–00 (12)

8 262 (2 136)

824 (1 604) 1 143 89–00 (12)

26 091 (6 413)

65 (264) 1 524 89–00 (12)

6 973 (5 085)

206 (560) 1 520 89 -00 (12)

176 902 (55 537)

83 (398) 1 003 93–00 (8)

1 151 (3 395) 158 89–99 (11)

8 084 (14 883) 74 94–97,99 (5)

10 400 (166) 1 548 89–00 (12)

10 200 (103) 1 548 89–00 (12)

112 498 49 850 (65 207) (59 270)

1 188 (2 613) 1 136 89–98 (9)

17 297 15 158 (11 208) (9 125)

184 (458) 1 134 89–97 (8)

–

3 613 (89) 1 548 89–00 (12)

266 (8) 1 548 89–00 (12)

(cont.)

57 300 (342) 1 548 89–00 (12)

650 211 (401 031)

15 034 (22 491) 519 91–00 (10)

– – – 82 (272) 1 528 89–00 (12) 10 457 (3 145)

14 515 (27 865)

3412 (6 251) 146 96–00 (5) 5 875 (5 149)

5 036 (2 660) 14 94–00 (7) 1 325 (1 514)

25 (96) 630 92–00 (9)

Germany> Greece> Hungary> Iceland> Ireland> Italy>

48 656 (12 478)

58 409 (19 760)

Sum of immigration flows ∗ ∗

2 029 (4 267) 1 046 89–00 (12)

Czech Republic> Denmark> Finland> France>

29 038 (29 310)

3 830 (4 320) 183 89–00 (12)

927 (1 707) 630 90–00 (11)

Belgium> Canada>

Immigration flows∗

1 508 (2 446) 238 86–00 (5)

Mean (std) No of obs years with information Australia> Austria>

Table 3.A.1 Descriptive statistics of basic variables for OECD destination countries (means, standard deviations, and number of years observed in the data set)

Appendix

21 219 (1 757) 1 548 89–00 (12)

8.153 (1.472) 1 548 89–00 (12)

29.073 (1.096) 1 419 89–99 (11)

5

GDP per capita PPP (constant 1995 int$)

Unemployment rate (% of the labour force)

Tax revenue (% of GDP)

No of years with complete info on all variables

3

8

12

12

2

12

8

5

9

0

5

41.633 (1.952) 1 548 89–00 (12) 33.275 (1.170) 1 548 89–00 (12) 32.633 (2.068) 1 548 89–00 (12) 42.620 (3.033) 1 548 89–00 (12) 32.142 (3.210) 1 548 89–00 (12) 36.800 (1.782) 1 548 89–00 (12) 44.158 (1.059) 1 548 89–00 (12) 45.642 (1.194) 1 548 89–00 (12)

48.892 (1.151) 1 548 89–00 (12)

40.025 (14.401) 1 032 93–00 (8)

37.167 (0.711) 1 548 89–00 (12)

44.575 (1.159) 1 548 89–00 (12)

42.533 (1.342) 1 548 89–00 (12)

11

10.624 (1.186) 1 548 89–00 (12) 11.613 (3.723) 1 548 89–00 (12) 3.044 (1.336) 1 548 89–00 (12) 8.562 (2.703) 1 419 90–00 (11) 9.509 (1.506) 1 548 89–00 (12) 7.592 (1.146) 1 548 89–00 (12)

10.794 (1.180) 1 548 89–00 (12)

11.007 (4.495) 1 548 89–00 (12)

7.146 (1.647) 1 548 89–00 (12)

4.823 (2.322) 1 419 90–00 (11)

9.142 (1.432) 1 548 89–00 (12)

8.431 (1.264) 1 548 89–00 (12)

5.013 (0.602) 1 548 89–00 (12)

5

20 989 (953) 1 548 89–00 (12) 18 259 (4 597) 1 548 89–00 (12) 22 866 (1 584) 1 548 89–00 (12)

4 612 (3 816) 89–00 (12)

13 225 (906) 1 548 89–00 (12)

30 047 (1 287) 1 548 89–00 (12)

Germany> Greece> Hungary> Iceland> Ireland> Italy>

21 095 (948) 1 548 89–00 (12)

19 864 (1 529) 1 548 89–00 (12)

23 899 (1 529) 1 548 89–00 (12)

11 743 (749) 1 290 91–00 (10)

22 480 (1 451) 1 548 89–00 (12)

Czech Republic> Denmark> Finland> France>

21 632 (1 259) 1 548 89–00 (12)

Belgium> Canada>

22 591 (1 597) 1 548 89–00 (12)

Mean (std) of obs years with information Australia> Austria>

Table 3.A.1 (cont.)

15 423 (16 862) 163 89–00 (12)

209 494 (36 018)

34 548 (110 834) 246 90,95, 97–00 (6)

708 234 (720 044)

Flows of immigrants to the country ∗

Sum of immigration flows to the country ∗ ∗

Stock of immigrants in the country ∗ ∗∗

Sum of immigration stock in the country ∗ ∗ ∗ ∗

Mean (std) No of obs years with information Japan

Table 3.A.1 (cont.)

695 195 (620880)

118 018 154 599 (206 090) (16 952)

62 295 135 780 (278 686) (34 750)

12 672 (12 812)

446 883 871 670 (243 560) (75 939)

1 256 522 (91 264)

(cont.)

2 815 967 15 500 (1452974) 000 (11 100 000)

300 190 (791 623) 620 90, 94–00 (8) 51 590 (90 234) 655 92–00 (9) 10 356 (40 218) 1456 89–00 (12) 9 056 (23 666) 1155 89–00 (12) 11 997 (19 930) 447 91–00 (10) 201 (696) 757 95–00 (6) 6 172 (8 300) 264 89–00 (12)

20 204 (55 505) 37 01 (1)

2 798 (4 072) 663 89–00 (12)

11 330 (35 080) 125 91,96,01 (3)

10 792 (31 109) 773 90,95–00 (7)

3 825 (9 460) 387 89–00 (12)

123 348 (17 246)

922 410 (332 837)

51 805 (29 475)

108 145 (35 081)

43 821 (11 941)

52 047 (86 327)

704 (998)

7 566 (35 316) 1463 89–00 (12)

4 153 (5 928)

879 (1 473) 707 91–00 (10)

889 (3 088) 1460 89–00 (12) 365 (1 216) 1440 89–00 (12)

5 118 (1 917)

1190 (5030) 525 89–00 (12)

United States

28 101 (4 969)

17 (90) 499 97–00 (4)

209 (534) 238 92–00 (9)

409 (516) 150 89–00 (12)

529 (936) 638 89–00 (12)

United Kingdom Switzerland

Sweden

34 996 (4 373)

3 962 (4 944) 106 89–00 (12)

Spain

64 350 (18 716)

938 (1 754) 823 95–00 (6)

Slovak Republic

Portugal

Poland

Norway

6 867 (394)

886 (814) 93 89–00 (12)

New Luxembourg Netherlands Zealand

22 476 (964) 1 548 89–00 (12)

3.12 (0.928) 1 548 89–00 (12)

28.158 (1.350) 1 548 89–00 (12)

GDP per capita (constant 1995 int $)

Unemployment rate (% of the labour force)

Tax revenue (% of GDP)

country i

12

1

9

4

10

12

12

9

8

26.942 (2.118) 1 548 89–00 (12)

35.492 (1.210) 1 548 89–00 (12) 32.717 (1.759) 1 548 89–00 (12) 50.925 (1.984) 1 548 89–00 (12) 33.692 (0.831) 1 548 89–00 (12) 33.256 (1.817) 645 96–00 (5) 32.058 (1.679) 1 548 89–00 (12)

37.584 (1.960) 1 290 91–00 (10)

41.492 (0.846) 1 548 89–00 (12)

36.900 (1.012) 1 548 89–00 (12)

42.892 (1.423) 1 548 89–00 (12)

41.100 (1.188) 1 548 89–00 (12) 3

5.574 (1.054) 1 548 89–00 (12) 7.552 (1.709) 1 548 89–00 (12) 3.048 (1.638) 1 548 89–00 (12) 5.675 (2.418) 1 548 89–00 (12) 18.868 (3.171) 1 548 89–00 (12) 12.886 (3.037) 1 290 91–00 (10)

5.464 (1.180) 1 548 89–00 (12)

12.507 (2.415) 1 548 90–00 (11)

4.709 (0.965) 1 548 89–00 (12)

7.698 (1.499) 1 548 89–00 (12)

5.591 (1.503) 1 548 89–00 (12)

2.417 (0.787) 1 548 89–00 (12)

6

28 069 (1 872) 1 548 89–00 (12) 20 020 (1 467) 1 548 89–00 (12) 25 670 (422) 1 548 89–00 (12) 19 930 (1 124) 1 548 89–00 (12)

15 214 (1 262) 1 548 89–00 (12)

9 301 (960) 1 548 89–00 (12)

13 460 (1228) 1 548 89–00 (12)

7 332 (975) 1 548 89–00 (11)

27 812 (2 912) 1 548 89–00 (12)

16 568 (1121) 1 548 89–00 (12)

21 770 (1 745) 1 548 89–00 (12)

35 351 (6 780) 1 548 89–00 (12)

12

262 000 (8 789) 1 548 89–00 (12) 58 500 (703) 1 548 89–00 (12) 697 (168) 1 548 89–00 (12)

8 744 (125) 1 548 89–00 (12)

39 200 (217) 1 548 89–00 (12)

5 366 (20 629) 1 548 89–00 (12)

9 894 (59) 1 548 89–00 (12)

38 500 (227) 1 548 89–00 (12)

4 350 (83) 1 548 89–00 (12)

3 617 (165) 1 548 89–00 (12)

15 400 (327) 1 548 89–00 (12)

United States

United Kingdom

Switzerland

Sweden

Spain

Slovak Republic

Portugal

Poland

Norway

409 (18) 1 548 89–00 (12)

New Luxembourg Netherlands Zealand

∗ mean and std for each particular migration flow from country j to country i ∗ ∗ mean and std for the sum of migration flows to country i ∗ ∗ ∗ mean and std for stock of immigrants originating from country j residing in ∗ ∗ ∗ ∗ mean and std for the sum of stocks of immigrants in country i

No of years with 6 complete info on all variables

125 000 (1 155) 1 548 89–00 (12)

Population (in thousands)

Mean (std) No of obs years with information Japan

Table 3.A.1 (cont.)

– –

21 600 (3 171) (12)

3 300 (85) (12)

28 100 (1 834) (12)

11 500 (1 232) (12)

35 000 (1 563) (12)

18 000 (725) (12)

7 959 (153) (12)

Afghanistan

Albania

Algeria

Angola

Argentina

Australia

Austria

8.15 (1.47) (12)

5.01 (0.60) (12)

225901 (1 756) (12)

– –

Chinese Taipei

1.10 (0.00) (12)

0.00 (0.00) (12)

14 300 (725 150) (12)

Chile

1.10 (0.00) (12)

0.00 (0.00) (12)

6 598 (672) (12)

Chad

2.03 (0.62) (12)

73 152 (115 012) (12)

29 100 (1 089) (12)

13 100 (1 131) (12)

10 800 (990) (12)

3.75 (0.39) (12)

12.19 (4.53) (12)

Canada

Cameroon

Cambodia

Mean (std) (Years with Population information) (in thous)

Cape Verde

6.45 (1.03) (12)

4.63 (1.49) (12)

7.52 (0.41) (12)

Freedom House Index

6.95 (0.53) (12)

52.00 (0.00) (12)

40.88 (4.96) (12)

19.44 (2.72) (12)

– –

Illiteracy rate

– –

24.96 (4.08) (12)

12.33 (3.69) (12)

– –

Unemployment rate

21220 (1760) (12)

10375 (1076) (12)

1974 (283) (12)

4811 (181) (12)

2725 (500) (12)

GDP per cap, PPP, (1995 int $)

Mean (std) (Years with Population information) (in thous)

– –

– –

20.00 (0.00) (12)

3.12 (1.15) (12)

2.46 (0.62) (12)

5.12 (0.60) (12) 6.13 (1.85) (12) 7002 (1132) (12)

6.71 (0.41) (12) 65.83 (5.19) (12)

– – 837 (60) (12)

2.09 (1.79) (12)

1.10 (0.00) (12)

6.96 (0.48) (12)

6.52 (1.01) (12)

(cont.)

Freedom House Index

31.44 (3.53) (12)

0.00 (0.00) (12)

36.00 (4.68) (12)

35.85 (2.07) (12)

Illiteracy rate

– –

9.14 (1.43) (12)

– –

– –

Unemployment rate

3512 (422) (12)

22480 (1453) (12)

1691 (146) (12)

1428 (143) (12)

GDP per cap, PPP, (1995 int $)

Table 3.A.2 Descriptive statistics of basic variables for source countries (mean, standard deviation, and number of years with information)

2098 (454) (9)

7 602 (312) (9)

121 000 (7 317) (12)

10 200 (74) (9)

10 100 (99) (12)

5 413 (536) (12)

7 343 (601) (12)

3 957 (340) (9)

160 000 (7 695) (12)

8 385 (177) (12)

Azerbaijan

Bangladesh

Belarus

Belgium

Benin

Bolivia

Bosnia Herzegovina

Brazil

Bulgaria

5754 (582) (12)

4 327 (209) (12)

14.26 (4.95) (11)

6.54 (2.06) (9)

39.28 (0.31) (4)

5.74 (2.08) (8)

2079 (84) (12)

– –

68.83 (3.77) (12)

– –

806 (47) (12)

2.23 (0.43) (12)

17.08 (1.52) (12)

1.50 (0.00) (9)

18.500 (2.54) (12)

0.00 (0.00) (12)

0.44 (0.07) (9)

Czech Republic

5.93 (0.56) (9)

2.82 (1.51) (12)

Dominican Republic

Denmark

Cyprus

2.06 (0.44) (12)

2.77 (0.52) (12)

11 000 (180) (12)

Cuba

3.51 (2.20) (12)

7 771 (458) (12)

5 227 (69) (12)

10 300 (18) (12)

723 (28) (12)

4 586 (177) (11)

13 700 (1 454) (12)

39 000 (2 548) (12)

1 210 000 (43 300) (12)

Croatia

Cote ˆ d’Ivoire

Colombia

China

Mean (std) (Years with Population information) (in thous)

1.14 0.05 (12)

5.60 (0.98) (9)

3.06 (0.98) (12)

6.28 (0.43) (9)

−

62.18 (2.21) (12)

Freedom House Index

Illiteracy rate

8.43 (1.26) (12)

1.88 (1.20) (9)

26.64 (14.52) (8)

0.69 (0.40) (9)

Unemployment rate

21632 (1261) (12)

4030 (542) (9)

1214 (118) (12)

GDP per cap, PPP, (1995 int $)

Mean (std) (Years with Population information) (in thous)

Table 3.A.2 (cont.)

4.82 (2.32) (11)

11743 (750) (11)

4453 (604) (12)

17.25 (2.25) (8)

7.15 (1.65) (12)

2.58 (0.56 (10)

12976 (1468) (12)

23899 (1531) (12)

– –

15.06 (3.82) (11)

– –

11.83 (4.29) (12)

2.78 (0.29) (12)

Unemployment rate

– –

4 454 (623) (11)

1530 (97) (12)

5618 (208) (12)

2397 (705) (12)

GDP per cap, PPP, (1995 int $)

1.90 (1.48) (12)

0.00 (0.00) (12)

18.66 (1.48) (12)

2.79 (0.873) (12)

1.10 (0.00) (12)

1.11 (0.03) (12)

4.33 (1.39) (12)

0.00 (0.00) (12)

7.7 (0.00) (12)

4.31 (0.78) (11)

6.46 (0.06) (12)

3.57 (0.80) (12)

7.68 (0.04) (12)

Freedom House Index

4.16 (0.53) (12)

2.47 (0.51) (11)

56.50 (3.46) (12)

10.01 (1.12) (12)

19.76 (2.51) (12)

Illiteracy rate

9 910 (814) (12)

6 077 (469) (12)

5 736 (411) (12)

1 464 (73) (9)

58 000 (4 605) (12)

10 600 (63) (12)

133 (209) (12)

5 088 (68) (12)

– –

Burkina Faso

Burundi

El Salvador

Estonia

Ethiopia

Fed Rep of Yugoslavia

Fiji

Finland

Former USSR

11.01 (4.50) (12)

19865 (1531) (12)

– –

5.83 (0.34) (7)

4314 (330) (12)

7122.58 (1457) (6)

23.42 (2.62) (11)

0.41 (0.89) (6)

5.95 (0.56) (2)

Iran

Indonesia

59 500 (3 236) (12)

196 000 (11 100) (12)

939 000 (57 4000) (12)

India

4.83 (1.22) (12)

9.43 (1.53) (12) 1.10 (0.00) (12)

266 (8) (12)

Iceland

6.11 (0.71) (12)

1.50 (0.00) (12)

0.00 (0.00) (12)

10 200 (103) (12)

Hungary

5.85 (1.17) (12)

66.73 (3.565) (12)

– –

6 264 (416) (12)

5 557 (531) (12)

58 800 (3 973) (12)

11 300 (825) (12)

Hong Kong

Honduras

Egypt

Ecuador

2.68 (1.69) (9)

0.20 (0.01) (9)

2.99 (0.49) (12)

7.30 (0.48) (12)

58.17 (3.85) (12) 24.55 (2.21) (12)

5.78 (0.80) (12)

80.42 (2.57) (12)

7.19 (4.46) (9)

8.21 (0.92) (10)

– –

– –

– –

567 (40) (12)

7984 (1149) (9)

3834 (362) (12)

682 (110) (12)

867 (43) (12)

3.04 (1.34) (12)

22866 (1586) (12)

4849 (419) (12)

2589 (306) (12)

– –

17.86 (2.48) (7)

– –

8.56 (2.70) (11)

9718 (856) (12)

1817 (242) (12)

2.86 (1.57) (12)

3.77 (0.63) (9)

9.32 (1.41) (9)

8.56 (1.92) (11)

20883 (1649) (12)

2336 (40) (12)

2830 (207) (12)

3127 (77) (12)

1.10 (0.00) (12)

0.00 (0.00) (12)

30.58 (4.50) (12)

16.99 (2.57) (12)

6.62 (0.08) (12)

6.42 (1.31) (12)

3.03 (0.89) (12)

1.71 (0.89) (12)

0.81 (0.10) (12)

47.12 (2.74) (12)

– –

2.72 (0.49) (12)

6.32 (0.48) (12)

2.37 (0.29) (12)

8.41 (1.26) (12)

28.75 (2.42) (12)

49.24 (2.84) (12)

10.50 (1.38) (12)

(cont.)

– –

– –

57 700 (739) (12)

– –

5 371 (67) (9)

78 400 (7 058) (12)

17 100 (1 432) (12)

9 890 (897) (12)

10 400 (166) (12)

Former Yugoslavia

France

Gaza Strip

Georgia

Germany

Ghana

Guatemala

Greece

3.97 (0.87) (12)

35.50 (2.70) (12) 1.27 (0.05) (12)

Kazakhstan

Jordan

3.76 (0.53) (12)

– –

3380 (157) (12)

9.51 (1.51) (12)

Japan

5.02 (1.35) (12)

– –

1625 (93) (12)

13226 (907) (12)

Jamaica

1.20 (0.00) (12)

0.00 (0.00) (12)

7.59 (1.15) (12)

21780 (1261) (12) 35.67 (4.65) (12)

Italy

Israel

Ireland

Iraq

16 000 (472) (9)

4 056 (569) (12)

125 000 (1 155) (12)

2 527 (788) (12)

57 300 (342) (12)

5 594 (511) (12)

3 613 (89) (12)

21 000 (1 783) (12)

Mean (std) (Years with Population information) (in thous)

4.57 (1.04) (9)

– –

1.20 (0.00) (12)

– –

Freedom House Index

– –

– –

0.00 (0.00) (12)

1.10 (0.89) (12)

Illiteracy rate

– –

– –

10.79 (1.18) (12)

– –

Unemployment rate

2327 (1348) (9)

– –

21094 (740) (12)

GDP per cap, PPP, (1995 int $)

Mean (std) (Years with Population information) (in thous)

Table 3.A.2 (cont.)

3.12 (0.93) (12) 14.40 (0.99) (3)

22476 (966) (12) 3621 (95) (12)

11.64 (2.27) (5)

15.93 (0.45) (11)

3445 (95) (12)

3938 (604) (9)

10.62 (1.19) (12)

8.76 (1.35) (12)

11.61 (3.72) (12)

– –

Unemployment rate

20989 (954) (12)

16279 (1478) (12)

18259 (4602) (12)

– –

GDP per cap, PPP, (1995 int $)

1.35 (0.38) (12)

0.00 (0.00) (12)

0.91 (0.21) (9)

6.26 (0.43) (9)

4.76 (0.90) (12)

2.24 (0.05) (12)

15.67 (1.69) (12)

14.25 (2.900) (12)

1.18 (0.06) (12)

1.59 (0.43) (12)

1.12 (0.04) (12)

7.69 (0.03) (12)

Freedom House Index

1.94 (0.27) (12)

7.44 (1.33) (12)

0.00 (0.00) (12)

49.94 (3.67) (12)

Illiteracy rate

7 265 (526) (12)

4 753 (400) (12)

2 534 (109) (9)

4 028 (239) (12)

4 732 (326) (12)

3 629 (69) (9)

Haiti

Laos

Latvia

Lebanon

Libya

Lithuania

Luxembourg 409 (18) (12)

823 (80) (12)

936 (403) (12)

GuineaBissau

35351 (6788) (12)

7634 (1373) (9)

– –

2.42 (0.79) (12)

11.52 (6.15) (9)

– –

0.00 (0.00) (12)

0.57 (0.09) (9)

26.33 (4.22) (12)

1.10 (0.00) (12)

Philippines

Peru

69 000 (5 062) (12)

23 800 (1 417) (12)

4 652 (378) (12)

Paraguay

7.61 (0.30) (12) 1.47 (0.45) (9)

124 000 (10 500) (12)

Pakistan

6.40 (0.30) (12)

17.04 (1.98) (12)

18.76 (0.21) (5)

3379 (774) (12)

4 350 (83) (12)

Norway

2.23 (0.85) (9)

0.21 (0.01) (9)

Korea South 44 900 (Rep. of (1 531) Korea) (12) 110 000 (10 600) (12)

11.43 (5.54) (9)

6669 (1707) (9)

26 400 (2 332) (12)

Korea North 21 100 (Dem. Rep.) (858) (12)

Kenya

Nigeria

6.04 (1.27) (12)

4.81 (1.48) (12)

6.68 (0.41) (12)

7.38 (0.39) (12)

55.76 (3.49) (12)

67.83 (3.83) (12)

– –

58.30 (4.29) (12)

– –

– –

– –

– –

1149 (162) (12)

1738 (272) (12)

1677 (86) (12)

6 505 (574) (12)

Guinea

3484 (105) (12)

4044 (337) (12)

8.70 (0.75) (12)

8.04 (1.17) (9)

5.86 (1.16) (7)

6.23 (1.02) (12)

12.46 (1.54) (12)

8.33 (1.03) (12)

61.07 (2.71) (12)

5.21 (1.05) (12) 1634 (72) (12) 4413 (92) (12)

1.10 (0.00) (12)

0.00 (0.00) (12) 4.71 (0.97) (12) 27812 (2915) (12)

(cont.)

2.74 (0.51) (12)

4.66 (1.19) (12)

4.24 (0.83) (12)

4.56 (1.21) (12)

6.42 (1.13) (12)

2.23 (0.05) (12)

7.70 (0.00) (12)

6.49 (0.78) (12)

44.42 (5.34) (12)

3.00 (0.00) (12)

3.17 (0.69) (12)

23.92 (4.00) (12)

7.05 (5.74) (4)

3.23 (1.57) (12)

– –

21.30 (0.00) (1)

798 (22) (12)

10856 (1861) (12)

– –

987 (36) (12)

782 (51) (12)

13 200 (1 332) (12)

9 231 (629) (12)

20 900 (1 768) (12)

9 516 (815) (12)

26 600 (1 606) (12)

89 300 (5 872) (12)

Madagascar

Malawi

Malaysia

Mali

Morocco

Mexico

– –

– –

43 900 (2 564) (12)

Myanmar (Burma)

17.44 (1.39) (12)

61.89 (3.59) (12)

– –

656 (83) (12)

Mozambique 16 000 (1 267) (12) – –

4.46 (1.48) (12)

3.86 (0.78) (12)

10.46 (1.268) (12)

3.52 (1.23) (12)

7286 (405) (12)

Senegal

8 468 (787) (12)

1259 (56) (12)

11395 (459) (12)

Saudi Arabia 32 800 (49 600) (12)

67.62 (3.10) (12)

29.50 (3.46) (12)

– – – –

37.00 (0.00) (1)

40.50 (4.78) (12)

0.62 (0.11) (9)

2.47 (0.38) (12)

10.41 (1.74) (12)

0.34 (0.06) (12)

Illiteracy rate

– –

– –

977 (121) (12) – –

8.84 (3.96) (10)

7.44 (2.14) (10)

5.46 (1.18) (12)

12.51 (2.42) (11)

7130 (1462) (9)

5379 (328) (12)

13460 (1230) (12)

7332 (976) (11)

GDP per cap, PPP, (1995 Unemployment int $) rate

Sao Tome 25 and Principe (38) (12)

6 843 (792) (12)

Rwanda

5.38 (0.52) (12)

56.63 (3.54) (12)

18.03 (2.44) (10)

3142 (105) (12)

22 700 (257) (12)

9 894 (59) (12)

38 500 (227) (12)

Russian Fed. 147 000 Rep. (1 232) (9)

Romania

Portugal

Poland

Mean (std) (Years with Population information) (in thous)

3.96 (1.86) (12)

5.05 (0.47) (12)

4.79 (2.37) (12)

3.41 (1.24) (12)

Freedom House Index

78.25 (2.42) (12)

16.14 (2.36) (12)

44.50 (2.82) (12)

38.17 (2.92) (12)

Illiteracy rate

– –

3.55 (1.16) (9)

– –

– –

Unemployment rate

654 (36) (12)

6658 (1107) (12)

510 (35) (12)

GDP per cap, PPP, (1995 int $)

Mean (std) (Years with Population information) (in thous)

Table 3.A.2 (cont.)

4.31 (0.29) (12)

7.59 (0.27) (12)

2.18 (1.77) (12)

7.09 (0.53) (12)

4.24 (1.10) (9)

4.02 (1.80) (12)

1.12 (0.04) (12)

1.88 (0.87) (12)

Freedom House Index

20 300 (1 675) (12)

15 400 (327) (12)

3 617 (165) (12)

9 031 (1 069) (12)

18 300 (811) (12)

134 (162) (12)

8 744 (125) (12)

697 (168) (12)

14 400 (1 410) (12)

Nepal

Netherlands

New Zealand

Niger

Sri Lanka

Suriname

Sweden

Switzerland

Syria

2847 (258) (12)

6.30 (0.51) (2)

3.05 (1.64) (12)

5.68 (2.42) (12)

19929 (1125) (12)

25670 (422) (12)

12.61 (2.84) (8)

11.91 (2.40) (11)

2595 (343) (12)

– –

– –

7.70 (1.50) (12)

5.59 (1.50) (12)

1.10 (0.00) (1)

743 (46) (12)

16568 (1123) (12)

21769 (1747) (12)

1070 (90) (12) 7 930 (562) (12)

5 366 (20 629) (10)

30.74 (3.35) (12)

7.70 (0.00) (12)

1.10 (0.00) (12)

1.10 (0.00) (12)

0.00 (0.00) (12) 0.00 (0.00) (12)

3.48 (1.12) (12)

30.00 (0.00) (1)

Venezuela

Uzbekistan

United States

22 100 (1 613) (12)

22 900 (1 455) (9)

262 000 (8 789) (12)

58 500 (703) (12)

51 000 (1 051) (9)

Ukraine

4.05 (0.53) (12)

9.95 (1.02) (12) United Kingdom

39 200 (217) (12)

Spain

South Africa 39 400 (2 611) (12)

Somalia

Slovak Republic

5.92 (1.58) (12)

1.10 (0.00) (12)

1.10 (0.00) (12)

3.392 (0.62) (12)

86.67 (1.55) (12)

0.00 (0.00) (12)

0.00 (0.00) (12)

64.50 (3.91) (12)

7.55 (1.71) (12) 5.57 (1.05) (12)

20020 (1469) (12) 28069 (1875) (12)

5537 (251) (12)

10.23 (2.10) (11)

5.79 (4.62) (8)

7.16 (4.38) (8)

5178 (1820) (9)

1349 (116) (9)

18.87 (3.17) (12)

1.19 (0.03) (12) 1.10 (0.00) (12)

0.00 (0.00) (12) 0.00 (0.00) (12)

9.36 (1.28) (12)

(cont.)

2.66 (0.97) (12)

7.18 (0.72) (9)

3.98 (0.99) (9)

0.49 (0.07) (9)

1.10 (0.20) (9)

1.17 (0.05) (12)

3.13 (2.14) (12)

7.617 (0.28) (12)

2.55 (1.34) (10)

3.08 (0.49) (12)

16.93 (1.41) (12)

60.00 (0.00) (1)

– – 21.61 (4.22) (8)

0.00 (0.00) (10)

12.89 (3.04) (10)

15832 (1264) (12)

8713 (311) (12)

– –

9300 (961) (10)

1139 (582) (9)

5 763 (326) (9)

29 200 (2 861) (12)

58 700 (1 652) (12)

8 977 (478) (12)

61 000 (3 721) (12)

19 000 (2 032) (12)

Tajikistan

Tanzania

Thailand

Tunisia

Turkey

Uganda

5.63 (0.84) (12)

38.83 (3.85) (12)

3.08 (2.27) (5)

945 (125) (12)

11 300 (828) (12)

Zimbabwe

4.03 (1.08) (12)

18.69 (2.52) (12)

7.47 (0.76) (12)

5194 (382) (12)

8 844 (806) (12)

Zambia

6.23 (0.48) (12)

35.73 (4.11) (12)

6.49 (0.69) (7)

4807 (515) (12)

44 600 (4 881) (12)

14 800 (1 843) (12)

73 300 (4 217) (12)

Zaire (Dem. Rep. of Congo)

Yemen

Vietnam

Mean (std) (Years with Population information) (in thous)

3.19 (1.10) (12)

6.01 (0.69) (12)

6.72 (1.32) (9)

Freedom House Index

6.15 (1.10) (12)

31.58 (4.18) (12)

1.33 (0.37) (9)

Illiteracy rate

1.87 (0.80) (12)

3.30 (0.14) (3)

1.63 (0.94) (6)

Unemployment rate

5247 (792) (12)

463 (12) (12)

GDP per cap, PPP, (1995 int $)

Mean (std) (Years with Population information) (in thous)

Table 3.A.2 (cont.)

2482 (135) (12)

5.97 (0.78) (3)

13.70 (1.31) (2)

787 (79) (12)

– –

11.50 (0.00) (1)

– –

Unemployment rate

1050 (323) (12)

660 (64) (11)

1415 (268) (12)

GDP per cap, PPP, (1995 int $)

5.80 (0.46) (12)

4.66 (1.54) (12) 27.50 (3.46) (12) 15.75 (2.81) (12)

7.33 (0.46) (12)

5.75 (0.57) (10)

7.69 (0.03) (12)

Freedom House Index

46.17 (4.82) (12)

60.92 (4.74) (12)

8.14 (1.01) (12)

Illiteracy rate

4

Divergent patterns in immigrant earnings across European destinations Alicia Adser`a and Barry R. Chiswick

Introduction Western Europe, which used to think of itself as a region of emigration, has experienced substantial net immigration in the last four decades from the lesser-developed countries of Asia, Africa, Latin America, and in the last decade, from the former Eastern bloc countries. As a result, immigration has become an important socioeconomic and public policy issue in all of the developed European economies. Understanding how well immigrants from different origins adapt to diverse labor markets across Europe, and whether labor market outcomes, such as earnings, occupation or unemployment among immigrants differ by country of destination, country of origin and gender is of public policy interest. The analysis of earnings of the immigrants began with the study of white men in the United States (Chiswick 1978), but was quickly followed by analyses for a range of racial and ethnic groups, countries of origin and for women (Chiswick 1979, 1980; Long 1980). While the literature has spread to nearly all destination countries, most studies deal with immigrants in the English-speaking developed countries of overseas settlement (United States, Canada, and Australia) and tend to concentrate on only one country, thereby making comparisons across destinations very difficult. This study builds on the existing literature, but takes a different approach. Unlike the standard literature on the adjustment of immigrants that focuses on one destination at a time, and often one origin and one gender at a time, this study emphasizes the comparison across destination countries, as well as across countries of origin and gender.1 Moreover, 1

Studies that compare immigrants across multiple destinations include Antecol, CobbClark and Trejo (2002); Boyle, Cooke, Halfacree, and Smith (2001); and Adser`a and Chiswick (forthcoming). Notable exceptions that study the earnings and/or labor supply of immigrant women include Chiswick (1980, chapter 9); Antecol et al. (2002); Baker and Benjamin (1997); Biswal (1999); Boyle et al. (2001); Cobb-Clark and Connolly (2001); Duleep and Sanders (1993); Evans (1984); LeClere and McLaughlin (1997); MacPherson and Stewart (1989); and Reimers (1985).

85

86

Immigration and the Transformation of Europe

while the literature on immigrant adjustment is dominated by research on the English-speaking countries of overseas settlement (i.e. the United States, Canada, and Australia), the analyses in this study focus on western Europe, a region that in the past few decades has changed from a major source to a major destination for international migrants. It does not consider immigrants of the same gender, age and years of schooling as homogeneous, but rather focuses on the differences by country of destination and by country of origin, and hence on differences by race and ethnicity. Finally, another differentiating feature is on the analysis of the earnings of both immigrant men and women, the latter being a topic that has received too little attention. This chapter uses the 1994–2000 waves of the European Community Household Panel (ECHP) to conduct a systematic analysis of individual earnings from work among immigrants as compared to native-born workers across destinations, origins and genders. The ECHP combines a variety of surveys for the European countries. It is the first household survey that provides the necessary data for a comparative analysis of the adjustment and impact of immigrants, not only across broad geographic areas of origin but also across the European destination countries. Data and methodology The ECHP This study uses the 1994–2000 waves of the European Community Household Panel (ECHP) Survey to conduct a systematic analysis of individual earnings from work of immigrants as compared to nativeborn workers, as well as to other immigrants from the same country of origin in different destination countries, and different origins in the same destination. The ECHP is a unique dataset produced by the European Union Statistical Office (Eurostat) that presents comparable micro-level (person/household) data on income, living conditions, demographic characteristics, migration, housing, health, and work for households across fifteen European Union Member States. The dataset also includes observations from the German socioeconomic household panel (SOEP), from the household panel from Luxembourg (PSELL) and from the British household panel (BHPS). Interviews in the European Community Household Panel (ECHP) were conducted simultaneously across all countries and data from national household panels were structured to mimic the rest of the ECHP. Thus, the European Community Household Panel is the first household survey that provides the data necessary for a comparative analysis of the adjustment of immigrants not only across

Divergent patterns in immigrant earnings

87

Table 4.1 Number of immigrants in sample by continent of origin for each destination country Country of destination

Asia

Non-Eng Eng Oceania Africa America America EU

Panel A –ECHP Data Denmark 320 4 Belgium 95 Luxembourg 25 France 390 UK 618 75 Ireland 59 18 Italy 34 34 Spain 23 2 Portugal 20 Austria 146 Sweden 609 10 Total 2,339 143

64 706 80 2,998 262 20 345 186 1,074 28 152 5,915

70 83 9 78 125 7 205 679 280 27 169 1,732

35 14 15 13 90 134 68 14 39 27 66 515

Non-EU Europe Total

405 264 1,983 401 1,711 103 2,530 621 489 150 1,884 25 728 581 715 131 622 33 693 1,917 1,611 1,492 13,371 5,718

1,162 3,282 1,943 6,630 1,809 2,147 1,995 1,750 2,068 2,838 4,109 29,733

Panel B – ECHP data deleting observations with missing values for relevant variables Denmark 101 0 24 45 15 104 74 363 Belgium 19 0 154 24 4 424 67 692 Luxembourg 14 0 31 4 6 498 18 571 France 83 0 576 32 10 612 129 1,442 UK 122 25 80 48 42 154 44 515 Ireland 5 4 1 7 40 560 11 628 Italy 10 0 62 50 23 193 149 487 Spain 6 0 35 140 14 122 31 348 Portugal 12 0 336 32 6 166 10 562 Austria 42 0 3 0 1 188 457 691 Sweden 148 3 28 65 17 605 405 1,271 Total 562 32 1,330 447 178 3,626 1,395 7,570 Note: Information on continents of origin is not available for Germany, The Netherlands, Greece or Finland. Source: ECHP – Waves 1–7.

broad geographic areas of origin, but unlike other data sources on immigrants, also across the major European destination countries. The data in the top panel of Table 4.1 reports the birthplace of immigrants for each of the eleven European countries for which the region of origin of the immigrants is known. The sample proportions should not be interpreted as population proportions due to the problem of sampling fractions that vary across the destinations. The bottom panel in Table 4.1 reports the sample sizes when observations with missing values for relevant variables are deleted from the sample. Appendix Table 4.A.1 reports these data as a relative frequency distribution for each

88

Immigration and the Transformation of Europe

geographic area of origin. The number of immigrants from Oceania is fairly small and they are mostly concentrated in the United Kingdom, Ireland, and Italy. Asians are almost absent from Southern Europe. Immigrants from Africa tend to choose France as their destination. The majority of them are natives of former French colonies, particularly in North Africa. Among those from the western hemisphere, English speakers move predominantly either to the United Kingdom or to Ireland, whereas those from South and Central America (primarily Romance language countries) choose mainly Spain, Portugal, and Italy. The ECHP sample contains around one million observations for individuals from the fifteen European Union countries, but only just over half of them work and report income from work. The top panel of Table 4.A.2 in the Appendix presents the distribution of observations, individuals of foreign birth and the percentage born in other European Union (EU) nations across destination countries in the total sample. After all explanatory variables are included, thereby deleting cases with missing values, the sample consists of 547,639 observations on individuals aged 18 and older – 231,457 women and 316,182 men.2 The bottom panel presents the distribution of individuals in the sample across countries. Foreigners represent 5.1 percent of the sample used in this study.3 Among foreigners, around 46 percent of those in the sample are citizens of other EU countries. The proportion of EU-born among migrants in the final sample can be compared to that of the general sample of migrants – working or not working. The proportions are quite stable for most countries, with three exceptions. Due to missing observations for some variables and to differences in labor force participation for immigrants from different origins, the proportion of EU-born in the sample for The Netherlands increases from 62 percent of the foreign-born in the sample to 95 percent. For the United Kingdom and Portugal the proportions of EU-born in the sample decrease from 52 to 28 percent and from 39 to 29 percent, respectively, largely due to missing many inactive EU-born migrants. The variables The natural logarithm of individual annual earnings from wages, salary and self-employment in purchasing power parity (PPP) terms is analyzed 2 3

Even if some individuals appear repeatedly in different years, observations are taken to be independent. The last column in the bottom panel of Table 4.A.2 presents the percent of foreign-born if the years since migration variable, a variable with a higher rate of missing values, was not included in the analysis. Note that the percentage of foreign-born increases to 6.64 percent of the sample and Luxembourg, the United Kingdom, and The Netherlands experience the largest changes in sample size.

Divergent patterns in immigrant earnings

89

both in a pooled sample and in a country fixed-effects model. Earnings are measured as total net income from work. Income data for France and Finland are in gross terms instead of net terms, and this needs to be taken into account in interpreting results. Welfare provisions such as housing and day-care subsidies, guaranteed income, unemployment benefits and others vary greatly across Europe and tend to be very generous in Nordic countries. Consideration of the effects of these sources of income and their effects on net earnings are beyond the scope of this study. The explanatory variables included in the analyses are: r Marital status This is measured by including two variables: one for currently married (Marry=1) and another for cohabitation (Cohabit=1). As a result, single and not cohabiting is the benchmark category. To the extent that a division of labor in the household has a different effect by gender on past labor supply and work effort, compared to being single, marriage is expected to be associated with lower earnings for women and moderately higher earnings for men. r No children The number of children present in the household is expected to have a negative impact on women’s earnings and a positive one on men’s earnings. r Education The educational attainment categories are less than upper secondary (Less secondary=1), upper secondary (the benchmark category) and at least some tertiary education (Tertiary =1). Unfortunately, a continuous variable for education, such as years of schooling, is not available. r Experience (Yrs experience & Sq experience) The survey reports the year when the individual worked for the first time. However, two considerations are warranted: first, data on this variable are not available for Sweden, and, second, on many occasions some of the reported answers are not consistent with responses to other questions in the survey. To create a more systematic, and perhaps less error-prone measure of experience, information on completed levels of education is used as follows: Potential experience is constructed as the age of the individual minus fourteen, eighteen, or twenty-three years, depending on the highest level of schooling (Age minus years of schooling minus six years). This measure of experience and its square are used in this study. r Foreign birth (Foreign=1) A variable is included to denote that an individual was foreign born. The ECHP includes several pieces of information on the migration history of each person surveyed. Since some questions were not included in the data for some countries, different data items are combined to construct this variable. This includes information on whether or not the person was foreign born (not readily available for Germany, part of Luxembourg and Sweden); whether

90

Immigration and the Transformation of Europe

or not the person was born in the EU (not available for Greece, The Netherlands, and the ECHP sample of Germany); and on their citizenship. r Years since migration (Yrs since migr & sq YSM) This variable is constructed from the year of arrival in the country of present residence. As indicated above, many observations were missing for these data for some countries. The square of years since migration is also included to reflect the non-linear effect on income of duration in the destination. r Geographic area of origin This variable distinguishes between those born within the EU or outside it. For Germany, The Netherlands, Greece, Finland, and for Luxembourg, in the PSELL sample, this is the only information available on the foreign country of birth. Information on the continent of origin is also available for all the other countries – Europe, Africa, Asia, America, and Oceania. As a result, when all the continents are included as explanatory variables, the coefficient for those born outside the EU corresponds to individuals from “other European origins,” mostly eastern Europe and the former Soviet Union. Where data on continents is available, except for those living in Italy and Austria, Americans can be divided into North and South/Central America. A variable for non-English speaking Americans is created which includes all persons classified as SouthCentral American, as well as those whose mother tongue is not English. Thus, Mexicans are in the non-English American group. Individuals for whom information is not available either on language or on continent are excluded from the sample when using continents as variables in the analysis. r Language spoken Individuals were asked: “What is your mother tongue?” and given a long list of choices. This information is not available for the United Kingdom, Sweden, and The Netherlands and is only available for the other countries in the year 2000 (seventh wave). Using the identification code of the individual, a mother tongue variable is created for other waves when the individual is present both in the seventh wave and previously. Two variables with language information were created. Same Language=1 if the language of the migrant and that of the country of destination match. Group Language=1 when the linguistic group of the language in the destination country and the migrant’s match, that is, they are close languages. The language groups are Romance (French, Portuguese, Spanish, Italian), English, Nordic (Danish and Swedish) and German/Dutch. Greek and Finnish are considered two separate language groups. In Luxembourg and Belgium both the Romance and Germanic language groups are accepted. Individuals for whom language information is not

Divergent patterns in immigrant earnings

91

available are not included in the analysis when language variables are used. All results are robust to the inclusion of survey year fixed-effects, that is, dichotomous variables for the year when the interview took place. Models are estimated with and without country of destination dichotomous variables. Further, foreign origin and birth outside of the EU are also interacted with country of destination to determine whether the effects on earnings of foreign birth in and outside the EU differ across the fifteen destination countries. Tables 4.2 and 4.3 present the means and standard deviations for the most relevant variables used in the analysis separately by country of destination and gender. The mean of the natural logarithm of the net earnings from work (in PPP terms) among men ranges from a high of 10.08 in Luxembourg to a low of 8.75 in Portugal. Among women, it ranges from 9.5 in Luxembourg to 8.44 in Portugal. Among migrants, the lowest net earnings are found in Sweden, with 8.61 for men and 8.31 for women. The relative level of education across countries and of migrants relative to natives varies widely. Tables 4.2 and 4.3 include the proportion of the sample with less than secondary studies and with tertiary (at least some college) education. Those with secondary schooling are the benchmark group. Educational attainment has a wider variance among migrants than among natives. In Denmark, for example, 41% of the immigrant men have some college education compared to 30% of all Danish men. The proportions with less than secondary schooling stand at 26% and 23% among immigrants and all Danes, respectively. Germany has a less educated pool of immigrants with only 6% that are college educated foreign-born men and 58% with less than secondary schooling. By contrast, the foreignborn living in Ireland are highly educated relative to the Irish-born in the sample: 25% of them have some college education as opposed to 17% of all men, and only 25% have less than secondary schooling compared to 43% of all Irish men. The education of both all women and foreign-born women is on average higher than that of men’s. This is not at all surprising since the sample is restricted to women who are working and on average they tend to be more educated. Years of potential labor market experience are relatively similar across the total sample of men in the fifteen European destinations. Yet the mean years of experience of the foreign-born living in southern Europe is substantially lower. This indicates that on average, migrants are younger in these destinations. The pattern is similar for foreign-born women. The mean years of experience for female immigrant workers range from a high of 22.8 in Sweden to a low of 15.9 in Luxembourg and 16.1 in Italy.

Italy

Ireland

United Kingdom

France

Luxembourg

Belgium

Netherlands

Denmark

Germany

Destination

9.47 (0.89) 9.28 (1.03) 9.51 (0.97) 9.44 (1.16) 10.08 (0.77) 9.35 (1.10) 9.34 (1.07) 9.26 (1.11) 9.25 (1.01)

Log work income

0.19 (0.39) 0.23 (0.42) 0.51 (0.50) 0.24 (0.43) 0.34 (0.47) 0.36 (0.48) 0.36 (0.48) 0.43 (0.49) 0.51 (0.50)

Less secondary educ

All

Table 4.2 Means of variables for men

0.26 (0.44) 0.30 (0.46) 0.15 (0.35) 0.35 (0.48) 0.20 (0.40) 0.23 (0.42) 0.43 (0.50) 0.17 (0.38) 0.10 (0.30)

Tertiary educ 21.27 (12.44) 22.43 (14.24) 19.70 (12.24) 19.77 (12.27) 18.75 (13.17) 20.11 (12.85) 21.24 (14.30) 22.49 (15.87) 20.74 (12.97)

Experience 9.50 (0.65) 9.17 (1.01) 9.58 (0.85) 9.59 (1.00) 9.85 (0.82) 9.37 (0.96) 9.36 (1.27) 9.30 (1.24) 9.24 (0.88)

Log work income 0.58 (0.49) 0.26 (0.44) 0.76 (0.43) 0.25 (0.43) 0.56 (0.50) 0.47 (0.50) 0.23 (0.42) 0.25 (0.43) 0.40 (0.49)

Less secondary educ.

0.06 (0.24) 0.41 (0.49) 0.13 (0.34) 0.34 (0.47) 0.21 (0.41) 0.19 (0.39) 0.41 (0.49) 0.27 (0.44) 0.15 (0.36)

Tertiary educ

Foreign-Born

23.09 (12.61) 19.34 (11.65) 18.21 (14.25) 20.35 (10.95) 21.03 (10.84) 25.76 (11.53) 22.09 (12.82) 20.01 (13.59) 16.16 (11.50)

Experience

21.86 (8.38) 19.36 (13.02) 24.10 (15.16) 27.45 (13.04) 15.54 (11.07) 27.51 (11.57) 24.20 (13.86) 23.75 (12.82) 25.02 (12.39)

Yrs since migr

9.09 (0.86) 9.13 (1.07) 8.75 (1.00) 9.43 (0.91) 9.11 (1.53) 8.87 (1.42)

0.50 (0.50) 0.59 (0.49) 0.84 (0.37) 0.19 (0.39) 0.29 (0.45) 0.23 (0.42)

Note: Standard deviations below the mean. Source: ECHP – Waves 1–7.

Sweden

Finland

Austria

Portugal

Spain

Greece

0.20 (0.40) 0.22 (0.42) 0.06 (0.23) 0.07 (0.26) 0.26 (0.44) 0.28 (0.45)

23.32 (13.98) 22.19 (14.05) 24.87 (15.71) 21.39 (12.70) 22.29 (14.83) 24.32 (14.68)

9.09 (0.82) 8.94 (1.14) 8.93 (0.98) 9.49 (0.75) 8.95 (1.66) 8.61 (1.55)

0.30 (0.46) 0.34 (0.48) 0.54 (0.50) 0.23 (0.42) 0.24 (0.43) 0.22 (0.41)

0.31 (0.46) 0.35 (0.48) 0.17 (0.37) 0.17 (0.38) 0.39 (0.49) 0.31 (0.46)

18.41 (13.14) 16.58 (10.51) 15.88 (12.99) 21.87 (11.91) 23.15 (15.47) 24.41 (13.51)

19.12 (12.83) 18.95 (10.40) 19.60 (9.56) 19.86 (14.77) 17.11 (13.20) 23.25 (16.20)

Italy

Ireland

United Kingdom

France

Luxembourg

Belgium

Netherlands

Denmark

Germany

Destination

8.85 (1.00) 8.95 (1.00) 8.70 (1.13) 8.96 (1.21) 9.49 (1.00) 8.90 (1.21) 8.71 (1.13) 8.63 (1.22) 8.93 (1.03)

Log work income

0.23 (0.42) 0.22 (0.41) 0.53 (0.50) 0.19 (0.39) 0.42 (0.49) 0.34 (0.48) 0.41 (0.49) 0.27 (0.44) 0.38 (0.49)

Less secondary educ

All

Table 4.3 Means of variables for women

0.19 (0.39) 0.33 (0.47) 0.14 (0.35) 0.43 (0.50) 0.16 (0.36) 0.28 (0.45) 0.38 (0.48) 0.22 (0.42) 0.12 (0.33)

Tertiary educ 20.18 (12.15) 20.36 (13.33) 16.40 (12.18) 16.51 (11.21) 15.91 (12.81) 18.49 (13.03) 20.70 (13.84) 16.82 (13.53) 16.11 (11.57)

Experience 8.89 (0.86) 8.89 (0.99) 8.63 (1.00) 8.94 (1.31) 9.27 (1.12) 8.95 (1.10) 8.86 (1.16) 8.65 (1.19) 8.78 (1.07)

Log work income 0.66 (0.47) 0.29 (0.46) 0.73 (0.45) 0.21 (0.41) 0.50 (0.50) 0.46 (0.50) 0.23 (0.42) 0.22 (0.42) 0.34 (0.48)

Less secondary educ

0.04 (0.20) 0.37 (0.48) 0.20 (0.41) 0.38 (0.49) 0.19 (0.39) 0.25 (0.43) 0.44 (0.50) 0.31 (0.46) 0.11 (0.32)

Tertiary educ

Foreign-Born

20.58 (11.84) 17.93 (11.88) 12.75 (13.13) 16.81 (11.28) 17.52 (10.81) 22.33 (11.56) 20.54 (12.39) 16.39 (11.63) 13.15 (10.15)

Experience

21.92 (7.85) 22.21 (13.31) 18.36 (14.45) 25.12 (12.66) 14.77 (10.70) 25.67 (11.78) 22.76 (12.87) 23.30 (11.47) 22.99 (13.19)

Yrs since migr

8.66 (0.98) 8.66 (1.15) 8.44 (1.08) 8.89 (1.03) 8.76 (1.50) 8.54 (1.34)

0.40 (0.49) 0.47 (0.50) 0.72 (0.45) 0.27 (0.44) 0.25 (0.43) 0.19 (0.40)

Note: Standard deviations below the mean. Source: ECHP – Waves 1–7.

Sweden

Finland

Austria

Portugal

Spain

Greece

0.29 (0.45) 0.32 (0.47) 0.11 (0.31) 0.09 (0.29) 0.36 (0.48) 0.32 (0.47)

16.22 (12.99) 16.75 (12.79) 19.34 (14.31) 19.78 (12.39) 21.11 (13.85) 22.79 (13.90)

8.55 (0.90) 8.71 (1.13) 8.74 (0.97) 8.98 (0.93) 8.61 (1.67) 8.31 (1.58)

0.24 (0.43) 0.22 (0.41) 0.40 (0.49) 0.31 (0.46) 0.22 (0.41) 0.24 (0.43)

0.38 (0.48) 0.44 (0.50) 0.29 (0.45) 0.19 (0.39) 0.37 (0.48) 0.32 (0.47)

13.40 (10.80) 15.56 (10.82) 14.40 (11.96) 21.64 (12.70) 22.73 (13.45) 24.31 (13.55)

16.86 (10.48) 19.25 (11.52) 19.66 (7.56) 18.94 (14.79) 15.75 (12.37) 23.84 (14.94)

96

Immigration and the Transformation of Europe

The mean years since migration for foreign-born individuals is reported in the last column in both tables. Not surprisingly Luxembourg has the lowest mean. Many of the foreign-born are temporarily assigned to positions in the institutions of the EU bureaucracy in Luxembourg. On the other hand, immigrants in Spain, Greece, and Portugal, countries that were traditionally a source rather than a destination for migrants, have relatively lower average years of migration. Overall mean duration is longer for men than for women, as many men migrate first and bring the family after they are established. Empirical analysis The effect of foreign origin across gender When the earnings equations are computed without country fixed effects, earnings of foreign born women and men at the time of arrival are estimated to be around 38 and 42 percent, respectively, lower than those of their native-born counterparts. When foreigners are split among those born in the EU and those born outside it, the EU-born women and men experience only 33% lower earnings at arrival with respect to natives, whereas women and men born outside the EU have around 41% and 56% lower earnings than natives, respectively. Differences in earnings of immigrants relative to natives of the same gender in each country vary on average from a low of 8 percent for women and 19 percent for men living in Germany, to a high of 62 percent and 67 percent for foreign women and men living in Sweden. Table 4.4 presents predicted differences in earnings for the foreignborn and for those born outside the EU as a percentage of the earnings of natives of the same gender in each of the fifteen destination countries, for individuals with a given set of characteristics. These characteristics are ten years of experience, a high school diploma, married and with one child, and, for the foreign-born, evaluated at the time of arrival. Table 4.4 columns (1) and (4), for men and for women, respectively, report the predicted percent age difference in earnings between immigrants and natives. Columns (2) and (5) are the predicted percent differences in earnings for immigrants born outside the EU compared to the native born. The third and sixth columns include the level of significance at which earnings for EU and non-EU migrants differ in each destination country. Here is how to read Table 4.4. The coefficient −0.204 for foreign men in Belgium (Table 4.4, column 1) means that men born outside Belgium on arrival earn about 20 percent less than men born in Belgium, other

Divergent patterns in immigrant earnings

97

Table 4.4 Predicted percent differences in earnings of immigrants at arrival and natives, with the same characteristics, by gender Men

Women

Destination

Foreign Diff Foreign Diff Foreign all non-EU EU/Non-EU Foreign all non-EU EU/non-EU

Germany Denmark Netherlands Belgium Luxembourg France United Kingdom Ireland Italy Greece Spain Portugal Austria Finland Sweden

−0.153 a −0.479 a −0.201 c −0.204 a −0.586 a −0.418 a −0.165 b −0.528 a −0.333 a −0.311 a −0.503 a −0.144 b −0.307 a −0.613 a −0.398 a

−0.181 −0.592 −0.488 −0.459 −0.675 −0.506 −0.539 −0.326 −0.377 −0.314 −0.658 −0.192 −0.409 −0.462 −0.823

− − − a − b a b − − c − − − a

−0.017 −0.362 a −0.342 a −0.361 a −0.537 a −0.337 a −0.221 b −0.415 a −0.517 a −0.484 a −0.311 a −0.278 a −0.441 a −0.619 a −0.500 a

−0.112 −0.461 −0.037 −0.412 −0.693 −0.346 −0.383 −1.141 −0.492 −0.499 −0.569 −0.142 −0.285 −0.462 −0.701

b − − − − − c a − − b − c − a

Note: Predictions are from Adser`a and Chiswick (forthcoming). Earnings are calculated for an individual with ten years of experience, high school diploma, married and with one child. For foreigners, earnings are measured at the time of arrival (Duration equal to zero years.) For significance levels, foreigners are compared to natives and non-EU born to all foreigners. Significance margins: (a) 1%; (b) 5%; and (c) 10%. Source: ECHP – Waves 1–7.

variables being the same.4 The coefficient −0.459 for men in Belgium born outside the EU means that, on arrival, they earn about 46 percent less than men born in Belgium, all things being equal. The entry in the third column indicates that there are significant differences between the earnings of foreign men in Belgium who were born within the EU and those who were born outside it. Individuals born in the EU but living in another European country have significantly lower earnings on arrival than natives in all cases, except for women in Germany. Among foreign-born men in the EU, differences relative to native men are only on the order of 15 percent for those living in Germany, the United Kingdom and Portugal, but the differences are much larger (up to 50% to 60%) for those living in Luxembourg, Ireland, Spain, and Finland. For EU-born women, except for those in 4

Actually, the percent difference in earnings (b) equals exp(b∗ )-1 where b∗ is the regression coefficient. The percent difference b is approximately equal to b∗ only when b is small.

98

Immigration and the Transformation of Europe

Germany whose earnings are comparable to German women, differences with respect to native women range from 22% (United Kingdom) to 62% (Finland) but are more homogenous than those for men. The United Kingdom is the “best” destination after Germany for that group of women. Countries where the relative earnings on arrival of the EUborn are the lowest (over 50 percent lower) with respect to their native counterparts are Finland, Ireland, Luxembourg, and Spain for men and Finland, Luxembourg, Sweden, and Italy for women. The earnings of migrants born outside the EU are significantly lower at migration than the EU-born in six of the fifteen countries presented in Table 4.4. Sweden is the country where the differences are the greatest and most highly statistically and economically significant overall. A man born outside the EU earns around 82% less at migration than a Swedish man and a non-EU-born woman around 70% less than a Swedish woman. The earnings differences evaluated at migration of these individuals with respect to migrants from the EU living in Sweden amount to 42% and 20% for men and women, respectively. The United Kingdom and Spain are the other countries where differences between the two groups of foreign workers (EU and non-EU) are significant and relatively sizable for both men and women. In Spain, those differences amount to 26% for women and 15% for men. In the United Kingdom, the numbers are 16% and 37%. In Belgium, men from outside the EU earn 25% less than their EU counterparts – a group that includes many EU bureaucrats – and those in France 9 percent less. Interestingly, for non-EU men in Ireland and non-EU women in Austria, predicted earnings on arrival are higher than those of migrants from EU countries. In Ireland, most of these immigrants are from the United States and Canada, and in Austria, they arrive mainly from the former Soviet Union. Around 77 percent of migrants to Austria are from outside the EU. Austria hosts a quarter of all non-EU European migrants in the sample. To sum up, differences in earnings by nativity vary greatly across destination countries, with migrants in Germany and Portugal receiving the highest earnings relative to natives and those in Sweden, Denmark, Luxembourg, and Spain the lowest – particularly among those not born in the EU. Table 4.5 reports the data in a different format. It presents predicted differences in the earnings of foreigners on arrival in different countries as compared to those arriving in Germany. This can be done because earnings are in purchasing power parity (PPP) terms. Germany is the benchmark and the values for other countries in the table represent percentage deviations from German earnings. For example, the coefficient −0.097

Divergent patterns in immigrant earnings

99

Table 4.5 Predicted percent differences in earnings of immigrants at arrival in different destination countries as compared to those arriving in Germany, by gender Men

Women

Destination

Native

Foreign

Non-EU

Native

Foreign

Non-EU

Germany Denmark Netherlands Belgium Luxembourg France United Kingdom Ireland Italy Greece Spain Portugal Austria Finland Sweden

0 −0.097 0.135 −0.039 0.756 −0.030 −0.017 0.113 −0.077 −0.228 −0.148 −0.384 0.139 −0.207 −0.372

0 −0.423 0.087 −0.090 0.323 −0.295 −0.028 −0.263 −0.257 −0.387 −0.498 −0.376 −0.015 −0.667 −0.617

0 −0.508 −0.173 −0.316 0.262 −0.355 −0.375 −0.032 −0.273 −0.361 −0.624 −0.395 −0.089 −0.489 −1.014

0 0.073 −0.025 0.066 0.828 0.090 −0.072 0.041 0.212 −0.035 −0.026 −0.102 0.202 −0.066 −0.289

0 −0.272 −0.350 −0.278 0.309 −0.230 −0.276 −0.357 −0.288 −0.501 −0.320 −0.364 −0.222 −0.668 −0.771

0 −0.276 0.051 −0.233 0.247 −0.144 −0.343 −0.987 −0.168 −0.422 −0.483 −0.133 0.029 −0.416 −0.878

Note: Predictions are from Adser`a and Chiswick (forthcoming). Earnings are calculated for an individual with ten years of experience, high school diploma, married and with one child. For foreigners, earnings are measured at the time of arrival. Source: ECHP – Waves 1–7.

for Danish native men means that men born in Denmark earn about 9.7 percent less than men born in Germany, other variables being the same, while male immigrants in Denmark at arrival earn about 42.3 percent less than male immigrants in Germany, other variables being the same. Among native men and women, earnings are the highest in Luxembourg. The large proportion of people living in Luxemburg working for the EU institutions who enjoy very high salaries may explain the Luxembourg pattern. The same reasoning also applies when the earnings of foreigners are compared across countries, particularly among EU-born migrants. Foreigners in Luxembourg have the highest earnings of all. Nonetheless, a close look at the composition of EU migrants in Luxembourg shows that there are both conventional economic migrants (mostly from Portugal) and those who work for the EU administration. Differences from the other countries narrow when only the non-EU group is examined. In any case, only around 10 percent of migrants in Luxembourg come from outside the EU and these are mostly from eastern Europe.

100

Immigration and the Transformation of Europe

Setting aside Luxembourg, among EU-born foreign men, those living in The Netherlands, the United Kingdom, Austria, and Germany receive the highest earnings. Among foreign EU women, Germany is by far the best destination, followed by Austria, Ireland, Denmark, Italy, and the United Kingdom – countries where foreign EU women typically earn around a quarter less than their counterparts in Germany. Sweden and Finland are the countries where the foreign EU-born have over two-thirds lower net earnings compared to those living in Germany. Earnings among foreign men born outside the EU are, again, much lower in Nordic countries (Denmark, Sweden, and Finland) and Spain in comparison to those in Germany where they fare the best. Earnings in Ireland and Austria closely follow those in Germany. Earnings for foreign women born outside the EU show smaller differences across countries than those of men. Austria, the Netherlands and Germany are the best destinations, followed closely by France, Italy and Portugal. The lowest earnings for the group are found in Sweden and Ireland. Gender differences in demographic variables This section explores whether there are some underlying gender differences in the factors that explain individual earnings from work. Table 4.6, columns (1) and (2), include only coefficients of the demographic control variables for the general specifications employed in the paper. Columns (3) and (4) include the same set of variables as well as a language variable. This dichotomous variable is unity if the language group of the origin is the same as that of the destination. Complete estimates for all specifications are available in Adser`a and Chiswick (forthcoming). Interesting gender differences arise from the analysis of Table 4.6. As expected, earnings increase with the level of education. Returns to education are higher for women than for men. Estimates in Table 4.6 columns (1) and (2) imply the earnings of women with tertiary education are 93 percent higher than those with less than upper secondary education. Differences for the same educational groups among men are only in the order of 66 percent. Interestingly, when the language group variable is included in estimates in columns (5) and (6), earnings differences between the lowest and highest education categories widen to 103 percent and 71 percent, respectively. The increase in the effect of schooling is due to the deletion of data from The Netherlands, the United Kingdom, and Sweden, countries for which the origin language was not available, rather than the statistical control for the language variable. Years of experience in the work place or years of potential experience, as estimated here, have a similar effect across gender. Women double their

Divergent patterns in immigrant earnings

101

Table 4.6 Selected partial effect of variables from a regression analysis of earnings for the native and foreign-born by gender(a) Explanatory variables Less secondary Tertiary educ Yrs experience Experience Sq Foreign Years since migr YSM Sq No children Marry Cohabit.

Women

Men

Women

Men

Women

Men

−0.331 (−60.5) 0.594 (103.4) 0.107 (162.7) −0.002 (−150.5) −0.454 (−15.38) 0.031 (12.84) −0.00038 (−8.45) −0.144 (−61.38) 0.075 (13.75) 0.240 (30.83)

−0.230 (−55.8) 0.425 (92.56) 0.101 (200.8) −0.002 (−201.7) −0.465 (−20.12) 0.032 (17.57) −0.00045 (−13.93) 0.008 (4.40) 0.321 (64.49) 0.288 (43.36)

−0.336 (61.26) 0.594 (103.39) 0.107 (163.14) −0.002 (−150.96)

−0.233 (−56.4) 0.427 (93.1) 0.101 (201.3) −0.002 (−202.1)

−0.408 (−65.4) 0.626 (95.5) 0.103 (135.7) −0.002 (−123.6)

−0.265 (−58.1) 0.449 (87.2) 0.096 (167.2) −0.002 (−167.9)

0.021 (8.43) −0.00021 (−4.47) −0.144 (−61.54) 0.075 (13.73) 0.241 (30.90)

0.024 (12.44) −0.00031 (−9.2) 0.008 (4.54) 0.319 (64.22) 0.287 (43.24)

316,182 0.28

231,457 0.20

316,182 0.28

0.016 (5.7) −0.00013 (−2.38) −0.115 (−42.9) 0.065 (10.4) 0.214 (22.5) 0.112 (4.04) 173,307 0.21

0.021 (10.1) −0.00028 (−7.1) 0.005 (2.8) 0.320 (57.1) 0.277 (34.7) 0.146 (6.4) 251,836 0.28

Group language∗ foreign No Obs 231,457 Adj R-Sq. 0.20

Note: Dependent variable: natural logarithm of earnings. T-ratios are below coefficients. Complete estimates shown in Adser`a and Chiswick (forthcoming). (a) The regression equations also include destination country dichotomous variables alone in all columns and, in columns (3)–(6), interacted both with foreign origin and Non-EU origin. Language information is not available for The Netherlands, the UK, and Sweden, therefore these countries are not included in columns (3) and (4). Source: ECHP – Waves 1–7.

initial earnings after twelve years of experience and men after twelveand-a-half years (Table 4.6 all columns). Interestingly, if the number of children in the household is not included in the specification, the effect of pre-migration experience is somewhat stronger for men than for women. In most previous research on earnings, marital status is identified by a dichotomous variable: currently married versus not currently married.

102

Immigration and the Transformation of Europe

Typically, married men earn more than single men, while married women earn less than single women. Here, however, there is a tricotomous marital status variable converted into two dichotomous variables.5 Are those cohabiting more like married or more like single individuals? Individuals with no children living with a spouse, whether married or cohabitating, have higher earnings than single workers. Cohabiting men earn about 28 percent more than single men and cohabiting women earn about 24 percent more than single women. Married men earn 3 to 4 percent more than those who are cohabiting and around 32 percent more than single men. Controlling for the number of children in the household, married women earn around 7 percent more than single women but around 16 percent less than those in a cohabiting status. If the number of children is excluded from the specification, married women on average earn about 3 percent less than single women. The number of children in the household has a strong negative effect on women’s earnings: around 14 percent per child. Thus, while married women without children earn more than single women without children, married mothers with one child earn 7 percent less than singles without children, which increases to about 21 percent lower earnings if there are two children. For men, the coefficient on the number of children is significant and positive but negligible in size (around 1 percent per child). The years that have lapsed since the migrant came to the destination and the square of years since migration are included in the regression in Table 4.6. Given the estimated coefficients on years since arrival and foreign birth in columns (1) and (2), it takes migrant women around nineteen years and migrant men twenty years to earn what native workers earn on average, other variables being the same for both genders.6 At fifteen years in the destination, the earnings disadvantage is 8 percent and 9 percent, respectively, for women and men. Table 4.A.5 in the Appendix presents the proportion of migrants in each country whose mother tongue belongs to the same language group as the destination country. The proportions are very high for most countries, except for Germany, followed by Denmark. Around 74% of those migrating to Portugal, 45% of migrants in France, and 52% of those living in Spain speak a language in the Romance group. This arises from the propensity among migrants to move to a country where, all things being equal, the cost of adjustment is lower. A smaller “linguistic distance” 5 6

Modeling of the tricotomous choice is needed, but is beyond the scope of this study. This is quite similar to the United States where the earnings catch-up for economic migrants is about fifteen years, other measured variables being the same (Chiswick 1979, 1986).

Divergent patterns in immigrant earnings

103

between the origin and destination languages reduces this cost (Chiswick and Miller 1994).7 Regression estimates of earnings in Table 4.6 columns (3) and (4) include the variable group language. This covariate controls for whether the language of the migrant belongs to the same language group as the country of destination. The sample does not include individuals living in The Netherlands, the United Kingdom, and Sweden. The coefficient on group language is negative in regressions estimated by pooling all countries together without country-fixed effects. This is explained by the fact that the prevalence of group language matches among immigrants is higher in countries with lower average earnings (southern Europe). Once country of destination dichotomous (fixedeffects) variables are included alone and interacted with foreign origin, however, a group language match provides for a statistically significant 11.2% increase in earnings for migrant women and a 14.5% increase for migrant men, compared to coming from a different linguistic group. Results are very similar if a variable for an exact language match is included instead: a gain in earnings of 11.7% for women and 12.5% for men. The unexpected high proportion of exact language matches in countries such as Greece suggests a high proportion of returning families among immigrants. However, when similar regressions are computed by only considering the subset of immigrants who are not citizens, as opposed to all immigrants, the positive effects are extremely robust for men, but a common language does not seem to make a difference for non-citizen women’s earnings. Earning differences across continents of origin The ECHP data on country of birth for broad geographic areas provides an opportunity to analyze whether the region of origin is relevant for explaining earnings differences across immigrants within a destination. As noted, information on region of birth for immigrants is not available for Germany, the Netherlands, Greece, and Finland. Thus, these countries are not included in this analysis. The effects of country of origin on the earnings of immigrants relative to the native born in their destination are reported in Table 4.7, column (1). Overall, other variables being the same, on arrival women born in an EU country earn about 33% less than their native-born counterparts, while the earnings gap is even greater (about 50%) for EU-born 7

A similar phenomenon is found in Canada where immigrants to French-speaking Quebec come disproportionately from French-speaking countries of origin (Chiswick and Miller 1994).

104

Immigration and the Transformation of Europe

Table 4.7 Partial effect of region of birth on earnings compared to native-born in that country, by gender Women

Men

Country of Birth

Diff from natives

Diff from EU migrants

Diff from natives

Diff from EU migrants

EU Europe Non-EU English Speaking America Not Eng. Speak. America Asia Africa Oceania

−0.326 a −0.484 a −0.378 a −0.519 a −0.418 a −0.210 a −0.071

a − a c a –

−0.477 a −0.608 a −0.545 a −0.660 a −0.696 a −0.524 a −0.030

a – a a c a

Note: Partial effects of country of birth controlling for country of destination, schooling, experience, duration in destination and marital status. Significance margins: (a) 1%; (b) 5%; and (c) 10%. Information on continents is not available for Germany, The Netherlands, Greece and Finland; therefore these countries are not included in the sample. Source: ECHP – Waves 1–7.

immigrant men. The earnings gap is even greater (by about 12 to 15 percentage points) for those from non-EU countries. The earnings gap is about the same for the European non-EU-born immigrants as for those from the Americas and is smaller for immigrants from Africa and women from Asia. Migrants from Oceania, highly geographically concentrated in the United Kingdom, Ireland, and Italy, are moving across high-income areas. Their earnings are only slightly lower (15% for women and 3% for men) than those of a native worker. Table 4.7 columns (2) and (4) indicate whether the earnings gap differs significantly between migrants of each continent and migrants born in the EU. Conclusion This chapter uses the 1994–2000 waves of the ECHP to conduct a systematic analysis across European countries of destination of individual earnings of immigrants in comparison to the native born and migrants from other countries. There is a significant negative effect of immigrant status on individual earnings of around 40 percent at the time of arrival in the pooled sample, although this is somewhat smaller in magnitude for women than for men. Those differences, however, vary greatly across destination countries with migrants in Germany and Portugal faring the best relative to natives and those in Sweden, Denmark, Luxembourg, and Spain the worst – particularly among those not born in the EU.

Divergent patterns in immigrant earnings

105

In absolute terms, however, immigrants in Luxembourg, Germany, The Netherlands, and Austria have the highest earnings. Earnings for individuals born in the EU but living in another EU country are only around a third lower both for men and women on arrival. Among those born outside the EU, the results indicate that women are doing relatively better than men. By continent of origin, Asian, Latin American, and eastern European men have the lowest earnings. Latin American and eastern European women are at the bottom of the women’s distribution. Immigrants tend to gravitate to countries with a close linguistic and cultural background, and immigrants earn more if their origin language is the same as, or close to, that of the destination. Language is more relevant for earnings for men than for women. Again, this point fits well with the main thrust of the preceding chapter, which found strong “network” effects in immigrants’ choice of destination. On average, about nineteen years after migration, the earnings of immigrants reach equality or parity with the earnings of natives. This is about the same as that reported in findings for the United States. While addressing many matters, this chapter also raises new questions. To what extent are the differences in the immigrant/native earnings ratios by gender in the destinations due to differences in the selectivity of immigrants from the various origins to the various destinations? To what extent do differences in the transferability of skills across origins and destinations play a role? To what extent do institutional differences in the destinations in terms of flexibility in labor markets, public assistance to immigrants, and receptivity to immigrants, and discrimination play a central role? These are issues that warrant further research.  Adser`a, Alicia and Barry R. Chiswick forthcoming “Are There Gender and Country of Origin Differences in Immigrant Labor Market Outcomes across European Destinations?” Journal of Population Economics. Antecol, Heather, Deborah Cobb-Clark and Stephen Trejo 2002 “Human Capital and Earnings of Female Immigrants to Australia, Canada and the United States.” Unpublished manuscript. Department of Economics, Claremont McKenna College. Baker, Michael and Dwayne Benjamin 1997 “The Role of the Family in Immigrants’ Labor-Market Activity.” American Economic Review 87(4) (September): 705–27. Biswal, Urvashi D. 1999 “Testing the Family ‘Common Preference’ Model for Immigrant and Non-Immigrant Women’s Labour Supply.” Canadian Public Policy 25(s1) (November): 95–114. http://economics.ca/cgi/ jab?journal=cpp&view=v25s1/CPPv25s1p095.pdf

106

Immigration and the Transformation of Europe

Boyle, Paul, Thomas J. Cooke, Keith Halfacree and Darren Smith 2001 “A Cross-National Comparison of the Impact of Family Migration on Women’s Employment Status.” Demography 38(2) (May): 201–13. Chiswick, Barry R. 1978 “The Effect of Americanization on the Earnings of Foreign-Born Men.” Journal of Political Economy 86(5) (October): 897–921. Chiswick, Barry R. 1979 “The Economic Progress of Immigrants: Some Apparently Universal Patterns,” in Contemporary Economic Problems, 1979, William John Fellner (ed.). Washington, DC: American Enterprise Institute, 357–99. Chiswick, Barry R. 1980 “An Analysis of the Economic Progress and Impact of Immigrants.” PB80–200454. Prepared for the Employment and Training Administration, US Department of Labor. Springfield, VA: National Technical Information Service, August. Chiswick, Barry R. 1986 “Human Capital and the Labor Market Adjustment of Immigrants: Testing Alternative Hypotheses.” Research in Human Capital and Development 4: 1–26. Chiswick, Barry R. and Paul W. Miller 1994 “Language Choice among Immigrants in a Multi-lingual Destination.” Journal of Population Economics 7(2): 119–31. Cobb-Clark, Deborah A. and Marie D. Connolly 2001 “A Family Affair: The Labor Market Experiences of Immigrant Spouses.” Social Science Quarterly 82(4) (December): 796–811. Cobb-Clark, Deborah A. and T. F. Crossley 2006 “Gender, Comparative Advantage and Labour Market Activity in Immigrant Families.” Journal of Labor Economics. Duleep, Harriet Orcutt and Seth Sanders 1993 “The Decision to Work by Married Immigrant Women.” Industrial and Labor Relations Review 46(4) (July): 677–90. Evans, M. D. R. 1984 “Immigrant Women in Australia: Resources, Family, and Work.” International Migration Review 18(4) (Special Issue) (Winter): 1063–90. LeClere, Felicia B. and Diane K. McLaughlin. 1997 “Family Migration and Changes in Women’s Earnings: A Decomposition Analysis.” Population Research and Policy Review 16(4) (August): 315–35. Long, James E. 1980 “The Effect of Americanization on Earnings: Some Evidence for Women.” Journal of Political Economy 88(3) (June): 620–29. MacPherson, David A. and James B. Stewart 1989 “The Labor Force Participation and Earnings Profiles of Married Female Immigrants.” Quarterly Review of Economics and Business 29(3) (Autumn): 57–72. Reimers, Cordelia W. 1985 “Cultural Differences in Labor Force Participation among Married Women.” American Economic Review 75(2) (May): 251–55.

Divergent patterns in immigrant earnings

107

Appendix Table 4.A.1 Share of immigrants in sample by origin and destinations Country of destination

Asia

Panel A –ECHP Data Denmark 13.68 Belgium 4.06 Luxembourg 1.07 France 16.67 UK 26.42 Ireland 2.52 Italy 1.45 Spain 0.98 Portugal 0.86 Austria 6.24 Sweden 26.04 Total

100

Oceania Africa

2.80

52.45 12.59 23.78 1.40

6.99 100

1.08 11.94 1.35 50.68 4.43 0.34 5.83 3.14 18.16 0.47 2.57 100

Non-Eng Eng America America

4.04 4.79 0.52 4.50 7.22 0.40 11.84 39.20 16.17 1.56 9.76 100

6.80 2.72 2.91 2.52 17.48 26.02 13.20 2.72 7.57 5.24 12.82 100

EU

3.03 14.83 12.80 18.92 3.66 14.09 5.44 5.35 4.65 5.18 12.05 100

Non-EU Europe Total

4.62 7.01 1.80 10.86 2.62 0.44 10.16 2.29 0.58 33.53 26.09 100

3.91 11.04 6.53 22.30 6.08 7.22 6.71 5.89 6.96 9.54 13.82 100

Panel B – ECHP data deleting observations with missing values by relevant variables Denmark Belgium Luxembourg France UK Ireland Italy Spain Portugal Austria Sweden Total

17.97 3.38 2.49 14.77 21.71 0.89 1.78 1.07 2.14 7.47 26.33 100

0 0 0 0 78.12 12.5 0 0 0 0 9.38 100

1.8 11.58 2.33 43.31 6.02 0.08 4.66 2.63 25.26 0.23 2.11 100

10.07 5.37 0.89 7.16 10.74 1.57 11.19 31.32 7.16 0 14.54 100

8.43 6,226 2.25 11.69 3.37 13.73 5.62 16.88 23.6 4.25 22.47 15.44 12.92 5.32 7.87 3.36 3.37 4.58 0.56 5.18 9.55 16.69 100

100

5.30 4.80 1.29 9.25 3.15 0.79 10.68 2.22 0.72 32.76 29.03 100

4.80 9.14 7.54 19.05 6.80 8.30 6.43 4.60 7.42 9.13 16.79 100

Note: Information on continents of origin is not available for Germany, Netherlands, Greece or Finland. Source: ECHP – Waves 1–7.

108

Immigration and the Transformation of Europe

Table 4.A.2 Number of individuals and of foreign-born in the sample by destination Country of destination

No of Obs

Panel A –ECHP data Germany 110,204 Denmark 33,031 Netherlands 63,533 Belgium 40,123 Luxembourg 39,785 France 85,057 UK 87,820 Ireland 49,093 Italy 115,759 Greece 76,329 Spain 103,815 Portugal 80,522 Austria 40,315 Finland 36,345 Sweden 37,726 Total

No foreign % Foreign % Foreign % Total born % Foreign female EU-born

11.03 3.30 6.36 4.01 3.98 8.51 8.79 4.91 11.58 7.64 10.39 8.06 4.03 3.64 3.77

999,457 100.0

13,797 1,351 844 4,276 12,293 7,428 2,991 2,163 2,083 3,571 1,823 2,429 2,978 1,332 4,132

12.52 4.09 1.33 10.66 30.90 8.73 3.41 4.41 1.80 4.68 1.76 3.02 7.39 3.66 10.95

47.97 57.59 53.67 52.71 51.84 51.24 55.27 56.96 58.57 59.59 55.29 52.66 57.96 47.52 51.62

38.39 41.45 61.26 65.19 88.23 39.58 52.16 87.70 37.97 90.79 43.01 39.65 25.62 25.83 39.38

63,491

6.35

52.44

55.00

Panel B –Working individuals deleting observations with missing values Germany Denmark Netherlands Belgium Luxembourg France UK Ireland Italy Greece Spain Portugal Austria Finland Sweden Total

67,422 23,745 38,230 22,257 14,657 48,467 56,213 27,882 52,952 32,866 44,403 40,780 22,071 27,641 28,053

12.30 4.34 6.98 4.06 2.68 8.85 10.30 5.09 9.67 6.00 8.11 7.45 4.03 5.05 5.12

547,639 100.0

Source: ECHP – Waves 1–7.

6,571 701 131 1,629 1,334 3,532 1,060 1,309 1,067 1,372 782 1,218 1,482 899 2,558

9.75 2.95 0.34 7.32 9.10 7.29 1.89 4.69 2.02 4.17 1.76 2.99 6.71 3.25 9.12

38.1 54.9 52.0 38.0 41.4 39.0 47.8 42.2 42.7 35.8 47.6 46.2 44.9 45.8 46.7

39.26 33.95 94.66 56.11 89.21 39.64 28.11 87.78 35.71 86.95 39.00 29.47 23.55 19.80 44.49

12.57 3.42 1.25 9.51 37.45 7.99 3.20 4.73 2.08 4.43 1.85 3.42 7.00 3.49 9.17

25,645

4.68

43.2

46.0

6.64

Divergent patterns in immigrant earnings

109

Table 4.A.3 Predicted natural logarithm of earnings at arrival for women: Natives, foreigners, and non-EU foreigners Destination

Native

Foreign–all

Foreign–non-EU

Germany Denmark Netherlands Belgium Luxembourg France United Kingdom Ireland Italy Greece Spain Portugal Austria Finland Sweden

7.731 7.804 7.706 7.797 8.559 7.821 7.659 7.772 7.943 7.696 7.705 7.629 7.933 7.665 7.442

7.714 7.442 7.364 7.436 8.023 7.484 7.438 7.357 7.426 7.213 7.394 7.350 7.492 7.046 6.943

7.619 7.343 7.670 7.386 7.866 7.475 7.276 6.632 7.451 7.197 7.136 7.486 7.648 7.203 6.741

Note: Predictions from estimates in Adser`a and Chiswick (forthcoming) considering an individual with ten years of experience, high school diploma, married and with one child. For foreigners, earnings are measured at the time of arrival (Duration “equals” zero). Source: ECHP – Waves 1–7.

Table 4.A.4 Predicted natural logarithm of earnings on arrival for men: Natives, foreigners and non-EU foreigners Destination

Native

Foreign–all

Foreign–non-EU

Germany Denmark Netherlands Belgium Luxembourg France United Kingdom Ireland Italy Greece Spain Portugal Austria Finland Sweden

8.049 7.952 8.184 8.010 8.805 8.019 8.032 8.162 7.972 7.821 7.901 7.665 8.188 7.842 7.677

7.896 7.473 7.983 7.806 8.219 7.601 7.868 7.633 7.639 7.509 7.398 7.520 7.881 7.229 7.279

7.868 7.360 7.695 7.552 8.130 7.513 7.493 7.836 7.595 7.507 7.244 7.473 7.779 7.379 6.854

Note: Predictions from estimates in Adser`a and Chiswick (forthcoming) considering an individual with ten years of experience, high school diploma, married and with one child. For foreigners, earnings are measured at the time of arrival (Duration equals zero). Source: ECHP – Waves 1–7.

110

Immigration and the Transformation of Europe

Table 4.A.5 Proportion of immigrants with group language match Destination

Group Language

Germany Denmark Netherlands Belgium Luxembourg France United Kingdom Ireland Italy Greece Spain Portugal Austria Finland Sweden

13.0 27.7 N/A 52.9 48.4 44.7 N/A 44.6 57.1 58.2 52.4 73.9 36.0 38.8 N/A

Total

40.1

Note: Group Language=1 when the linguistic group of the language from the country and the migrant’s match. See text for details. Source: ECHP – Waves 1–7.

5

Economic consequences of immigration in Europe Herbert Br¨ucker, Joachim R. Frick and Gert G. Wagner

Introduction The migration policies of the European Union (EU) and the other Member States of the European Economic Area (EEA) are characterized by a two-fold approach. On the one hand, the free movement of labor has been defined since the Treaty of Rome (which established the EU in the 1950s) as one of the fundamental freedoms of the Common Market and has been subsequently implemented by the supranational legislation of the European Community. This integrative approach distinguishes the EU and the EEA from other regional trade agreements in the world such as NAFTA. On the other hand, the individual Member States of the EU and the EEA decide on immigration policies vis-`a-vis third-country nationals. Most Member States have pursued a restrictive migration policy since the first oil-price shock in 1973. Aggregate migration figures reflect this restrictive approach: although the income gap on the European continent and between Europe and its neighboring regions resembles that between North and South America (Brucker ¨ 2002), annual net immigration rates in the EU and the EEA have only been half those of the United States and Canada (2.2 persons per thousand versus 4.4 per thousand) during the 1990s and early 2000s.1 The restrictive immigration policies of the EU and the EEA vis-`a-vis third countries face three main challenges today. First, the fall of the Berlin Wall has removed the barriers to emigration in an area with almost 400 million people and a per capita Gross Domestic Product (GDP) of only 20 to 50 percent of those in the EU and the EEA. Second, more and more neighboring countries with relatively low per capita incomes aspire to become members of the EU. As a consequence, the EU and the EEA is less and less a club of rich countries with homogeneous income levels. Measured at purchasing power parities, the income levels of the new Member States (NMS) that joined the EU in May 2004 1

Authors’ calculations based on data from the UN Population Division (2004).

111

112

Immigration and the Transformation of Europe

amount on average to only 50 percent of those of the old Member States. The income levels of other (potential) accession candidates (Bulgaria, Romania, Turkey) is even lower, at around 20 percent of the old European Union-15. Third, the populations of the EU and the EEA will age rapidly in the coming years. As Paul Demeny elaborated in his chapter, the age dependency ratio – the ratio of the population above 65 to the population in working age – will double during the next 40 years. As a consequence of these developments, international migration both within the enlarged EU and between the EU and its neighboring regions may increase in the future. The objective of this chapter is to analyze the implications of increasing migration in Europe on welfare and its welfare states from an economic perspective. In theoretical terms two main issues, including several sub-issues, are addressed. The next part of the chapter discusses whether international labor migration is a zero-sum game or whether it yields a net welfare gain for the economies involved. We analyze the impact of labor migration on income, wages, unemployment and the distribution of income in host and source regions, using theoretical and simulation models. Then we compare the results from the simulations with the empirical evidence from econometric studies. The third part of the chapter discusses some stylized facts on the status of immigrants in Europe and attempts to answer the question of how the economic gains and losses incurred from immigration by subgroups of natives affects the overall acceptance of migration in immigration societies. We also provide some evidence on the incidence and level of remittances to immigrants’ home countries. The fourth part of the chapter discusses the fiscal implications of migration in the context of the rapidly ageing populations in European countries. A generational accounting approach is used to analyze whether international migration can help to alleviate the sustainability gap in public finances. The findings of the study are then summarized and conclusions are drawn for European migration policies. The impact of migration on wages, employment, and income distribution The public debate on the economic benefits and costs of migration is characterized by a distinction between its short- and long-term effects. In the short term, fears are widespread in the receiving countries that migration is harmful, since it ostensibly reduces wages and increases unemployment of native workers. In the long term, many observers expect that migration is beneficial because it may mitigate the costs of population ageing. The short-term fears have been very important in discouraging

Economic consequences of immigration in Europe

113

European countries from opening their labor markets for the immigration of third-country nationals. For an economic tally of the benefits and costs of migration, however, this focus on the destination countries is at best incomplete, since it leaves out the impact of migration on the sending countries and on the migrants themselves. In this section we discuss the impact of migration on aggregate income, income distribution and employment. The effects of migration in closed economies Public concerns about labor migration often seem to reflect the most simple case of an otherwise isolated economy, where only one good is produced and the labor supply of natives is inelastic. In this case, the labor market bears the whole burden of adjustment to migration. Assume that the good is produced with capital, skilled labor, and unskilled labor, and that the production technology has constant returns to scale. Capital and both types of labor are complements, while high-skilled and low-skilled labor are imperfect substitutes. In this setting, an additional supply of low-skilled labor through migration will reduce the wages of low-skilled workers, raise the income of capital, and expand production. The impact of migration on high-skilled labor is ambiguous: the fall in wages for low-skilled workers may lead to the substitution of high-skilled workers by less skilled ones, while the scale effect increases the demand for high-skilled labor. The total effect on the income of natives is positive. The converse holds for the sending country: wages for low-skilled labor increase, income from capital falls, and the total effect on the income of those left behind is negative (Wong 1995; Bauer and Zimmermann 1997; Brucker ¨ 2002). Thus, in the absence of remittances and other transfers to the sending countries, the receiving countries benefit from immigration while the sending countries lose. However, inequality in the distribution of factor income increases in the receiving country and is reduced in the sending country. The result, that immigration increases the aggregate income of natives, relies on the assumption that labor markets clear. This changes if we consider wage rigidities and unemployment. Assume that wages of manual workers are fixed above equilibrium levels by a bilateral bargaining monopoly of trade unions and employer federations and that wages for high-skilled workers are flexible. As a consequence, part of the unskilled labor force is unemployed and the wage for skilled labor is below its equilibrium level. The impact of migration on wages and employment depends then on the collective wage setting: in the most extreme case, wages do not respond to unemployment at all, such that an additional

114

Immigration and the Transformation of Europe

supply of unskilled labor through migration simply increases aggregate unemployment of unskilled workers and the aggregate income of natives falls.2 If wages adjust partially to the additional labor supply, the effect of the immigration of unskilled labor on the income of natives is ambiguous: production expands, the rate of return to capital increases, the wage for unskilled labor falls, and the unemployment rates increase. In contrast, the immigration of skilled labor reduces unemployment and so increases the aggregate welfare of natives more than in the case of flexible labor markets. However, the latter result depends on the assumption that wages for skilled labor are flexible. Altogether, the impact of migration on welfare in the receiving and sending countries in closed-economy models depends highly on the assumptions about the wage-setting mechanism. The implications of East–West migration in Europe have been simulated in a closed-economy framework by Brucker ¨ (2002) and Boeri and Brucker ¨ (2005), employing actual differences in factor endowments and factor productivities between the EU-15 and the new Member States of the EU. Under the assumption that labor markets clear, the migration of 1 percent of the population in the East would result in a GDP gain of 0.3 percent in the total region (sending and receiving countries). This is a substantial gain, presumably higher than that resulting from an additional liberalization of goods and capital markets. However, these gains are not equally distributed across countries and different factors of production. The income of migrants increases by around 120 percent, which implies that some 90 percent of all gains from migration accrue to the migrants and their families. In the sending countries, the GDP falls by 0.8 percent, wages increase, and capital income falls. The net effect for native income is negative but negligible (0.001 percent). In the receiving countries, the GDP increases by 0.7 percent, the wages of manual workers fall by 0.5 percent and those of non-manual workers by 0.1 percent, while the incomes of capital owners increase. The net gain for the native population is only 0.002 percent. The picture changes if we make realistic assumptions about wage rigidities. In this case, the GDP gain in the total region shrinks to 0.2 percent and the total income of natives in the receiving countries falls by 0.2 percent, while that of natives in the sending countries increases by 0.3 percent. The reason for this result is that the unemployment rate increases in the receiving country by 0.2 percentage points and falls in the sending countries. This creates a net loss for the population in the receiving countries both by increasing the unemployment of natives and redistributing income to the migrants through welfare benefits, since the 2

This case is considered by the famous Harris-Todaro (1970) model.

Economic consequences of immigration in Europe

115

migrant population is more than proportionally affected by unemployment under realistic assumptions. As chapter 3 summarized, whether or not welfare benefits distort the incentives for migration (and so reduce welfare) is widely debated. In simulations it can be shown that a higher level of welfare benefits reduces the income of natives in the receiving countries both directly, by redistributing income from natives to migrants, and indirectly, by increasing the scale and changing the composition of the migrant population. But under realistic assumptions about the elasticity of migration with respect to (expected) income differentials, simulation results by Brucker ¨ (2002) and Boeri and Brucker ¨ (2005) show the differences in welfare benefits across the EU Member States have only a negligible impact on the scale and composition of migrants. More importantly, a higher level of welfare benefits tends to increase welfare in the total region under the assumption of realistic unemployment rates of migrants. The reason for this counterintuitive result is that higher welfare benefits increase the incentives to migrate not only for those who will later become unemployed, but also for those who will later be employed, since they reduce their social risks. Consequently, we have a higher scale of migration, which in turn creates a higher income in the total region. The principle of equal treatment in the EU, which guarantees migrants the same access to welfare benefits as natives, is therefore justified if the target is to increase total income in the Community. Another important feature of migration is that it can “grease the wheels of the labor market” in economies with regional wage and employment disparities (Borjas 2001). Consider the case of an economy that is characterized by a certain gap in wages and unemployment levels, but where the net benefit from moving from the poor to the rich region is zero for the marginal migrant. Hence, domestic migration has ceased. Note that this is a realistic scenario for many EU countries. The average GDP and wage levels of many regions are at 75 percent of the country average or below, but domestic (internal) migration is almost negligible. International migration can in this case create additional gains. If migrants move to the prosperous regions with high wages and low unemployment, the total gains for GDP tend to increase relative to the case of a country with homogeneous regions. Moreover, the increase in the unemployment rate of the receiving country is much lower. If we assume that wage setting is centralized, migration can even reduce the national unemployment rate. The additional labor supply in the high-wage region tends to reduce wages there, which in turn also reduces wages in the low-wage region if minimum wages are centrally determined. As a consequence, the unemployment rate falls in the low-wage region and the national income of

116

Immigration and the Transformation of Europe

natives tends to increase (Boeri and Brucker ¨ 2005). Thus, in the case of regional wage and employment differentials, the gains from migration for the total region are higher, and the negative consequences for the receiving countries are mitigated or even reversed into positive income effects. Migration can create problems for the receiving countries if we relax the assumption of clearing labor markets and consider the case of wage rigidities and unemployment. However, if we also consider the effects of migration on the welfare of migrants and natives in the sending countries, the total net effect of migration is positive and substantial. If we furthermore consider the realistic case that migrants tend to move into the prosperous regions of the receiving countries, it is likely that the negative effects for natives are small or even positive there. The growth effects of migration So far we have analyzed the effects of migration from a comparative-static perspective, ignoring the dynamic effects of migration that result from the accumulation of physical and human capital. Adding dynamics to the comparative-static model does not change the overall picture. Consider first the case of a one-time migration shock. In standard neoclassical growth models of both the Solow type (1956) with a constant saving rate and the Ramsey-Cass-Koopmans type with an endogenous saving rate, the capital stock will grow until we eventually achieve the same proportion between capital and labor as before the shock. As a consequence, the receiving country will experience a higher growth rate for a transitional period, and, at the end, wages and profits will be the same for all natives as before. The effects of migration we have sketched above will disappear over time. Consider now the case of a permanent inflow of labor through migration. Permanent immigration reduces the per capita endowment of capital and, hence, increases the marginal productivity of capital in the receiving country. In the steady state, the physical capital stock, output, and consumption per capita of the total population fall in the receiving country as long as the endowment of migrants with physical and human capital is below that of natives. However, the capital stock, income, and consumption per head of the native population increase with the immigration of labor. The converse holds for the sending country. Thus, under the assumption of clearing-labor markets, natives in the receiving countries gain from a permanent inflow of labor through migration. In case of nonclearing labor markets, the results of dynamic models are similar to those of comparative-static models.

Economic consequences of immigration in Europe

117

The effects of migration in open economies So far we have assumed that the economy is closed, in which case labor markets bear the whole brunt of adjustment. Immigration, however, does not necessarily affect wages and employment in the case of an open economy. In the standard Heckscher-Ohlin-Samuelson (HOS) model, factor prices depend on the prices of traded goods, not on domestic factor endowments. An influx of labor is completely absorbed by decreasing imports of labor intensive goods, reducing exports of capital intensive goods, and shifting the output-mix toward labor-intensive goods (the socalled Rybczynski effect). As long as immigration is not large enough to change the patterns of specialisation, it does not affect relative wages and the distribution of income in either the receiving or the sending country (Wong 1995). If the standard assumptions of the HOS model are relaxed, the picture may change. Differences in the level of technology, complete specialisation in the production of different kind of goods, immobility of factors between sectors, and the existence of sectors that produce non-tradable goods can imply that trade can complement migration, and, hence, even enforce the wage and employment effects (for a discussion, see Venables 1999; Trefler 1997). Moreover, if foreign and domestic goods are only imperfect substitutes in consumption, trade can mitigate the effects of changes in factor endowments only to a limited extent. Nevertheless, models that ignore the integration of national economies in international markets may tend to exaggerate the impact of migration on labor markets. Whether the open- or closed-economy models are more appropriate to analyze the impact of migration is an open question. Despite the global trend of internationalization, real economies have large service sectors, which still produce largely non-tradable goods, as well as manufacturing sectors, which produce largely for international markets. The critical question is which sector determines the marginal demand for labor, or, more precisely, the marginal demand for the type of labor that immigrates. If it is a manufacturing sector, then the open-economy model would be more appropriate. If it is a service sector, then a closed-economy model would be suitable. Interestingly enough, complex CGE models that consider a large number of sectors that produce both tradable and non-tradable goods and apply an open-economy framework, yield very similar results to those in the closed-economy simulation model discussed above. For example, a number of studies have calibrated the impact of the EU’s eastern enlargement on GDP, wages, and employment in complex computable

118

Immigration and the Transformation of Europe

equilibrium models. Keuschnigg and Kohler (2002) assess the impact of the eastern enlargement on Austria in an open-economy CGE model, assuming, inter alia, that the number of unskilled workers increases by 10.5 percent and the number of skilled workers by 2.1 percent. As a result, the wages for unskilled workers drop by 5 percent while those of skilled workers increase by 2.7 percent. The respective elasticities are similar to those we found above if we consider that our simulations are based on a 1 percent increase. Heijdra et al. (2002) calibrate the effects of migration, trade, and fiscal transfers for national welfare in an openeconomy CGE model, and find, inter alia, that migration in the context of enlargement increases the German GDP by 0.7 percent. Given that the simulation relies on the assumption that the manual-labor force increases by 6.2 percent and the non-manual labor force by 0.8 percent, this result is slightly below our estimates. In another study, Kohler (2003) finds overall GDP gains from migration in the context of eastern enlargement of 1.2 percent for Germany, using the same migration scenario as the study as Hejdra et al. (2002). Thus, the GDP effect here is slightly higher than in our projections. Finally, Brucker ¨ and Kohlhaas (2004) have simulated the impact of migration for Germany in a CGE model employing different assumptions about the education levels of the migrant population. They find that wages can decline by 0.5 to 0.6 percent at an immigration level of 1 percent of the labor force, while the unemployment rate increases by 0.02 to 0.1 percentage points. Again, these results are in the range of our findings. Altogether, although more complex CGE models capture both the dynamic effects of migration and its effects on the structure of production and trade, they yield very similar results as the comparative-static model of a closed-economy. Evidence from econometric studies The simulation results of the CGE models are relatively consistent and robust with regard to the underlying assumptions, as the preceding sections have shown. Nevertheless, all CGE models rely on a number of arbitrary assumptions and may therefore over- or understate the actual effects of migration. It is therefore instructive to confront the simulation results with empirical evidence from econometric studies. A large empirical literature has evolved in Europe which assesses the impact of migration on wages and employment in the receiving countries. Most of these studies rest explicitly or implicitly on the one-good, closed-economy framework. One branch of the empirical literature estimates traditional wage equations, in which the wage is explained by a vector of human capital characteristics, macro-variable controls for regional

Economic consequences of immigration in Europe

119

or sectoral demand shocks, and a measure for the density of migrants. Another branch of the literature estimates the elasticity of substitutionality of different types of labor in CES- or translog-production functions. Both approaches rely on the theoretical assumption that a change in the factor proportions through migration involves a change in the income of the respective factor. The income of the respective factor will fall if migrant labor is a net substitute and increase if it is a net complement. The empirical literature focuses mainly on the impact of migration on wages and employment, leaving aside its impact on capital income and GDP growth since detailed data on capital stocks are usually not available.3 Most studies rely either on a cross-section of regions or sectors and use variations in migrant density to identify the impact of migration on wages and employment. This approach suffers from two fundamental problems (Borjas 1995). First, economies are not considered to be able to adjust to an increasing supply of labor through a change in the structure of traded goods and output mix. Even if migration does affect wages or employment at the national level, the presence of trade between regions along with capital flows and the migration of natives may eventually result in the equalization of factor prices across regions. As a consequence, the actual impact of migration cannot be measured using the variance of wages across regions. However, these studies may capture the short-term impact of migration on labor markets, since it takes time for regional wage and employment differentials to disappear. Second, migrants tend to move into prosperous regions or sectors. A naive regression of local wage levels or unemployment rates against the share of foreign citizens might yield the spurious result that migration leads to higher wages or increasing employment. This problem can be circumvented by estimating using first differentials, by regressing the change in wages or employment against the change in the migrant density, to see if migrants choose their location on the basis of present wage and employment levels. In this case, migrant density is not affected by changes in wages or employment, such that we can attribute the correlation between the changes to the impact of a change in migrant density on the change in wages or employment. However, if location decisions are affected by expectations for wage or employment growth, the problem remains (Friedberg and Hunt 1995). Many estimates rely therefore on the instrumental variable technique to control for the endogeneity of migration decisions. The challenge is to 3

As a consequence, studies relying on a production function framework assume that labor and capital are additive-separable factors. Many authors argue that this is not imprudent, because the assumption of separability between capital and labor is confirmed by empirical evidence (e.g., Bauer 1997).

120

Immigration and the Transformation of Europe

find suitable variables that are sufficiently correlated to migration density but also uncorrelated to wage or employment variables. The choice of suitable instruments is a controversial topic in the literature, such that a good deal of uncertainty surrounds all empirical findings. Differences in empirical findings can largely be traced back to different methods to control for endogeneity. Table 5.1 summarizes the findings of a number of empirical studies on the wage impact of migration in Europe. Columns (6) to (9) report the elasticity between a change in wages as a percentage and an increase in the proportion of foreign citizens of 1 percentage point, which is equal to an increase in employment through migration by 1 percent. Some studies report the elasticity between a percentage change in wages and the increase of the proportion of foreign citizens as a percentage, even though it is hard to interpret, as the proportion of immigrants varies greatly across countries and different time periods. The elasticity between an increase in wages and the proportion of foreign citizens of 1 percent has therefore been converted at the sample mean. Altogether, the empirical findings summarized in Table 5.1 indicate that, with the exception of few outliers, a 1 percent increase in the labor force through migration yields a change in native wages in a range between minus and plus 1 percent. The majority of the studies indicate that the change in native wages is in a range between minus and plus 0.3 percent. It is moreover important to note that many of the non-instrumented studies find that a complementarity exists between immigrant and native labor, the wages of natives tending to increase with labor migration. Note that most of the recent studies, which are more careful with the use of instrumental variables, find lower results. However, due to the methodological problems of measuring the impact of migration, a good deal of uncertainty surrounds the empirical conclusions. The finding that immigration has only small, if any, effects on native wages is not surprising per se. If the European system of wage determination involves wage compression, one would expect labor markets to adjust to the influx of foreign labor by increasing unemployment, on the basis of the one-good framework sketched above. However, there is also little evidence that immigration leads to higher unemployment. Neither studies based on macro data nor those based on micro data indicate that migration has a large impact on employment. Conversely, Gross (1999) finds that an increase in the immigration rate by 0.1 percentage points reduces the unemployment rate in the long term by 0.15 percentage points, while in the short term, the unemployment rate may increase by up to 0.05 percentage points. Gross argues that this counterintuitive result is consistent with a number of models in which either migrant workers

Economic consequences of immigration in Europe

121

complement native workers or the demand of migrants for goods and services creates a net gain in jobs for host labor markets independent of their participation in the labor market. A number of the studies based on micro data sets find no, or only weak, correlations between the density of migrants in a region or branch and individual displacement risks, while others find larger effects. Again, differences in the empirical findings can largely be traced back to different methods to control for the endogeneity problem. The findings of the empirical studies on the employment impact of migration cannot easily be compared to the results of the simulation model, since most of the findings refer to individual employment risks instead of unemployment rates. Nevertheless, we observe a similar trend as in the case of the studies on the wage impact (Table ??). The calibration of models with rigid wages indicates that the total unemployment rate may increase between 0.2 and 0.4 percent, and that of manual workers by 0.4 to 0.8 percent, given an increase of the workforce by 1 percent through migration (Boeri and Brucker ¨ 2005; Brucker ¨ 2002). With the exception of the instrumented regressions in Winter-Ebmer and Zweimuller ¨ (1994) and Hofer and Huber (1999), the empirical studies find that the individual displacement risks increase between zero and 0.2 percent. Thus, although results vary according to the methodologies employed, the empirical findings again indicate that the impact of migration is below that predicted by simulations of the closed-economy models of international migration. Unsurprisingly, the open economy framework might be more appropriate for analyzing the impact of migration on labor markets than models that rely on the case of an isolated economy. Overall, the empirical evidence indicates that the impact of migration on wages and employment in host labor markets is much smaller than is widely believed. The empirical evidence provided for Europe is largely consistent with that for the United States (see the review of empirical studies in Borjas [1995] and Friedberg and Hunt [1995]), though the assessment of the labor market impact of migration is controversial there as well. Hanson and Slaughter (2002) find evidence for the existence of Rybczynski effects, suggesting that the actual impact of migration is relatively small. In contrast, Borjas (2003) estimates, on the basis of a factorproportions approach at the national level, that a 10 percent increase in the labor supply through migration reduces wages for unskilled workers by 4 percent, within the range suggested by the simulation of the wage effects of migration in a closed-economy setting. Still, Borjas’ result is an outlier in the empirical literature. The simulation results of the CGE models reported above form a bottom-line with respect to the wage and

Country

Germany

Germany

Germany

Germany

Italy

Study

DeNew/ Zimmermann (1994a)

DeNew/ Zimmermann (1994b)

Haisken-DeNew/ Zimmermann (1995)

Bauer (1997)

Gavasto/Venturini/ Villosio (1999)

Household panel (GSOEP, 1982–1989)

Household panel (GSOEP, 1982–1989)

Household panel (GSOEP, 1982–1989)

Database

Wage equation

Panel, Social Security Archive (SSA).

Production Cross-section, function (translog) IAB-Employment Sample (1990).

Wage equation

Wage equation

Wage equation

Approach

Estimation in first differences

Instrumental variables

Instrumental variables

Instrumental variables

Instrumental variables

Control for endogeneity

–

–

0.00

0.01

−0.04

– –

– 0.00

–

−0.05b −0.06b

0.00b 0.00c

0.00b

0.00e

0.00f,g

0.01e

–

0.01e

Change in wages at an increase of the foreigner share by 1%-pointa

Medium- Highskilled skilled

Impact on native wages

All less-skilled

Table 5.1 Review of empirical studies: The wage impact of international migration on employment

Austria

France

Austria, Germany

Hofer/Huber (1999)

Hunt (1992)

Winter-Ebmer/ Zimmermann (1999)

Panel regression across industries

Differenced cross-sectional regression (natural experiment)

Wage equation

Wage equation

Wage equation

Industry-level data (Austria: 1987–1994, Germany: 1986–94).

Cross-section of regions, 1962–1968

Social security records

IAB-employment sample (1990–1995)

IAB-employment sample (1990–1995)

Instrumental variables

Instrumental variables

Instrumental variables

Estimation in first differences

Instrumental variables

–

–

−0.32 (Austria), 0.02 (Germany)

c Unskilled

0.00b

0.03

−0.02b

0.0 – 0.8

0.00

0.01

−0.01

elasticities from some studies are converted at the sample mean by the author. b Blue-collar workers. workers. d Skilled blue-collar workers. e White-collar workers. f High-skilled white-collar workers. g Derived from the cross-elasticities for several foreigner groups weighted by their present shares in foreign employment.

Germany

Trabold/Trubswetter ¨ (2003)

a Point

Germany

Brucker/Kreyenfeld/ ¨ Schr¨apler (2001)

–

–

0.01

–

–

–

0.00e

0.01

0.00e

Country

EU-12

EU-12

France

Germany

Italy

Germany

Study

Gang/Rivera-Batiz (1996)

Gang/Rivera-Batiz (1999)

Hunt (1992)

Muhleisen/ ¨ Zimmermann (1994)

Venturini/Villosio (2002)

Winkelmann/ Zimmermann (1993)

Estimation of individual (un-)employment probability (Poisson-distribution)

Estimation of individual (un-)employment probability (Probit)

Estimation of individual (un-)employment probability (Probit)

Differenced cross-sectional regression (natural experiment)

Estimation of individual (un-)employment probability (Probit)

Estimation of individual (un-)employment probability (Probit)

Approach

Household panel (GSOEP, 1974–1984)

Cross-section of individuals (Labour Force Survey, 1993–1997)

Household panel (GSOEP, 1982–1989)

Cross-section of regions, 1962–1968

Cross-section of individuals (Eurobarometer 1988)

Cross-section of individuals (Eurobarometer 1988)

Database

Weak, but significant positive correlation between unemployment probability and foreigner density

Positive, but insignificant correlation between unemployment probability and foreigner density

Impact on employment

No

No, tests reject endogeneity

No

No significant increase of unemployment frequency of natives

Ambiguous signs for both, displacement risks and entry probability

No significant impact of foreigner share on unemployment probability

Instrumental An increase of the foreigner share by 1% variables point increases native unemployment rate by 0.2% points

No

No

Control for Endogeneity

Table 5.2 Review of empirical studies: the impact of international migration on employment

Germany

Germany

Austria, Germany

Brucker/ ¨ Schr¨apler/ Kreyenfeld (2001)

Pischke/Velling (1997)

Winter-Ebmer/ Zimmermann (1999)

Panel regression across industries

Estimation of different panel models

Estimation of individual (un-)employment probability (Probit)

Estimation of individual (un-)employment probability (Probit)

Estimation of individual (un-)employment probability (Probit)

Industry level data 1987–94 (Austria), 1986–94 (Germany)

Regional data (labour market areas and counties), 1985–1989

Cross-section of individuals (IAB-employment sample, 1990–1995)

Cross-section of individuals (Social security records 1991–1994)

Cross-section of individuals (Social security records 1988–89, 1991–92)

have been converted at the sample mean by the author.

Austria

Hofer/Huber (2001)

a Elasticities

Austria

Winter-Ebmer/ Zweimuller ¨ (1994)

Instrumental An increase in the foreigner share by 1% variables point increases (decreases) total employment by −0.002% (Austria) and 0.025 % (Germany), and native employment by −0.25 % (Austria) and −0.04% (Germany).

Instrumental Weak correlation between foreigner variables share and unemployment rate with ambiguous signs

Instrumental An increase in the foreigner share by 1% variables point increases individual displacement risks by 0.2%

Instrumental An increase in the foreigner share by 1% variables point increases individual displacement risks by 0.8%

Instrumental An increase of the foreigner share by variables 1% point increases individual displacement risks by −1% to 1.1% (instrumented), 0.1% – 0.4% (not instrumented)

126

Immigration and the Transformation of Europe

employment effects of migration in the receiving countries, although the actual impact may be considerably smaller. The economic status of migrants, the acceptance of migrants, and remittances The economic status of immigrants in Europe This section adds some insight from a micro perspective by comparing migrants to the native-born population with respect to their economic performance and position in the host society. We make use of micro data from various European panel data sets: the European Community Household Panel (ECHP; see Wirtz and Mejer [2002], and Adser`a and Chiswick, chapter 4 in this volume), the British Household Panel Survey (BHPS; see Taylor 1993) and the German Socio-Economic Panel (SOEP; see Wagner et al. 1993, Haisken-DeNew and Frick 2005). The primary focus of our analysis is on the self-supporting capacities of migrants as proxied by their pre-government income position and on their role in the income redistribution process induced by taxation and social security systems.4 The main aim of this exercise is to provide information about the “net” economic performance differential between immigrants and native-born inhabitants of various EU countries. To measure the individual state of economic integration (which is mostly labor-market oriented) we analyze households with prime-aged heads only (20 to 60 years of age) to reduce the impact of different age structures among immigrants and natives across Europe. We categorize immigrant households into two groups, “mixed” and “non-mixed” households, to reflect their state of integration. Whereas all adult members of non-mixed households are foreign-born, in mixed households at least one adult is native-born and at least one other is foreign-born (these are mainly couples with one immigrant and one native partner). We expect people living in mixed immigrant households to perform better than those in non-mixed households because they are more integrated into the host society. Buchel ¨ and Frick (2004) suggest that this concept of measuring the individual degree of integration is valid and adds to

4

The effect of redistribution is measured by subtracting the relative income position based on pre-government income from that based on post-government income for each individual. This metric measure is positive (negative) for those who improve (worsen) their income position as a result of the redistribution process, subtracting taxes and social security contributions on the one hand and adding public transfers (including public pensions) on the other.

Economic consequences of immigration in Europe

127

Table 5.3 Population living in private households with prime-aged heads1) in selected EU countries, 1994–982) , by household immigrant status (in percent) Native-born

Denmark Luxembourg 3) Ireland Italy Spain Austria 4) Great Britain 5) West Germany 6)

Immigrant

Total

All adult household members are native-born

All adult household members are immigrants (“non-mixed” HH)

At least one adult household member is immigrant (“mixed” HH)

Total

89.6 53.8 88.7 95.4 96.3 86.6 88.8 79.2

4.6 28.5 1.8 0.7 0.6 6.9 3.8 12.2

5.8 17.7 9.5 3.9 3.1 6.5 7.4 8.6

10.4 46.2 11.3 4.6 3.7 13.4 11.2 20.8

100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0

1) Head

of household is 20 to 60 years of age. 2) Average over the 1994–98 observation years 1994–96 only. 4) Observation years 1995–98 only. 5) Based on period. BHPS data 1994–1998. 6) Based on SOEP data 1995–99. Source: ECHP-UDB waves 1–5; BHPS waves 4–8; SOEP waves 12–16 (weighted). 3) Observation

commonly used indicators such as country of origin and duration of stay in host country. Table 5.3 shows that both the proportion of immigrants5 and their state of integration vary markedly across the countries under consideration. The highest proportion of immigrants is found in the small country of Luxembourg, which lies at the heart of the EU. This may be due to the high concentration of foreigners working in Luxembourg’s financial and banking sector, as well as to the rather large group of migrant workers from Portugal. In the Mediterranean states, represented by Italy and Spain, there are relatively few (legal) immigrants. Because of their relatively weak economic performance in the 1960s and 1970s, these countries were long characterized by emigration rather than by immigration (e.g. the “guest worker” movement of low-skilled labor to the German automobile industry and mining sector). The few immigrants to these countries are typically well integrated; most of them live 5

The figures in Table 5.3 may deviate from official statistics for several reasons: defining the term foreigners in the present study to mean foreign-born rather than non-citizens, restricting the sample to prime-aged heads, and defining immigrant status at the household level, which overstates the number of individual immigrants.

128

Immigration and the Transformation of Europe

with members of the native population (see also Adser`a and Chiswick, chapter 4). As a result of this out-migration from Mediterranean countries as well as the massive influx of immigrants with German ancestry (Aussiedler) from eastern European countries since the late 1980s, West Germany has a rather high share of immigrants, most of whom live in non-mixed households. Surprisingly, the proportion of immigrants in the United Kingdom is markedly lower than in Germany, and, with two-thirds of them living in the same household as a member of the native population, these immigrants appear to be somewhat better integrated. By means of regression models we analyze the determinants of pregovernment income position (top panel in Table 5.4) and the impact of redistribution (bottom panel in Table 5.4).6 Control variables include various socioeconomic characteristics such as sex and age, household composition, health status, education, and labor market status. Immigration is controlled in a first specification (A) by a simple dummy, assuming immigrants to be a homogenous group within each country; in a second specification (B), we control for heterogeneity among immigrants with respect to their region of origin (EU versus non-EU countries) and integration status (measured in terms of years since migration and the mixed/non-mixed status). Immigrants to Denmark and West Germany in particular have a much lower pre-government income position than the native-born population. When taking social differences into account, immigrants to Luxembourg have a significantly higher pre-government income position than the native-born population. The same is true for Ireland, but only at the 10 percent significance level. The redistribution analysis provides an almost perfect counterpart to these findings. As a result of redistribution, the outperforming immigrant populations in Luxembourg and Ireland lose income, whereas the under-performing immigrant populations of Denmark and West Germany gain. Although immigrants to Great Britain do not show a significant income advantage over the native-born population, they do pay significantly more into the system as a whole. The difference between the market performance of the Italian, Spanish, and Austrian immigrant populations and the respective native-born populations is non-significant, which is also reflected in the redistribution process. 6

To make use of the panel nature of the data, random-effects GLS models are applied to control for unobserved heterogeneity occurring in the context of repeated observation of households over time (cf. Buchel ¨ and Frick 2005).

−4.626 5.168 −22.574 7.834 42.579 −1.093 2.838 11046 3225 .0340

−13.882 −17.404∗ −13.368 −2.574 0.298 0.146 2223 835 .1602

13.365 9.837 16.959∗ 36.67∗∗ −0.973 2.338 10633 2896 .1798

0.014 −0.363+ 0.098 −1.499∗∗ 0.011 −0.014 .2319

−9.030∗

0.160 0.127 0.273 −0.265 0.012 −0.036 .2799

−0.243 −1.169∗∗ −0.603∗∗ −1.918∗∗ 0.045∗ −0.096∗ .2175

0.109+

Ireland

15.372∗∗

0.155∗

−0.596∗∗

Lux.

−0.276 −0.191 0.281 −0.327 0.015 −0.052 .1830

−0.035

Spain

12.105 17.537 −3.958 13.630 −0.869+ 1.503∗ 22218 6178 .1049

−4.386

7.030 13.398 −1.005 23.721+ −2.274∗ 7.699∗∗ 13688 3890 .1427

2.580

Redistribution

−0.250 −0.715∗ 0.065 0.037 0.011 −0.018 .1349

0.044

Italy

All estimations include various control variables for socioeconomic status and time effects. +significant at 10%; ∗ significant at 5%; ∗∗ significant at 1% Source: ECHP-UDB waves 1–5; BHPS waves 4–8; SOEP waves 12–16.

Specification A Immigrant household Specification B Origin: EU/mixed HH Origin: EU/non-mixed HH Origin: Non-EU/mixed HH Origin: Non-EU/non-mixed HH Years since migration Years since migration (sq) Observations Groups R2

Specification A Immigrant household Specification B Origin: EU/mixed HH Origin: EU/non-mixed HH Origin: Non-EU/mixed HH Origin: Non-EU/non-mixed HH Years since migration Years since migration (sq.) R2

DK

Pre-government income

7.425 22.182∗ 18.826∗ 8.423 −2.047∗∗ 3.628∗∗ 9162 2932 .0903

−2.646

−0.074 −0.185 −0.358∗ −0.313∗ 0.030∗∗ −0.038∗ .1497

0.058

Austria

−1.028 7.545 −2.906 −1.040 0.151 −0.657 16683 5077 .2541

−2.386∗

−0.011 −0.595∗∗ 0.064 −0.226+ 0.002 0.012 .3910

0.054

GB

Table 5.4 Pre-government income positions and redistribution effects for households with prime-aged heads in selected EU countries, 1994–98

5.756∗ 6.897∗∗ 9.308∗∗ 10.588∗∗ −0.849∗∗ 1.934∗∗ 19403 5336 .2055

2.050∗∗

−0.064 −0.254∗ −0.288∗∗ −0.662∗∗ 0.039∗∗ −0.105∗∗ .2299

−0.174∗∗

Germany

130

Immigration and the Transformation of Europe

Differentiating the immigrant groups according to their region of origin and state of integration allows for a better control of the heterogeneity of the immigrant population across Europe.7 In all countries analyzed, mixed households in which an immigrant from the EU resides with an adult member of the native-born population do not show any significant differences in economic performance compared to households of nativeborn individuals only. This holds for both of the performance indicators analyzed – pre-government income position and change in the relative income position due to the redistribution process – with the exception of West Germany in the latter case. Those who immigrate to Denmark, Ireland, Italy, the United Kingdom, and West Germany from EU countries and who live alone or with a partner from the same ethnic origin (non-mixed immigrants) are worse off economically than the native-born population, even when controlling for duration of stay and numerous other socioeconomic background variables. Since we are not able to measure the skill potential of individuals in our immigrant sample perfectly – that is, we cannot measure language fluency – we hesitate to interpret this result as an indicator that immigrants from the EU to these countries are discriminated against. We prefer to interpret it to indicate atypically low (unmeasured) skill levels within these groups. Furthermore, it appears that non-mixed immigrants lose out during the redistribution process in Luxembourg, but profit from it in Austria and in West Germany. While we expect reduced self-supporting capacities among immigrants in non-mixed households, the unexpected result in Luxembourg may well reflect the presence of high-performing employees in the country’s financial and banking sector. In general, integration in the sense of living with a native-born adult shows no significant income differential between immigrants from nonEU countries and the native-born population. Exceptions are Denmark, Austria and West Germany, where non-EU immigrants in mixed households show a significantly lower market performance, and accordingly profit from redistribution. 7

Concerning the additional control variables, very similar patterns can be observed across countries: households with a middle-aged, well-educated head who is in good health and who has not previously been affected by unemployment fare better economically than others. Two-parent households have higher market incomes, and the presence of (many) children in the household is negatively linked to income. Finally, the European tax and contribution systems seem to be “fair” to the extent that those socioeconomic groups with a weaker pre-government income position tend to profit from the redistribution process. The effects of the additional control variables remain essentially unchanged when compared to the results of the simple dummy control for immigration status described above.

Economic consequences of immigration in Europe

131

The group assumed to be least well assimilated are immigrants from non-EU countries who live in non-mixed households. In terms of pregovernment income, their households are outperformed by the nativeborn population in almost all countries under consideration; this effect is statistically significant in Denmark, Ireland, Austria, the United Kingdom, and West Germany. However, it is only in Denmark, West Germany and Spain that they profit substantially from the redistribution process. A general hypothesis predicts that duration of stay in the host country exerts a positive effect on the economic performance. However, our data only confirm this improved market performance with time among immigrants to Denmark, Austria, and West Germany. Notably, these are essentially the countries for which we identified the highest economic penalties for immigrant status. In these countries, more recent immigrants are the main economic under-performers (consistent with Adser`a and Chiswick). With increasing duration of stay, however, their economic position improves rapidly, signalling successful societal integration. As expected, this effect is not linear but diminishes over time. The results for the redistribution analyses are essentially in line with the results on pre-government income, but with two notable exceptions. First, immigrants who live in the host country for longer are net payers to the social system in Italy and Spain, although their pre-government income position is not affected by duration of stay. In Denmark, given the extremely poor economic performance of immigrants to this country, there is no statistically significant indication that their need for support via redistribution is reduced over time. Immigrants to Denmark remain dependent on public transfers. Summing up, there is considerable variation in the economic performance of immigrant populations in different EU countries. This may be linked to the heterogeneous conditions of entry to the EU states, which strongly influence the distribution of socioeconomic characteristics of the immigrant population, as well as to differences in country-specific strategies to promote immigration integration. These cross-country differences persist when we control in detail for socioeconomic characteristics of the individual in the household context and for indicators designed to tap the individual state of integration, such as years since migration and immigrant/native intermarriage. This suggests that institutional aspects such as access to the labor market and parts of the social security system that are restricted by citizenship or immigration status plays an important role in limiting the economic performance of immigrants. For further development of these themes, see the discussion in the next four chapters.

132

Immigration and the Transformation of Europe

The acceptance of immigrants by those who gain and lose from immigration in a society For a comprehensive evaluation of immigration it is important to take into account short-term distributional effects. People live mostly in the short term and gains from immigration that will be earned by future generations may not increase the utility of the current population. Therefore it is worthwhile analysing whether clear winners are more in favor of immigration than other groups of the population.8 We can do this in the case of Germany, where we have special data at hand. The German Socio-Economic Panel Study (SOEP) contains unique indicators that allow us to identify specific winners and losers of immigration, going beyond common indicators such as high educational and occupational level (Bauer et al. 2000). Like any indirect indicator they are imperfect, but they can provide valuable additional information. Two indicators of gains from immigration are how frequently respondents dine out in a restaurant and whether they employ a maid in their household. As many foreigners run restaurants, those who like to dine out gain, on average, from immigration. In many cases maids are of foreign origin. Thus, on average, households who employ a maid have a special advantage from immigration. Because not all restaurants are run by immigrants and not all maids are immigrants, we estimate a lower bound for the effect of maids of foreign origin and restaurants run by immigrants. An indicator for losses due to immigration is the concentration of foreign children in school classes. Given widespread public debate in Germany about how immigrants perform poorly in school, we assume that parents will be concerned by the presence of non-German speaking pupils in a class. Because it can be assumed that people with a higher level of education are more aware of an economy’s overall gains from immigration, we control for the respondent’s level of education. Age is a control variable because it is likely that with increased age, people become less and less flexible. The character of the place of residence is controlled for because in urban areas there is, on average, more experience with immigrants than in rural areas. In addition, equivalent household income is controlled for, calculated on the basis of the new OECD scale. The dependent variable “Concerns about immigration” is coded as a dummy variable where the categories “Somewhat concerned/Not concerned at all” are set to the value zero and the category “Very concerned” is equal to one. With regard to the effects of the share of foreigners in school classes we can analyze only parents with children aged seventeen 8

We are grateful to Ingrid Tucci for conducting these empirical analyses.

Economic consequences of immigration in Europe

133

Table 5.5 Odds ratios predicting the concerns of German adults about immigration in 2000 Odds ratio

z-value

0.760 ∗∗

2.69

0.929 0.869 ∗

1.40 2.25

Education level (Reference: low) High Middle

0.555 ∗∗ 0.896 +

7.74 1.84

Equivalent HH income Unemployed

1.000 ∗∗ 0.843 ∗∗

3.17 2.79

Community size (Reference: Middle) Small Big

0.982 0.890 ∗

0.35 2.18

Constant

−0.919 ∗∗

4.79

Observations Pseudo R2

9667 0.02

Maid in the HH Dining outside (Reference: Never) Sometimes Frequently

Absolute value of z statistics. +significant at 10%; ∗ significant at 5%; ∗∗ significant at 1%. Source: SOEP 1998 and 2000.

or eighteen, as only children of these ages were asked about the situation at school in the underlying SOEP survey. Table 5.5 shows that the coefficients for “dining out” and” “maid in household” are both significant, and reduce the probability of being worried about immigration.9 Table 5.6 shows that the probability of being very concerned about immigration is significantly higher for parents whose children attend schools with a high concentration of foreigners than for those whose children attend schools where there are only a few pupils of foreign origin.10 The importance of emigrants’ remittances All standard economic models of migration ignore the fact that emigration is not a one-way street. Many migrants continue to have extensive 9

10

Moreover, women are less likely to worry about immigration than men and less-educated respondents tend to be more concerned about this subject. Controlling, among other characteristics, for the educational level and income, unemployed respondents are less likely to worry about immigration than the rest of the population. We are grateful to Ingrid Tucci for providing the regression results. As expected, parents with a high level of education have significantly less concern than less-educated parents.

134

Immigration and the Transformation of Europe

Table 5.6 Odds ratios predicting the concerns of parents about immigration Variable

Odds ratio

z value

High Foreigners’ concentration Age Age2 Female

1.543∗∗ 0.928 ∗∗ 1.003 0.738 ∗∗

2.53 −2.29 2.26 −2.20

Educational level (reference: low) High Middle

0.519 ∗∗ 0.823

−2.72 −0.95

Constant

6.153 ∗∗

2.20

Observations Pseudo R2

1177 0.03

Absolute value of z statistics. +significant at 10%; ∗ significant at 5%; ∗∗ significant at 1%. Source: SOEP 2000–2003.

relationships with their home country. Some of them even go back (return migrants) or move back and forth as repeat migrants (see chapter 11, and Constant and Zimmermann 2003). In this context emigrants’ remittances are an important factor. They are not easy to measure, but several approaches capture them at least partly from macro or micro perspectives. Aggregated data from official statistics show that the overall volume of remittances from migrants more than doubled between 1988 and 1999. Just during the period 1999–2002, remittances by labor migrants to their (mostly developing) home countries increased by nearly 20 percent – from $67 billion to $80 billion (IMF 2003). Official estimates find that financial transfers crossed the $100 billion mark in 2003 (Migration News 2004). According to the IMF, financial transfers by migrants residing permanently in foreign countries are the second largest source of income to developing and newly industrialized countries, after foreign direct investment (FDI). In 2001, emigrants’ remittances worldwide ($72.3 billion) exceeded the total global volume of private FDI and public development aid by approximately 42 percent (IMF 2003). Emigrants’ remittances comprised between 20 and 50 percent of the national income in Eritrea, Cape Verde, India, Yemen, Jordan, Lesotho, Lebanon, Nicaragua, Palestine, Pakistan, and the Sudan. The ratio of migrants’ income transfers to export revenues was between 25 and 50 percent in Egypt, Bangladesh, Burkina Faso, Jamaica, Malawi, Morocco,

Economic consequences of immigration in Europe

135

Sri Lanka, and Turkey (International Organization for Migration 2004). In Turkey, for example, the volume of emigrants’ remittances is four times larger than that of FDI. Such information can be linked to micro data from the German Socio-Economic Panel Study (SOEP) which for 2002 shows that nearly 17 percent of all people of Turkish and former Yugoslavian origin living in Germany made private monetary transfers back to their home country (authors’ calculations). From a macroeconomic perspective, emigrants’ remittances can help relieve poverty and reduce trade deficits. Capital inflows from emigrants’ remittances can lessen debt pressure by increasing foreign exchange holdings and helping to pay off debts. Furthermore, emigrants’ remittances can improve the overall investment climate in the recipient country. The money that migrants transfer to their countries of origin provides an incentive for FDI, when combined with innovative entrepreneurial ideas and the necessary funds. And even if emigrants’ remittances only make their way into the economy through consumption by the recipient households, the increased purchasing power can still support domestic productivity. Highly qualified emigrants return to their countries of origin Indian emigrants to the United States provide a prime example of other benefits, in addition to emigrants’ remittances, which migration brings to the countries of origin. After working in the United States, many Indian IT specialists return to India (brain circulation), using their American connections to get contracts for work that they can carry out in India. In 2002, India registered revenues of $10 billion from the export of technology products and IT services and from outsourcing by numerous foreign companies. From India’s perspective, the initial outward migration and resulting loss of human capital (brain drain) is thus transformed into profit (brain gain). The effects of outward and inward migration play themselves out in the long term, and migration functions as a bridge between the countries of origin and the countries of immigration. The outward migration of highly qualified workers helps the recipient countries, but it also helps the countries of origin through the expansion and strengthening of trade relationships. Will immigration help alleviate the demographic burden? Populations are ageing rapidly around the globe. This secular process of ageing is, at 1.3 births per woman, particularly apparent in Germany

136

Immigration and the Transformation of Europe

and other countries of the EU. As Paul Demeny showed in chapter 2 there is consensus in the literature that international migration can mitigate, but not solve, the demographic problem in the receiving countries. According to a study by the United Nations, an average annual net immigration of 1.4 million people will be needed to keep the proportion of the working-age population in the EU stable until 2050 (UN 2000). This corresponds to an increase in the net annual inflow from 0.8 persons per thousand in western Europe between 1950 and 2000 to 3.7 per thousand over the next five decades. It is hardly realistic that immigration will accelerate to this level, and some authors argue that migration cannot slow the effects of a country’s ageing population sufficiently to make it a reasonable policy instrument for this purpose (Coleman 1992). We, however, tend to side with the many other authors who argue that migration can help mitigate the effects of population ageing to some degree (Borsch-Supan ¨ 1994; Straubhaar and Zimmermann 1993). A number of studies also try to assess the fiscal impact of migration on the source countries. Several cross-sectional studies have tried to estimate the fiscal contribution of migrants relative to natives (Blau 1984; Simon 1984; Simon 1994; Riphahn 1998), but they fail to capture the impact of migration on the demographic structure of the population. Others focus on the impact of migration on pay-as-you-go pension systems by changing the demographic structure of the population, but they do not consider the overall impact of migration on the revenues and expenditures of public finances and social security systems (Felderer 1994; Borsch¨ Supan 1994). We employ a more comprehensive approach, in which we calculate the net present value of the fiscal contribution of migrants in the framework of the generational accounting approach for Germany (Bonin et al. 2000). Similar results have been obtained in an estimate of the lifetime net fiscal contributions of migrants in Sweden (Storesletten 2000). Generational accounting is based on the intertemporal budget constraint faced by the public sector. If chain-letter (Ponzi) games are ruled out, the present value of future taxes must equal the present value of future government consumption and debt servicing. Generational accounting assesses the intertemporal sustainability of public finances by calculating both the expected taxes and expenditures of public finances, based on long-term projections of the underlying fiscal and demographic variables. The difference between the expected government consumption and debt servicing on the one hand, and the expected tax revenue on the other, is referred to as the sustainability gap. In the case of Germany with no

Economic consequences of immigration in Europe

137

migration, the sustainability gap has been calculated to be 6.1 percent of GDP per year. Migration can affect the sustainability gap of public finances in two ways. First, the net tax contributions of migrants directly affect the governmental budget balance. If the net contribution of migrants is positive, the sustainability gap declines. Second, immigration increases the number of potential taxpayers on whom future tax increases can be levied. Thus migration can reduce the sustainability gap by increasing the tax base, even if migrants’ contributions to the present budget are negative. Consider first the net effect of migrants to the balance of the public finances and social security systems. According to Bonin et al. (2000) and Bonin (2001), the net contributions of migrants to public finances vary with the age of the respective cohorts. While net tax payments – all social security system transfers and governmental budget expenditures – are positive over the remaining life cycle of migrants who immigrate between ages 11 and 48 years, those of the younger and older cohorts are negative. As an example, average net tax payments over the life cycle of a migrant who immigrates at age 30 amounts to €110,000, while a migrant who immigrates before his first birthday creates a net burden to the public finances of €60,000. At present, around 78 percent of immigrants belong to cohorts that contribute to a budget surplus over their remaining life cycle. The net contribution of the representative migrant to the public finances amounts to some €50,000. Consider now the overall impact of migration on the sustainability gap in the government budget in Germany. As stated earlier, the sustainability gap in public finances can be estimated at 6.1 percent of the German GDP under the assumption of zero migration. An annual lumpsum tax of around €1,300 per capita is needed to close this gap. An immigration of 200,000 persons per year reduces this gap by 5 percent through the migrants’ net contribution to the budget and by increasing the number of taxpayers. This implies that the lump-sum tax falls to €1,060 or by 18 percent per capita due to 200,000 more immigrants per year over a longer period. The net gain for the native population is even higher; as the sustainability gap declines by 24 percent overall (Bonin et al. 2000). These estimates are based on the present tax payments and transfer income patterns for the foreign population in Germany. As a consequence of relatively low human capital endowments, the net present value of the income earned by foreign citizens is around 20 percent below that of natives, and transfers for unemployment and social welfare exceed those of natives by 65 percent. However, higher transfers from unemployment

138

Immigration and the Transformation of Europe

insurance and social insurance do not cancel out the lower transfers made to foreign citizens relative to natives by German pension schemes. The impact of immigration on public finances will depend on the human capital endowments of future immigrant cohorts. If the skill levels of future migrant cohorts continue to decline, then their impact on public finances will be below that presented by the calculations here. However, recent immigrant cohorts from central and eastern Europe possess better human capital endowments than the average representative of the foreign population, such that the human capital endowments and labor market prospects of future migrant cohorts may improve. In this case, the net contribution of migrants to the public budget may even exceed the €50,000 found in the Bonin et al. (2000) and Bonin (2001) studies. It is reasonable to assume that the fiscal gains from migration in the other host countries in western Europe are similar to those in Germany, since they are experiencing the same demographic trend. Altogether, this generational accounting exercise demonstrates that, against the background of rapidly ageing populations, the fiscal gains for natives from future immigration in Europe are substantial. With reasonable assumptions for potential migration, it is rather unlikely that migration can seriously slow the population ageing process, but it can mitigate the fiscal consequences of population ageing significantly. Migration will aggravate the fiscal burden in the source countries of European migration, in particular in central and eastern European countries. The age of the population will increase even more rapidly there than in western Europe, since fertility rates have been declining sharply since the start of transition (Figure 5.1). This does not imply that migration is inefficient: the economic consequences of population ageing can be alleviated, if production factors – physical capital as well as human resources – move to those countries and regions where their productivity is highest. However, migration may lead to externalities. For example, if the education of a skilled migrant who is hired in western Europe is financed by government expenditure in central and eastern Europe, then migration creates a negative externality for the budget in the source country and a positive one for that in the host country. To compensate the source country for these expenditures would not only be fair, it would also help to improve the efficiency of the allocation of human resources. Summary: Economic benefits and costs of immigration for western Europe in the short and long term While most economists agree that migration will tend to increase aggregate income and the welfare of the migrants, it is an open empirical

Share of 20–29 aged in total population (males)

0.00

5.00

10.00

15.00

20.00

25.00

EEA

2005

Southern EU

2010

2020

traditional immigration countries

2015

CEEC-10

2025

2030

other CEEC

2035

2040

2045 Developing Countries

Figure 5.1. Share of the 20–29 age group in selected European regions. Source: Authors’ calculations based on the population projection of the World Bank (2002).

2000

2050

140

Immigration and the Transformation of Europe

question how migration will affect the welfare of certain groups of natives in the host and source countries. A further question is how aggregate income will end up being distributed between host and source countries. In this chapter, we have discussed these issues from different perspectives. The key results can be summarized as follows: r Our simulation exercises show that for the given difference in income levels in Europe, the gains in aggregate income from international labor migration can be substantial. Most of these gains accrue to the migrants. Wage rigidities affect the size of the gain in aggregate income, but even in the case of perfectly rigid labor markets a substantial gain remains. r The calibration of the impact of international migration in a closedeconomy model indicates that international labor migration can have an important impact on relative wages and, hence, can affect the distribution of income. While the inequality in income distribution increases in the host countries, it decreases in the source countries and in the total region. However, the empirical evidence available suggests the impact of migration on relative wages is much lower than that predicted by those migration models that neglect the impact on trade. In fact, migration may not have any effect on relative wages at all. r International migration can increase unemployment in the host countries and reduce unemployment in the source countries, according to our simulation of the closed economy model. Again, the empirical evidence suggests that the impact of migration on the unemployment rates and displacement risks of natives is small. Thus, fears that international migration affects the wages and employment opportunities of natives are largely exaggerated. r Under reasonable assumptions about the unemployment rates of foreigners, host countries win more from migration if the migrant population is relatively high-skilled, since less-skilled migrants are more affected by unemployment and, as a consequence, receive higher welfare benefits than high-skilled foreigners. Note, however, that in the case of a full-employment economy, host countries would win more from low-skilled migrants, since they complement domestic production factors. However, without going into detail about different immigration policies, we note that different states within the EU act very differently with respect to the labor market performance of immigrants. r The traditional brain-drain argument is not as important as it appeared to be when discussed in the older theoretical literature, due to the important role of emigrants’ remittances. According to the IMF,

Economic consequences of immigration in Europe

141

financial transfers by migrants residing permanently in foreign countries are the second largest source of income to developing and newly industrialized countries, after foreign direct investment (FDI). In 2001, emigrants’ remittances worldwide exceeded the total global volume of private FDI and public development aid by approximately 40 percent. r Even if immigration creates gains for the host country, those gains will not be distributed equally, and the acceptance of immigration will be different in different subgroups in the host country. Using a unique analysis of the determinants of immigration acceptance (within Germany), we show that acceptance is in fact closely related to the perception of personal winning or losing due to immigration. r There is considerable variation in the economic performance of immigrant populations in different EU countries. This can be assumed to be linked to the variety of conditions of entry to the EU states, which strongly influence the distribution of socioeconomic characteristics of the immigrant population, as well as to differences in country-specific strategies to promote the integration of the existing immigrant population. r Although migrants are more than proportionally affected by unemployment and social assistance, they tend to contribute on average more to public finances and social security systems than they receive. Moreover, increasing the population through migration increases the number of future taxpayers who will contribute to the financing of public goods. Immigration thus reduces the individual debt burden in future generations. Yet the question remains whether international migration will be large enough to mitigate the demographic pressure on public finances and social security systems in Europe. Altogether, host countries are likely to win from international migration – even in the case of “permanent” unemployment in the host country. And migration can aggravate demographic pressures on the welfare state in the source countries. Recall that the populations in central and eastern Europe are ageing at almost the same speed as in western Europe. Thus, the source countries of European migration may lose in terms of aggregate welfare and public finances when their demographic ageing is accelerated by emigration. Restricting international migration is not an efficient answer to these problems. The source countries of European migration have a cost advantage in reproducing human labor and developing human resources, while the productivity of human resources is higher in the host countries. Thus, compensating the source countries for public investment in human capital

142

Immigration and the Transformation of Europe

and fiscal losses caused by international migration is better policy in economic terms. Enlarging the EU can be interpreted as such a measure. However, the short-term losers from immigration in the host countries, such as non-skilled workers, should be compensated as well. Even though the positive effects of international trade are larger than the negative effects of increased competition, the gains from immigration are distributed unequally on the labor market. Cheaper maids, for example, are not employed by low-income households, and better restaurants have no positive impact on low-income people who seldom dine out. Thus the benefits of immigration must be promoted at all levels of society, particularly through educational measures. More education not only helps young people to earn an economic benefit from immigration, but better education can foster open-mindedness and a climate of cultural diversity that in turn promotes immigration.

 Bauer, Thomas 1997 “Do Immigrants Reduce Native Wages? Evidence from Germany.” Munchner ¨ Wirtschaftswissenschaftliche Beitr¨age 97. Munich: University of Munich. Bauer, Thomas and Klaus F. Zimmerman 1997 “Looking South and East, Labor Market Implications of Migration in Europe and Developing Countries,” in Globalization of Labor Markets; Challenges, Adjustment and Policy Response in the European Union and the Less Developed Countries, Olga Memedovic, Arie Kuyvnhoven, and Willem Molle (eds). Dordrecht: Kluwer Academic Publishers, 75–103. Bauer, Thomas, Magnus Lofstrom and Klaus F. Zimmerman 2000 “Immigration Policy, Assimilation of Immigrants, and Natives’ Sentiments towards Immigrants: Evidence from 12 OECD Countries.” Swedish Economic Policy Review 7(2): 11–53. Blau, Francine D. 1984 “The Use of Transfer Payments by Immigrants.” Industrial and Labor Relations Review 37(2): 222. Boeri, T. and H. Brucker ¨ 2005 “Why are Europeans so tough on migrants?” Economic Policy 44: 629–703. Bonin, Holger 2001 “Fiskalische Effekte der Zuwanderung nach Deutschland: Eine Generationenbilanz.” IZA Discussion Paper No 305. Bonn: Institute for the Study of Labor, June. Bonin, Holger, Bernd Raffelhuschen, and Jan Walliser 2000 “Can Immigration Alleviate the Demograhic Burden?” Finanzarchiv 57(1): 1–21. Borjas, George J. 1995 “The Economic Benefits from Immigration.” The Journal of Economic Perspectives 9(2): 3–22. Borjas, George J. 2001 “Does Immigration Grease the Wheels of the Labor Market?” Brookings Institution: Brookings Papers on Economic Activity, 1.

Economic consequences of immigration in Europe

143

Borjas, George J. 2003 “The Labor Demand Curve is Downward Sloping: Reexamining the Impact of Immigration on the Labor Market.” Quarterly Journal of Economics 118(4) (November): 1335–74. Borsch-Supan, ¨ Axel H. 1994 “Migration, Social Security Systems, and Public Finance Source,” in Migration: A Challenge for Europe, Horst Siebert (ed). Tubingen: Mohr, 111–14. Brucker, ¨ Herbert. 2002 “Can International Migration Solve the Problems of European Labor Markets?” Economic Survey of Europe (2): 109–42. Brucker, ¨ Herbert, J. P. Kreyenfeld and M. Schr¨apler 2001 “Trade and Migration: Impact on Wages, Employment and Labor Mobility in Western Germany.” Mimeo. Berlin: DIW. Brucker, ¨ Herbert and Michael Kohlhaas 2004 Migration, Qualification, and Labor Markets – International Immigration to Germany: How much is it Worth? Mimeo. Berlin: DIW. Buchel, ¨ Felix and Joachim R. Frick 2004 “Immigrants in the UK and in West Germany: Relative Income Positions, Income Portfolio, and Redistribution Effects.” Journal of Population Economics 17(3) (August): 553–81. Buchel, ¨ Felix and Joachim R. Frick 2005 “Immigrant’s Economic Performance Across Europe – Does Immigration Policy Matter?” Population Research and Policy Review 24(2): 175–212. Cass, David 1965 “Optimal Growth in an Aggregative Model of Capital Accumulation.” Review of Economic Studies 32(3) (July): 233–40. Coleman, David A. 1992. “Does Europe Need Immigrants? Population and Work Force Projections.” International Migration Review 26(2): 413– 61. Constant, Amelie and Klaus F. Zimmermann 2003 “Circular Movements and Time Away from the Host Country,” IZA Discussion Paper No 960. Bonn: Institute for the Study of Labor. DeNew, John H. and Klaus F. Zimmermann 1994a “Native Wage Impacts of Foreign Labor: A Random Effects Panel Analysis.” Journal of Population Economics 7: 177–92. DeNew, John H. and Klaus F. Zimmermann 1994b “Blue Collar Labor Vulnerability: Wage Impacts of Migration,” in The Economic Consequences of Immigration to Germany, Gunter Steinmann and Ralf E. Ulrich (eds). Heidelberg: Physica, 81–9. Felderer, Bernhard 1994 “Can Immigration Policy Help to Stabilize the Social Security Systems?” in Economic Aspects of International Migration, Herbert Giersch (ed). Berlin: Springer, 197–226. Friedberg, Rachel M. and Jennifer Hunt 1995 “The Impact of Immigrants on Host Country Wages, Employment and Growth.” The Journal of Economic Perspectives 9(2): 23–44. Gang, Ira N. and Francisco L. Rivera-Batiz 1996 “Labor Market Effects of Immigration in the United States and Europe: Substitution vs Complementarity.” Journal of Population Economics 7(2): 157–75. Gang, Ira N. and Francisco L. Rivera-Batiz 1999 “Immigrants and Unemployment in Europe.” IZA Discussion Paper 70. Bonn: Institute for the Study of Labor.

144

Immigration and the Transformation of Europe

Gavasto, A., Venturini, A. and Villosio, C. 1999 “Do Immigrants Compete with Natives?” Labour 13: 603–22. Gross, Dominique 1999 “Three Million Foreigners, Three Million Unemployed? Immigration and the French Labour Market.” International Monetary Fund, Working Paper 99/124. Washington, D.C. Haisken-DeNew, John P. and Joachim R. Frick 2005 “Desktop Companion to the German Socio-Economic Panel Study (GSOEP)” Version 8.0, update to Wave 21, Berlin: DIW. Haisken-DeNew, John P. and Klaus F. Zimmermann 1995 “Wage and Mobility Effects of Trade and Migration.” CEPR Discussion Paper 1318. London. Hanson, Gordon H. and Matthew J. Slaughter 2002 “Labor Market Adjustment in Open Economies: Evidence from US States.” Journal of International Economics 57: 3–29. Harris, John R. and Michael P. Todaro 1970 “Migration, Unemployment and Development: A Two-Sector Analysis.” American Economic Review 60: 126– 42. Heijdra, Ben J., Christian Keuschnigg and Wilhelm Kohler 2002 “Eastern Enlargement of the EU: Jobs, Investment and Welfare in Present Member Countries.” CESifo Working Paper No 718. Munich: CESifo, May. Hofer, Helmut and Peter Huber 1999 “Eastward Enlargement, Trade and Migration: Impact on the Austrian Labor Market.” Vienna: Institute of Advanced Studies (IAS). Hunt, Jennifer 1992 “The Impact of the 1962 Repatriates from Algeria on the French Labor Market.” Industrial and Labor Relations Review 45 (April): 556–72. International Monetary Fund (IMF) 2003 “Balance of Payments Yearbook – World Development Indicators,” in Global Development Finance: Striving for Stability in Development Finance, World Bank (ed). Washington, DC: World Bank. Internationale Organisation fur ¨ Migration (IOM) 2004 “Die wirtschaftliche Bedeutung der Ruck ¨ uberweisungen ¨ von Migranten,” in Gutachten im Auftrag des Sachverst¨andigenrates f¨ur Zuwanderung und Integration. Berlin: Internationale Organisation fur ¨ Migration (IOM), 53–9. Keuschnigg, Christian and Wilhelm Kohler 2002 “Eastern Enlargement of the EU: How Much is it Worth for Austria?” Review of International Economics 10(2): 324–42. Kohler, Wilhelm 2003 “Eastern Enlargement of the EU: A Comprehensive Welfare Assessment.” Department of Economics Working Paper No 0316. Linz: Johannes Kepler University, December. Koopmans, Tjalling C. 1965 “On the Concept of Optimal Economic Growth,” in Pontifical Academy of Sciences, The Econometric Approach to Development Planning. Amsterdam: North-Holland, 255–87. Migration News 2004 11(1). Davis, CA: University of California. Muhleisen, ¨ Martin and Klaus F. Zimmerman 1994 “A Panel Analysis of Job Changes and Unemployment.” European Economic Review 38 (April): 793– 801.

Economic consequences of immigration in Europe

145

Pischke, Jorn-Steffen ¨ and Johannes Velling 1997 “Employment Effects of Immigration to Germany: An Analysis Based on Local Labor Markets.” The Review of Economics and Statistics 79(4): 594–604. Ramsey, Frank P. 1928 “A Mathematical Theory of Saving.” Economic Journal (38) (December). 543–59. Riphahn, Regine 1998 “Immigrant Participation in the German Welfare Program.” Finanzarchiv 55: 163–185. Simon, Julian L. 1984 “Immigrants, Taxes and Welfare in the United States.” Population and Development Review 10 (March): 55–69. Simon, Julian L. 1994 “The Economic Consequences of Immigration: Lesson for Immigration Policies,” in Economic Aspects of International Migration, Herbert Giersch (ed). Berlin: Springer, 227–48. Simon, J. 2004 “The Economic Consequences of Immigration: Lesson for Immigration Policies,” in Economic Aspects of International Migration, Herbert Giersch (ed). Berlin: Springer-Verlag, 227–48. SOEP 1998 German Socio-Economic Panel Study 1998. Berlin: DIW. SOEP 2000 German Socio-Economic Panel Study 2000. Berlin: DIW. Solow, R M. 1956 “A Contribution to the Theory of Economic Growth.” Quarterly Journal of Economics 70: 65–94. Storesletten, Kjetil 2000 “Sustaining Fiscal Policy through Immigration.” Journal of Political Economy 108(2): 300–23. Straubhaar, Thomas and Klaus F Zimmermann 1993 “Towards a European Migration Policy.” Population Research and Policy Review 12(3): 225– 41. Taylor, Marcia 1993 British Household Panel Survey User Manual: Introduction, Technical Report and Appendices. Vol. A. London: Economic and Social Research Council (ESRC). Trabold, H. and P. Trubswetter ¨ 2003 “Sch¨atzung der Besch¨aftigungs- und Lohneffekte der Zuwanderung,” in Migration: Potential und Effekte f¨ur den deutschen Arbeitsmarkt, H. Brucker, ¨ H. Trabold, P. Trubswetter ¨ and C. Weise (eds). Baden Baden: Nomos, 101–37. Trefler, Daniel 1997 “Immigrants and Natives in General Equilibrium Trade Models.” NBER Working Paper No 6209. Cambridge, MA: NBER, October. UN Population Division 2004 “World Population Reports – Population Database.” Accessed 16 April 5 AD Available from http://esa.un.org/unpp/ United Nations 2000 Replacement Migration: Is It a Solution to Declining and Aging Populations? New York: United Nations. Venables, A.J. 1999 “Trade Liberalization and Factor Mobility: An Overview,” in Migration: The Controversies and the Evidence, Riccardo Faini, Jaimi De Melo and Klaus F Zimmermann (eds). Cambridge: Cambridge University Press, 23–48. Venturini, Alessandra and Claudia Villosio 2002 “Are Immigrants Competing with Natives in the Italian Labor Market? The Employment Effect.” IZA Discussion Paper No 467. Bonn: IZA April. Wagner, G. G., R. V. Burkhauser and F. Behringer 1993 “The English Language Public Use File of the German Socio-Economic Panel.” Journal of Human Resources 28(2) (Spring): 429–33.

146

Immigration and the Transformation of Europe

Winkelmann, R. and Klaus F. Zimmerman 1993 “Aging, Migration, and Labor Mobility,” in Labor Markets in an Aging Europe, Paul Johnson and Klaus F. Zimmerman (eds). Cambridge: Cambridge University Press, 255–83. Winter-Ebmer, R. and Josef Zweimuller ¨ 1994 “Do Immigrants Displace Native Workers? The Austrian Experience.” CEPR Discussion Paper 991. London: Centre for Economic Policy Research. Winter-Ebmer, Rudolf and Klaus F. Zimmermann 1999 “East–West Trade and Migration: the Austro-German Case,” in Migration: The Controversies and the Evidence, Riccardo Faini, Jaimi De Melo and Klaus F. Zimmermann (eds). Cambridge: Cambridge University Press, 296–328. Wirtz, Christine and Lene Mejer 2002 “European Data Watch.” Schmollers Jahrbuch 122(1): 143–54. Wong, Kar-yiu 1995 International Trade in Goods and Factor Mobility. Cambridge, MA: MIT Press.

6

Occupational status of immigrants in cross-national perspective: A multilevel analysis of seventeen Western societies1 Frank van Tubergen Introduction

It is well documented that micro-level factors like human capital and demographic position are important for the incorporation of immigrants in the labor market. Immigrants with a higher education, who are more proficient in the destination language, who have more work experience, and who remained longer in the destination country, generally have a better position in the labor market (McAllister 1995; Poston 1994; Raijman and Semyonov 1995). Recently macro factors have also received increasing attention from students of the economic integration of immigrants (Model and Lin 2002). One research tradition has shown that, even after taking into account human-capital variables, the country of origin affects immigrants’ economic position (Borjas 1999; Jasso and Rosenzweig 1990; Poston 1994; Raijman and Semyonov 1995). Another line of research suggests that the attainment of immigrant groups also differs between destination countries (Borjas 1988; Model 1997; Model, Fisher, and Silberman 1999; Model and Lapido 1996; Reitz 1998, 2003). Van Tubergen, Maas, and Flap (2004) combined both macro approaches and suggested that three groups of macro effects strongly affect the economic attainment of immigrants. First, there exists what they call “origin effects,” the impact of countries of origin irrespective of the destination of immigrants. Second, they outline “destination effects,” the impact of receiving societies, notwithstanding immigrants’ origins. Third, there is an influence of the combination of origin and destination, which they called “setting” or “community” effects. Van Tubergen et al. (2004) studied the role of these groups of macro factors in the

1

This paper was written in part while I was an academic visitor at Nuffield College, University of Oxford (UK), February–March 2004. The visit was funded by a grant from the Netherlands Organization for Scientific Research. I would like to thank Anthony Heath, Henk Flap and Ineke Maas for helpful comments on earlier versions of this paper.

147

148

Immigration and the Transformation of Europe

labor-force participation and employment of immigrants in eighteen Western countries. This chapter builds on this earlier collaborative study in two ways. First and foremost, I extend their study by examining another sociologically relevant aspect of immigrants’ economic incorporation: occupational status. Participating in the labor market and having a job are only the first steps towards full incorporation in the labor market. A logical follow up question is to study the status of the jobs that immigrants occupy. From a theoretical perspective, it is tempting to examine if the empirical findings of van Tubergen et al. on labor-force participation and unemployment equally apply to occupational status. Second, I formulate new hypotheses. In addition to the macro-level variables characteristics considered in Van Tubergen et al., I examine the effects of racial composition and the proportion of the immigrant group that is employed.

Theories and hypotheses I use two kinds of explanations to understand the influence of the country of origin, the country of destination, and the immigrant setting on the economic incorporation of immigrants. The first idea is that of human capital (Borjas 1987; Chiswick 1978), which argues that macro differences can be explained in terms of composition effects. This view holds that economic incorporation is affected by human capital that is not equally distributed across macro units. Another line of reasoning suggests that social contexts can be important. Contextual effects occur when characteristics of macro units have a direct effect on individuals’ outcomes, over and above the effects of individual characteristics. These contextual effects are linked to processes of discrimination that foster or hinder the economic mobility of immigrants (Model and Lapido 1996; Portes and Rumbaut 1996, 2001).

Composition effects The idea that macro differences in the economic attainment of immigrants can be interpreted in terms of composition effects has been worked out most notably by the economists Chiswick (1978) and Borjas (1987, 1988). They argue that human capital factors like education, work experience, language, and individual talents determine immigrants’ economic attainment, and that systematic group differences in the composition of these skills explain macro differences. Selection can either be positive (or favorable), selecting immigrants with high (un)observed human capital,

Occupational status of immigrants

149

or negative (unfavorable), selecting immigrants with lower productivity and skills. One potentially important factor in the selection of immigrants is associated with political conditions in the sending country. Chiswick (1978, 1979, 1999) suggested that political suppression and instability in the origin country may induce people to migrate for other than pure economic reasons. Hence, Chiswick argued that people who move from politically unstable societies are less favorably selected than economic migrants, which in turn results in a lower occupational status (H1). Immigration policy is another determinant of skill selection. Borjas (1988) hypothesized that immigrants in countries with a strict immigration policy, such as Australia and Canada, are more favorably selected. In these countries migrants who apply for an entry visa have to meet specific requirements. This so-called “point-system” rates migrants according to their skills and additionally selects those who are thought to be more of use to fulfil labor market shortages (Borjas 1988; Reitz 1998). Immigrants who pass are assumed to be more favorably selected than those who fail; these immigrants will perform better in the labor market than immigrants in countries without such an entrance test. In view of these ideas, it is predicted that immigrants have a higher occupational status in countries that use a point system than they have in other countries (H2). Borjas (1988) has suggested that the selection of immigrants is determined by the income inequality in the countries of origin and destination. In societies that have a very skewed income distribution, emigration tends to become concentrated among the less talented, who have much to gain by migrating. On the other hand, emigration from societies with more equally distributed incomes tends to be largely concentrated at the upper end of the home country’s income distribution. Similarly, host countries with high-income inequality attract migrants with high abilities and talents. The more dispersed the income inequality in the home country relative to that of the destination country, the lower will be the occupational standing of immigrants (H3). Skill selection of immigrants is also associated with the level of economic development in the origin and destination country (Borjas 1987, 1988; Chiswick 1978, 1979; Jasso and Rosenzweig 1990). All things being equal, migrants from developing countries have a lower average education than those originating from more advanced economies. In addition, migrants from rich countries have an edge over migrants from poorer countries regarding the transferability of their skills and subsequent rewards. Educational diplomas obtained in developing nations are more difficult to transfer to economically advanced nations than

150

Immigration and the Transformation of Europe

diplomas obtained in equally or more advanced origin nations. Therefore, it is hypothesized that the less economically advanced the origin country is relative to the destination country, the lower the occupational status of immigrants (H4). Geographic distance between origin and destination countries could also affect skill selection. The literature generally assumes that greater distance increases migration costs (Borjas 1987; Jasso and Rosenzweig 1990) and diminishes the likelihood of return migration (Borjas 1987). As a consequence, the more talented persons are overrepresented among migrants who move over long distances (Blau and Duncan 1967) and these immigrants also have more incentives to invest in human capital, such as acquiring the destination language (Chiswick and Miller 2001). Thus we might expect that greater geographic distance between the origin and destination countries has a positive impact on immigrants’ occupational status (H5). A final property that may affect the skill selection of immigrants is the exposure to the language of the destination country in the country of origin. Language skills differ between immigrant groups (van Tubergen and Kalmijn 2005) and proficiency in the destination language has an important impact on immigrants’ economic attainment (Chiswick and Miller 2002; Shields and Price 2002). Immigrants who have been exposed to the destination language in their country of origin would be expected to have more skill in the destination language than immigrants who have not been exposed to the host language before migration. This leads to the prediction that immigrants will have a higher occupational status given related official languages of origin and destination than when languages are different (H6). Context effects A second line of thought argues that the position immigrants obtain in the labor market depends on societal conditions that are linked to processes of discrimination (Model and Lapido 1996; Portes and Rumbaut 1996). Although there is no overarching discrimination theory, the broad range of hypotheses within this framework share their emphasis on societal conditions that produce in-group preferences and out-group prejudices. It is argued that such preferences and prejudices cause positive and negative discrimination, such as better jobs being offered to members of the in-group or outright refusal to employ a member of the out-group. According to this research tradition, religious and racial characteristics of the country of origin are important. Research has shown that natives’ social distance towards ethnic groups overlaps with a distinction

Occupational status of immigrants

151

in race and religion, ranking non-whites, Islamic, Buddhist, and other non-Christian groups at the top of the social distance scale (Owen, Eisner, and McFaul 1981; Pineo 1977). Because the Western countries examined in this study are predominantly white and Christian, I predict that migrants from predominantly non-Christian origins have a lower economic status than those of Christian nations (H7), and non-white groups have a lower status than white groups (H8). Receiving nations provide an important social context for immigrants’ occupational position. It has been suggested that the election of socialdemocratic parties in the government (in contrast to liberal, conservative, and Christian-democratic parties) would lead to less economic inequality in a country (Lenski, Lenski, and Nolan 1991), including inequalities between immigrants and natives (van Tubergen et al. 2004). Thus we can hypothesize that immigrants in countries with a stronger presence of social-democratic parties in the government have a higher occupational status (H9). With respect to the immigrant community, the relative size of an immigrant group can be relevant in a variety of ways. First, it is often suggested that prejudice increases with the relative size of the immigrant group. Sizable groups are more visible and more likely to be perceived as a potential threat to the native population in terms of political and economic power (Blalock 1967; Quillian 1995). This ‘ethnic-threat’ hypothesis predicts a negative effect of the relative size of an immigrant group in a host society on the occupational status of its members in that society (H10). Alternatively, it is argued that immigrants in sizable groups perform better economically (Portes and Bach 1985; Wilson and Portes 1980; Zhou and Logan 1989). Under the assumption that immigrants are more willing to help co-ethnics – in the same way that natives prefer in-group members – immigrants could profit from the presence of country-fellows, sharing their “ethnic capital” (Borjas 1992; and chapter 3 by Pedersen et al. in this volume). Members from the same immigrant group help each other by offering jobs, buying goods, and lending money. Therefore, it is predicted that the relative size of an immigrant group has a positive effect on the occupational status of the members of that group (H11). Because the relative size of an immigrant group in a country only refers to the number of people of the own group that could be of help, the resources they could generate should be considered as well. Migrants who belong to economically powerful groups could benefit more from their economic potential than migrants from disadvantaged groups. To examine this idea, I use two indicators of economic resources of the immigrant group that are independent of occupational status: the mean educational level of the group (Borjas 1992), and the percentage of the labor

152

Immigration and the Transformation of Europe

force population of the group that is employed. It is hypothesized that immigrants who belong to groups with a higher average education have a higher occupational status (H12), and groups with a larger proportion employed population will have a higher status (H13). Data and methods Data I collected and standardized existing surveys containing individual-level information on the economic position of immigrants. The surveys were pooled in a single cross-national data set: the International File of Immigration Surveys (van Tubergen 2004). The core of the data file consists of 121 labor-force surveys conducted between 1991 and 2002 in fourteen countries of the European Union(EU): Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Luxembourg, The Netherlands, Portugal, Spain, Sweden, and the United Kingdom. These surveys were collected and standardized by Eurostat (2003), the statistical office of the EU, into the so-called European Union Labour Force Survey (EULFS). Because Eurostat takes great care in standardizing the laborforce surveys, the designs of these surveys became much alike in the 1990s, which reduced problems of comparability (Eurostat 1998). In addition to the EULFS data set, I collected comparable surveys for Australia, Canada, and the United States (van Tubergen 2004). For Australia, I relied on a specific immigrant survey conducted in 1988 (Australian Office of Multicultural Affairs 1988). For Canada, I used the 1991 and 1996, 3 percent public-use census (Statistics Canada 1991, 1996). For the United States, I used the 1980 and 1990 census, 1 percent files (United States Census Bureau 1980, 1990). The analysis is restricted to first-generation immigrants, defined as those born outside the country of residence. For Canada and the United States the census samples were much larger than the samples of the new immigrant countries, so I restricted the number of respondents in large immigrant groups to a maximum of 2,000 per survey. I analyzed the economically employed population between the ages of twenty-five and fiftyfour, thus excluding inactive and unemployed immigrants. The analysis includes females as well as males, which is a significant contribution to the research literature, because earlier research on economic attainment of immigrants was largely restricted to males (Cobb-Clark 1993; Schoeni 1998). All in all, the cross-national data-file consists of seventeen destination countries, 181 origin groups, 859 combinations of origins and destinations (i.e. settings), and 239,619 immigrants.

Occupational status of immigrants

153

Dependent and independent variables The dependent variable is the status of the current job, as measured in terms of the International Socio-Economic Index (ISEI) (Ganzeboom, De Graaf, Treiman 1992). To obtain ISEI scores for the occupations, I used tools that convert the ISCO-88 classification into ISEI (Ganzeboom and Treiman 1996, 2003). The EULFS provided three digit ISCO-88 scores. The original codings of occupations in the United States census and the Australian survey were translated into four digit ISCO-88 using conversion tools (Ganzeboom 2003; Lambert 2003). For the fourteen occupational categories provided in the Census of Canada, I relied on the weighted ISEI scores provided by Model and Lin (2002).2 The independent variables are related to origins, destinations, settings, and individuals. Political suppression I used information collected by Freedom House (Karatnycky and Piano 2002) on political rights and civil liberties in the countries of origin. Political rights varied from one (e.g. free and fair elections, power for opposition parties, etc.) to seven (e.g. oppressive regime, civil war). Civil liberties varied from one (e.g. freedom of expression and religion, free economic activity) to seven (e.g. no religious freedom, political terror, no free association). I used the sum score for each country (two to fourteen) and computed averages for the 1972–80 period. Christian origin I included a dummy variable for origin countries that have a predominant Christian population, using predominantly nonChristian countries as a reference. Those countries with more than 50 percent Christian adherents in the 1960–80 period were assumed to be predominantly Christian. This information was obtained from Brierley (1997). White origin I relied on the racial self-identification question in the 1990 census of the United States to obtain figures on the racial composition of countries. I used the proportion identifying as whites of all immigrants from a certain country as a measure of the proportion of whites in that country. Point system I set up a dummy to indicate whether destinations had a point system or not. Australia and Canada have had such immigration policies since the 1960s (Borjas 1988), so I did not vary this dummy over time. 2

See the Appendix in Wanner (1998) for the specific ISEI scores assigned to the categories.

154

Immigration and the Transformation of Europe

Presence of social-democratic parties in the government I counted the number of years in which social-democratic parties were present in the government during the ten years preceding the survey year. The annual presence of social-democratic parties in the government was rated as one when they formed a one-party government, 0.5 when they joined a coalition, and zero when they were absent from the government. Information on the presence of social-democratic parties in the government was obtained from various Internet sources. Income inequality (ratio) This variable measures the income inequality of the origin country relative to that of the destination country. To measure the income inequality within a country, I used the Gini formulae. Information was obtained from a publication of the World Bank (2001), which gives Gini scores per country in the 1990s. Economic development (ratio) I used gross domestic product (GDP) per capita as a measure of economic development and calculated GDP ratios for the origin country relative to the destination country. GDP was measured in constant dollars per capita for 1980 and was obtained from OECD (2000). Geographic distance Geographic distance between origin and destination was measured as the distance in kilometres between the capital cities of the origin and destination countries. Calculations were based on the so-called “great circle distance method” (Byers 2003). Official language I use a dummy to indicate whether the official language of the origin country was the same as the official language of the destination country, based on the language situation at the end of the twentieth century (Grimes 2000). An official language is the language used in schools and formal settings. I further aggregated individual-level information on immigrants included in the surveys of our data set to obtain information on other setting or group-specific variables. ISEI natives I included aggregate variables that controlled for differences in labor-market opportunities between countries and between time periods. I computed the annual ISEI of native males and females between twenty-five and fifty-five years old. Relative group size I constructed a variable for the size of an immigrant group relative to the total population of the host country. I estimated averages for the 1980–90 period.

Occupational status of immigrants

155

Mean level of education group I used a three-category classification of education ranging from low to high (see below), and computed the mean education of immigrant groups. Employment group This variable indicates the fraction of the group that is employed in the labor market of all economically active members of that group (i.e. both employed and unemployed) who are between twenty-five and fifty-four years old.3 I also included individual-level control variables. Some surveys contain precise information on all relevant variables, whereas other surveys have cruder measures or do not contain some variables. The European Union Labour Force Survey (EULFS), for example, does not provide precise information on duration of residence and schooling and has no information on language skills. Therefore, I had to make some concessions to make the variables cross-nationally comparable. Age Age was measured in years and by using midpoints for surveys using broader age categories. Duration of residence I constructed three categories: zero – five years, six – ten years, and eleven years or more. Education In accordance with the classification in the EULFS I used three categories for education: low (primary education and first stage of secondary education); middle (second stage of secondary education); and high (higher education). Surveys using measures of schooling (years of full-time education) were recoded using information on the years needed to obtain certain educational levels (OECD 1999). Marital status I contrast married with all others. Analysis and models I made use of random intercept models with two levels. At the “lowest,” or micro, level, occupational status is affected by individual characteristics, such as education and duration of residence in the host society. 3

A potential problem of contextual analysis is the high correlation between macro-level variables. Bivariate Pearson correlations at the setting level show that correlations are generally not higher than .40. Exceptions are the negative associations between political suppression in the country of origin and the relative economic development (r = −.59) and Christian origin (r = −.47). Another moderately strong association exists between geographic distance and percent white (r = −.43). In summary, there is no a priori reason to doubt the results on grounds of multi-collinearity.

156

Immigration and the Transformation of Europe

At the macro level, immigrants’ economic status is an outcome of their origin, destination, and setting. These macro-level components affect economic attainment at the same level, so the multilevel structure is non-hierarchical. I therefore relied on so-called “cross-classified” models (Raudenbush and Bryk 2002; Snijders and Bosker 1999), in which at the higher level origins and destinations constitute “parallel” levels. Because the variance of settings is tapped by the variance of origins and the variance of destinations, it is not independently assessed; however, setting effects are estimated at the appropriate origin-by-destination level. It is important to emphasize that I made destination countries time variant in the analysis. This resulted in more than a hundred “destinationyear” cases as the destination component instead of only seventeen destination countries. This design includes precise macro-level control variables that measure the average occupational status among native males and females for each destination country in each survey year. In addition, I have a better estimate of the time-varying predictor in the analysis – the presence of left-wing parties ten years preceding the survey year. A drawback, however, is that the standard error of the time invariant destination variable (i.e. point system) is underestimated in this design. The reason is that the within-nation observations of surveys are not independent from one another. I will therefore also analyze models in which destinations are treated as time invariant. I made use of Markov Chain Monte Carlo (MCMC) estimation procedures, provided in the software program MlwiN (Browne 2002). Results Descriptive analyses To give a descriptive account of macro effects, Table 6.1 (males) and Table 6.2 (females) present the mean occupational status (in ISEI) of immigrants by origin, destination, and setting. Because such information could, of course, not be presented for the more than 800 settings included in the data set, I illustrate how seven origin groups fare in the seventeen destination countries. Table 6.1 shows that the mean occupational status (in ISEI scores) of all immigrants, averaged over the seventeen destination countries, is 43.1. Immigrants from the United States have an average ISEI of 53.2, which is ten points above the mean ISEI of immigrants from all countries, and almost twenty points higher than the average ISEI of immigrants from Morocco (35.0) and Turkey (35.2). There are also pronounced differences in immigrants’ occupational standing between destination countries. I find a low occupational status among immigrants in Austria

. 40.8a 52.7a 47.2 . . 46.5a . . . 39.3a 46.9a . 40.5a 66.8a 50.6 47.8 47.4

Australia Austria Belgium Canada Denmark Finland France Germany Greece Ireland Luxembourg Netherlands Portugal Spain Sweden UK USA Mean

. = N < 25. a = 25 < N < 200.

China

Country of destination Italy 44.0a 48.1a 37.2 41.3 . . 39.1 34.8 49.4a 48.1a 42.0 41.6 . 43.8a . 45.5 39.6 40.3

Germany

46.9a 51.9 49.0 45.7 48.8a 59.3a 51.4 na 41.1 51.9 52.9 49.1 44.4a 50.3 49.8 50.5 48.1 49.1 . . 35.9 . . . 34.0 29.7a . . 53.2a 36.8 . 28.2 33.9a 38.4a 48.7a 35.0

Morocco . 39.1 44.5a 41.2 45.3a . 39.6a 40.5 32.8 . 46.4a 42.0a . 35.2a 50.2a 50.6a 41.4 40.5

Poland . 32.9 34.9 . 35.9a . 31.8 32.0 43.0 . 45.1 36.4 . . 37.3a 44.8a 49.4 35.2

Turkey

Country of origin

. 51.2 60.5 50.7 53.3 . 62.7 46.9 51.8 50.9 59.7 53.0 . 58.2 53.6 56.2 na 53.2

USA 40.1 37.8 42.6 43.6 44.9 46.9 37.2 37.8 37.8 46.1 42.7 44.8 43.3 39.5 41.6 50.1 45.1 43.1

Mean all groups

47.3 43.1 45.0 43.1 45.0 44.3 42.9 43.8 41.5 42.1 44.7 49.0 39.8 40.5 44.7 46.9 44.3 44.2

Mean natives

Table 6.1 Occupational status (ISEI) of male immigrants by country of origin, country of destination, and setting

. 35.9a 46.9a 42.7 . . 40.7a . . . 38.9a 42.1a . 35.0a 50.6a 49.5 44.2

43.7

Australia Austria Belgium Canada Denmark Finland France Germany Greece Ireland Luxembourg Netherlands Portugal Spain Sweden UK USA

Mean

. = N < 25. a = 25 < N < 200.

China

Country of destination

47.9

. 46.7 48.7 47.6 44.4a . 50.1a na 47.8 50.9 49.8 47.2 55.2a 49.4a 47.3a 48.6 46.9

Germany

41.7

. 54.0a 37.6 42.6 . . 41.9a 32.7a 56.7a 51.0a 39.4 47.2a . 53.7a . 45.2 42.4

Italy

32.6

. . 33.9 . . . 28.9 34.8 . . . . . 23.2 . . 49.5a

Morocco

40.5

. 38.9 41.3a 42.7 41.1a . 37.5a 41.7 31.2 . 40.9a 40.2a . 36.7a 42.4 46.7a 41.2

Poland

32.4

. 27.2 32.5 . 28.0a . 33.2a 28.9 45.1 . . 34.8 . . 36.1a 47.6a 50.9a

Turkey

Country of origin

54.1

. 53.0a 57.0a 52.4 52.8a . 60.1a 51.0a 54.5a 53.6 55.8a 54.9a . . 54.4a 55.5 na

USA

42.3

42.3 34.2 44.0 44.7 41.3 45.7 34.1 37.8 37.4 48.7 38.7 44.1 46.6 39.7 41.5 48.3 43.7

Mean all groups

45.6

47.4 42.7 45.6 47.7 42.2 44.9 42.4 43.7 42.7 47.3 45.9 46.8 38.7 42.0 43.7 46.1 47.1

Mean natives

Table 6.2 Occupational status (ISEI) of female immigrants by country of origin, country of destination, and setting

Occupational status of immigrants

159

(average ISEI is 37.8), France (37.2), Greece (37.8), and Germany (37.8), whereas the occupational status is considerably higher in the United Kingdom (50.1). Table 6.1 also provides some clues for the impact of settings. Compare, for instance, the occupational status of Turks in Germany and the United States. In Germany, Turks have an average ISEI of 32.0, which is below the mean ISEI of Turks in general (35.2) and below the mean of all immigrants in Germany (37.8). In contrast, in the United States, Turks have a much higher occupational status (49.4), which is far above their general rate and also above the overall pattern observed among immigrants in the United States (45.1). Apparently, then, the specific situation of Turks in Germany and the United States determines their deviance from the pattern expected from general origin and destination effects. Hypotheses testing I constructed three cross-classified multilevel regression models of immigrants’ occupational status to test the hypotheses. Model 1 includes the macro-composition factors plus a macro-control variable for the occupational status of natives. In Model 2, individual controls are included. Model 3 adds the contextual factors. The results are presented separately for males and females (Table 6.3). Note that in these models, I use destinations per year as the destination level.4 Composition effects To examine the results for the hypotheses on composition effects, we have to look at Models 1 and 2. Because Model 1 contains only macro-composition factors (and a macro control), it is logical to examine the results of this model for the discussion of composition effects. Adding human capital factors in Model 2 normally reduces the strength of composition effects. However, by no means should the macro effects in the present study disappear. This is because several “observable” skills were not measured (e.g. labor-force experience) or only partly measured (e.g. length of residence) and “unobserved” skills (i.e. talents, productivity, ambitions) were, of course, completely omitted. Moreover, some macro-composition factors may be positively (or negatively) selective, but their effects may be suppressed due to associations with other selection mechanisms that are negatively (or positively) selective. Taking individual-level factors into account could therefore show an increase in some macro-level effects. 4

In additional analyses (available from the author) I examined how sensitive the findings were to different multilevel designs, i.e. treating destinations as time-constant instead of looking at destinations by year. The overall conclusion is that of a strong similarity between the original and the additional analyses, suggesting that the findings are robust for differences in multilevel designs.

Constant Origin Political suppression Predominantly Christian Origin White (%) Destination Point system Mean occupational status, native reference group Social-Democratic parties in government Setting Gini origin/Gini destination GDP origin/GDP destination Geographic distance (per 1,000 km) Official language Relative group size (%) Mean educational level group Employment rate group (%)

9.008 −.259∗∗

−5.965∗∗ .617∗∗

−3.610∗∗ .659∗∗ .391∗∗ .341∗

−.222

−4.137∗∗ .753∗∗

−3.644∗∗ .942∗∗ .521∗∗ .085

Model 2

14.518

Model 1

Males

−2.094∗∗ .420 .289∗∗ .699∗∗ −.284∗∗ 6.366∗∗ .027∗∗

.117

−6.621∗∗ .585∗∗

−5.181∗∗ 3.381∗∗ .426∗∗ .726∗∗

−.765 .610∗∗

−.281∗

20.051

−6.900 −.233∗∗ −.929 .021∗∗

Model 1

Model 3

−3.703∗∗ 2.363∗∗ .357∗∗ .950∗∗

−2.832∗∗ .436∗∗

−.187∗

17.233

Model 2

Females

Table 6.3 Cross-classified multilevel regression of socioeconomic status (ISEI) in seventeen Western countries, 1980–2002, immigrants between 25 and 54 years old∗

−2.046∗ 1.968∗∗ .240∗∗ 1.433∗∗ −.687∗∗ 6.422∗∗ −.010

−.101

−4.359∗ .379∗∗

−.129 −.230 .021∗∗

6.392

Model 3

p < .05; ∗∗ p < .01 (two-tailed tests).

126 181 859 138,472

Number of observations Destination year Origin Setting Individual

∗

Ref. 4.621∗∗ 20.756∗∗ .450∗∗

Ref. 4.973∗∗ 21.501∗∗ .374∗∗ 126 181 859 138,472

Ref. −.097 .930∗∗

Ref. −.207 .602∗∗

126 181 859 138,472

.078∗∗

.086∗∗

Individual Age Duration of stay 0–5 years 6–10 years 10+ years Education Low Middle High Married 126 170 809 100,697

126 170 809 100,697

Ref. 7.634∗∗ 20.893∗∗ −.618∗∗

.615∗∗ 2.262∗∗

Ref.

−.042∗∗

126 170 809 100,697

Ref. 7.157∗∗ 20.098∗∗ −.569∗∗

.731∗∗ 2.613∗∗

Ref.

−.050∗∗

162

Immigration and the Transformation of Europe

The first factor that relates to the skill composition of immigrants is the political suppression in the sending nation. According to hypothesis one, people from more suppressive societies have a lower occupational standing. The results are significant for males in Model 2, and for females in Models 1 and 2. These findings support the hypothesis and are in line with the cross-national analysis of van Tubergen et al. (2004), who found a negative relationship between political suppression and the likelihood of immigrants’ labor-force activity and employment. It was hypothesized that immigrants in countries with a point system should have a higher occupational status (H2). The analysis shows that the opposite is true: immigrants in countries having a point system have a lower occupational status than immigrants in countries without such policies. Earlier studies showed that immigrants in Australia and Canada have higher earnings relative to the native population than immigrants in the United States (Borjas 1988; Reitz 1998), but van Tubergen et al. (2004) found no positive impact of the point system. The third group of macro effects pertains to settings. It was hypothesized that those who migrated from countries with a more unequal income distribution to countries with less inequality have a lower occupational position than immigrants who moved in the opposite direction (H3). The analysis confirms this prediction, both for the male and female samples. The effect of the ratio of the Gini score of the country of origin to the Gini score of the country of destination is significantly negative in Model 1. It becomes less strong after individual controls are included, but remains significant. These findings are in line with Borjas’ (1988) analysis of immigrant earnings in Australia, Canada, and the United States, and with the analysis of immigrants’ employment status by van Tubergen et al. (2004). Studies of immigrants’ earnings in the United States have found no significant effect of the inequality ratio for males (Borjas 1987), whereas the predicted negative relationship was supported for females (Cobb-Clark 1993). Hypothesis four stated that people who moved from less economically advanced nations to more advanced nations have a lower position in the labor market than immigrants who moved from relatively rich countries. This idea is confirmed. For both males and females, Model 1 shows a significantly positive effect of the GDP ratio on occupational status. In Model 2, the GDP ratio is still significant, but the coefficient has become somewhat smaller. This suggests that part of the relationship is due to favorable selection of educational diplomas, whereas the remaining effect underscores the idea that immigrants’ human capital obtained in advanced economies is valued more in the receiving countries than skills obtained in more developing nations. In line with these findings,

Occupational status of immigrants

163

earlier studies in the United States showed that the per capita income in the origin country has a positive effect on the occupational earnings and wages of male immigrants (Borjas 1987; Jasso and Rosenzweig 1990) and female immigrants (Cobb-Clark 1993). These findings are also in line with Borjas’ (1988) analysis of immigrant earnings in Australia, Canada, and the United States, but do not agree with the cross-national study of van Tubergen et al. (2004). They found that the GDP ratio decreases the likelihood of labor-force activity and employment, for males and females. It was further hypothesized that the geographic distance between the country of origin and the country of destination has a positive impact on immigrants’ occupational status (H5). In accordance with this prediction, Model 1 shows that the distance between the capital cities of the origin and destination countries has a significantly positive impact on the occupational status for males and females. The effect diminishes once individual variables are taken into account, but remains significant. Earlier studies have found mixed results concerning the relationship between geographic distance and economic performance. Borjas (1987) and Cobb-Clark (1993) found no, or even a negative, effect of geographic distance on earnings of immigrants in the United States. By contrast, Jasso and Rosenzweig (1990) found a positive effect of distance on occupational earnings and wages of immigrants in the United States. Van Tubergen et al. (2004) showed that geographic distance decreases the likelihood of labor-force activity, but, once immigrants are in the labor force, increases the chance of employment. A final selection factor of the immigrant setting is whether the official languages of the origin and destination resemble each other. I hypothesized that groups who, in this way, were exposed to the official language of the host country before migration have a higher occupational status (H6). Model 1 indeed shows a significantly higher score for this group among females, but not for males. However, in Model 2, after taking individual factors into account, the results are significant for males as well. The effect in Model 2 is stronger for females, and inspection of Model 3 reveals very significant effects of language exposure for both males and females. This suggests that immigrants who moved to a destination with the same official language tend to be negatively selected on other skill factors – suppressing the positive skill selection in terms of language proficiency. All in all, the analysis confirms the idea that language exposure improves the occupational standing of immigrants. Earlier research in the United States has shown that male immigrants from countries where English is an official language perform better in the labor market (Borjas 1987; Jasso and Rosenzweig 1990). Van Tubergen et al.(2004) found a

164

Immigration and the Transformation of Europe

beneficial effect of language in regard to the analysis of male employment, but could not find a similar effect for labor-force participation and female employment. Context effects To test the context effects we have to look at the findings of Model 3. I hypothesized that people from dominant non-Christian societies have a lower occupational status than those from mainly Christian nations (H7). Surprisingly, the analysis does not show significant differences between immigrants from Christian and non-Christian societies. The cross-national analyses of van Tubergen et al. (2004) showed that immigrants from Christian nations are more active in the labor market and have lower unemployment rates than immigrants from non-Christian nations. Another idea was that predominantly white groups have a higher occupational status (H8). In line with this idea, I find that the percent white in the country of origin has a positive and significant impact on the occupational status of immigrants. I formulated one contextual hypothesis on the impact of receiving nations. I predicted that the longer the presence of social-democratic parties in the government, the higher the occupational status of immigrants (H9). I find no support for this idea, however. The analyses for the male and female samples show that the presence of social-democratic parties in the government in the ten years preceding the survey year had no significant effect on immigrants’ occupational standing. Such a positive effect of social-democratic parties in the government was found in van Tubergen et al. (2004). With regard to the relative size of immigrant groups I formulated opposing ideas. According to one notion, there is a positive relationship between size and occupational status (H10); the alternative idea predicts a negative relationship (H11). For both samples, I find evidence for the latter idea: the larger the size of an immigrant group relative to the total population in the destination country, the lower the occupational status of the members of that group. Findings on the relationship between group size and economic performance have been mixed in studies within a single country (Wilson and Portes 1980; Zhou and Logan 1989). Van Tubergen et al. (2004) observed a significantly positive effect of relative group size on labor-force participation, but this had no effect on employment. Another contextual factor of the immigrant community is the average education of the group members. I hypothesized that a higher level of education among the immigrant group has a positive effect on the occupational status of the members of the group (H12). The findings

Occupational status of immigrants

165

of Model 3 clearly confirm this prediction, for both males and females. It is important to realize that these are true contextual effects, because the education of immigrants at the individual level is controlled for. Van Tubergen et al.(2004) reported a significantly negative effect of the average education of the group on the likelihood of participating in the labor market, and a significantly positive effect on employment. A final contextual characteristic of the immigrant community is the percentage of the group that were employed. I predicted that the higher the percent employed in an immigrant group, the higher the occupational status of the members of that group (H13). I found different results by gender. For male immigrants, the percent employed of a group directly varies with their economic chances. However, the relationship between percent employed and occupational status is not significant among females. Thus, I find some support for hypothesis 13.

Explained variance Analysis of random intercept models without explanatory variables (“empty model”) shows that the variance at the individual level for males is 83 percent of the total variance observed, the variance between origins is 14.5 percent and between destinations only 2.4 percent (figures not presented). Comparing the empty model with the full model for males (i.e. Table 6.3, Model 3) shows that 36 percent of the total variance is explained. More specifically, the final model explained 6 percent of the variance among destinations, 28 percent of the variance among individuals, and 84 percent of the variance among origins. The results for the female sample are very similar. Apparently, then, from a macro perspective, I was better able to explain the variance between origin countries than between destination countries.

Conclusion and discussion In this chapter, I studied macro effects on the occupational status of immigrants. The analysis shows that the country of origin, the country of destination, and the combinations thereof (i.e. settings, communities) all play a role. More precisely, about 17 percent of the total variance of immigrants’ occupational status was observed at the macro level. It appears that the occupational standing of immigrants varies quite strongly among origin groups. I find little variation among receiving nations, which suggests that the occupational status of immigrants is quite similar across nations (though see chapter 9 for qualifications to this broad observation).

166

Immigration and the Transformation of Europe

In order to understand macro differences, hypotheses were derived from human capital theory and notions on discrimination. From the perspective of host countries this study does not confirm the idea that societies aiming to select the more favorable immigrants to their economy succeeded in this goal. On the contrary, in countries with a point system, immigrants have a lower occupational status than similar immigrants in other societies. In addition, I find no support for a contextual effect of receiving nations: the idea that the presence of social-democratic parties in the government has a beneficial impact on the occupational status of immigrants. Van Tubergen et al.(2004) found such a positive effect for immigrants’ labor-force participation and employment. These combined findings suggest that an active state policy towards integrating immigrants succeeds in increasing levels of participation and employment, but not in their socioeconomic standing. This may not seem too surprising, since government policies are more often directed towards changing levels of social welfare, schooling, and positive labor-market discrimination than on the level of immigrants’ occupation once they are active in the labor market. I was better able to understand differences among origin groups. People who moved from suppressive countries have a lower occupational status. These immigrants are less favorably selected, and, in turn, have a lower occupational status than immigrants who move for economic reasons. Further, the percent white of the country of origin has a positive impact on the occupational status. Non-white groups meet stronger discrimination in the labor market than white groups in the predominantly white receiving nations, which results in a lower occupational status. I do not find an effect of religious origin, however. Immigrants from non-Christian origin do not have the expected lower occupational status than immigrants from mainly Christian societies. Apparently, then, social distance and the resulting discrimination is stronger on racial than on religious grounds. A possible explanation for not finding an effect of religious origin is that the discrimination effect is suppressed in terms of unobserved selection. The social costs of migration are lower for those who moved from Christian origin societies to Christian host societies than for those from non-Christian nations. This would imply that immigrants from Christian countries are less talented and less productive than immigrants from non-Christian nations. It also implies that the effect of percentage white in the country of origin found in this study is stronger once (negative) selection is controlled for. The community plays a key role in immigrants’ occupational status. People who moved from societies with an unequal income distribution

Occupational status of immigrants

167

to more egalitarian societies have a lower occupational status than people who moved in the opposite direction. This supports the assumption that the less talented and less productive move from unequal to more egalitarian societies, whereas the opposite is true for the more talented and productive immigrants. I also find that the less economically advanced the country of origin relative to the country of destination, the lower the occupational status of immigrants. People who moved from less economically advanced societies to more advanced societies are less skilled than people who moved from more economically advanced nations. This study also shows that after taking into account educational attainment at the individual level, the relationship becomes weaker, but remains significant. Hence, the beneficial effect of a higher relative economic development is partly due to educational sorting, but also to higher rewards of educational diplomas. Immigrant communities are also important, because they are selective in terms of the travel distance between origin and destination. The greater the distance between the country of origin and the country of destination, the higher the occupational status of immigrants. Travel distance favorably selects immigrants and also increases the benefits of human capital investments after migration – people who travelled over longer distances are less likely to return. A final selection mechanism that plays a role at the setting level is the exposure to the language. Groups that had the same official language in the country of origin and destination, have a higher occupational status. Immigrants in these groups have a better command of the destination language, which is an important determinant of occupational success. Discrimination processes are also important for explaining the role of the immigrant community. I find that the larger the size of the immigrant group relative to the total population in a country, the lower the occupational status of the immigrants of that group. Larger groups are perceived by the native majority as more economically powerful and threatening, and are more discriminated against in the labor market. It opposes the alternative suggestion that immigrants of larger groups could profit from the ethnic capital available in their own group. One interpretation of this anomaly is to question the idea of ethnic solidarity and the supposed beneficial outcomes of belonging to a large immigrant group (Li 1977; Sanders and Nee 1987). However, the absence of a positive effect of group size on occupational status is not a strong case against the ethnic capital idea. Group size only indicates how many co-ethnics are available for help, but does not measure the resources available in the community, which are crucial for providing positive social support. In view of these considerations, I

168

Immigration and the Transformation of Europe

included two other contextual characteristics of immigrant settings that measure the resources available in the group: mean average education, and percent employed. I find that the higher the education of the immigrant community, the higher the occupational status of the immigrants of these groups. In addition, I find, for the male sample, a positive effect of the percent employed of the group on immigrants’ occupational status. These findings generally support the idea that the occupational chances of immigrants depend on the social capital available in their ethnic community.  Antecol, Heather 2000 “An Examination of Cross-Country Differences in the Gender Gap in Labor Force Participation Rates.” Labour Economics 7(4) (July): 409–26. Australian Office of Multicultural Affairs 1988 “Issues in Multicultural Australia.” Distributed by Australian Social Science Data Archive, Canberra Australia [data file]. Blalock, Hubert M. Jr. 1967 Toward a Theory of Minority-Group Relations. New York: John Wiley and Sons. Blau, Peter Michael and Otis Dudley Duncan 1967 The American Occupational Structure. New York: John Wiley and Sons. Borjas, George J. 1987 “Self-Selection and the Earnings of Immigrants.” American Economic Review 77(4) (September): 531–53. Borjas, George J. 1988 International Differences in the Labor Market Performance of Immigrants. Kalamazoo, MI: W. E. Upjohn Institute for Employment Research. Borjas, George J. 1992 “Ethnic Capital and Intergenerational Mobility.” Quarterly Journal of Economics 107(1) (February): 123–50. Borjas, George J. 1999 Heaven’s Door: Immigration Policy and the American Economy. Princeton, NJ: Princeton University Press. Brierley, Peter W. 1997 World Churches Handbook: Based on the Operation World Database by Patrick Johnstone, WEC International. London: Christian Research. Browne, W. J. 2002 “MCMC Estimation in MLwiN.” Accessed 2002. Available from http://multilevel.ioe.ac.uk/ Byers, John A. 2003 “Great Circle Distances Calculated between Points on Earth.” Accessed 2002. Last updated on 7 August. Available from http://www.wcrl.ars.usda.gov/cec/moregen.htm Chiswick, Barry R. 1978 “The Effect of Americanization on the Earnings of Foreign-Born Men.” Journal of Political Economy 86(5) (October): 897– 921. Chiswick, Barry R. 1979 “The Economic Progress of Immigrants: Some Apparently Universal Patterns,” in Contemporary Economic Problems, 1979, William John Fellner (ed). Washington, DC: American Enterprise Institute, 357– 99.

Occupational status of immigrants

169

Chiswick, Barry R. 1999 “Are Immigrants Favorably Self-Selected?” American Economic Review 89(2) (May): 181–85. Chiswick, Barry R. and Paul W. Miller 2001 “A Model of Destination-Language Acquisition: Application to Male Immigrants in Canada.” Demography 38(3) (August): 391–409. Chiswick, Barry R. and Paul W. Miller 2002 “Immigrant Earnings: Language Skills, Linguistic Concentrations and the Business Cycle.” Journal of Population Economics 15(1) (January): 31–57. Cobb-Clark, Deborah A. 1993 “Immigrant Selectivity and Wages: The Evidence for Women.” American Economic Review 83(4) (September): 986–93. Eurostat 1998 The European Union Labour Force Survey: Methods and Definitions 1998. Luxembourg: Eurostat. Eurostat 2003 “European Union Labour Force Survey, 1992–2002.” Luxembourg: Eurostat [data file]. Ganzeboom, Harry B. G. 2005 “Tools for Standardizing Occupation Codes into ISCO-88.” Accessed 2003. Available from http://home.scw.vu.nl/%7 Eganzeboom/occisco/ Ganzeboom, Harry B. G., Paul M. De Graaf and Donald J. Treiman 1992 “A Standard International Socio-Economic Index of Occupational Status.” Social Science Research 21(1) (March): 1–56. Ganzeboom, Harry B. G. and Donald J. Treiman 1996 “Internationally Comparable Measures of Occupational Status for the 1988 International Standard Classification of Occupations.” Social Science Research 25(3) (September): 201–39. Ganzeboom, Harry B. G. and Donald J. Treiman 2003. “Conversion Tools ISCO-88 into ISEI.” Accessed 2003. Available from http://home.scw.vu.nl/ %7Eganzeboom/occisco/ Grimes, Barbara F. (ed) 2000 Ethnologue: Languages of the World. 14th edn. Dallas, TX: SIL International. Jasso, Guillermina and Mark Richard Rosenzweig 1990 The New Chosen People: Immigrants in the United States. The Population of the United States in the 1980s. New York: Russell Sage Foundation for the National Committee for Research on the 1980 Census. Karatnycky, Adrian and Aili Piano (eds) 2002 Freedom in the World: The Annual Survey of Political Rights and Civil Liberties, 2001–2002. Piscataway, NJ: Transaction Publishers. Lambert, Pat 2003 “Conversion Tools Occupations US 1990 Census into ISCO-88.” Accessed 2003. Last updated on 27 May. Available from http://www.cf.ac.uk/socsi/CAMSIS/occunits/distribution.html Lenski, Gerhard Emmanuel, Jean Lenski and Patrick Nolan 1991 Human Societies: An Introduction to Macrosociology. 6th edn. New York: McGrawHill. Li, Peter S. 1977 “Occupational Achievement and Kinship Assistance among Chinese Immigrants in Chicago.” Sociological Quarterly 18(4) (Autumn): 478–89. McAllister, Ian 1995 “Occupational Mobility among Immigrants: The Impact of Migration on Economic Success.” International Migration Review 29(2) (Summer): 441–68.

170

Immigration and the Transformation of Europe

Model, Suzanne 1997 “An Occupational Tale of Two Cities: Minorities in London and New York.” Demography 34(4) (November): 539–50. Model, Suzanne and David Ladipo 1996 “Context and Opportunity: Minorities in London and New York.” Social Forces 75(2) (December): 485–510. Model, Suzanne and Lang Lin 2002 “The Cost of Not Being Christian: Hindus, Sikhs and Muslims in Britain and Canada.” International Migration Review 36(4) (Winter): 1061–92. Model, Suzanne, Gene Fisher and Roxane Silberman 1999 “Black Caribbeans in Comparative Perspective.” Journal of Ethnic and Migration Studies 25(2) (April): 187–212. OECD 1999 Classifying Educational Programmes: Manual for ISCED-97 Implementation in OECD Countries. Paris: Organisation for Economic Co-operation and Development. OECD 2000 International Migration Statistics. Paris: Organisation for Economic Co-operation and Development. Owen, Carolyn A., Howard C. Eisner and Thomas R. McFaul 1981 “A HalfCentury of Social Distance Research: National Replication of the Bogardus Studies.” Sociology and Social Research 66(1) (October): 80–98. Pineo, Peter C. 1977 “The Social Standing of Ethnic and Racial Groupings.” Canadian Review of Sociology and Anthropology 14(2) (May): 147–57. Portes, Alejandro and Robert L. Bach 1985 Latin Journey: Cuban and Mexican Immigrants in the United States. Berkeley, CA: University of California Press. Portes, Alejandro and Rub´en G. Rumbaut 1996 Immigrant America: A Portrait. 2nd edn. Berkeley: University of California Press. Portes, Alejandro and Rub´en G. Rumbaut 2001 Legacies: The Story of the Immigrant Second Generation. Berkeley, CA: University of California Press. Poston, Dudley L. Jr. 1994 “Patterns of Economic Attainment of Foreign-Born Male Workers in the United States.” International Migration Review 28(3) (Autumn): 478–500. Quillian, Lincoln 1995 “Prejudice as a Response to Perceived Group Threat: Population Composition and Anti-Immigrant and Racial Prejudice in Europe.” American Sociological Review 60(4) (August): 586–611. Raijman, Rebecca and Moshe Semyonav 1995 “Modes of Labor Market Incorporation and Occupational Cost among New Immigrants to Israel.” International Migration Review 29(2) (Summer): 375–94. Raudenbush, Stephen W. and Anthony S. Bryk 2002 Hierarchical Linear Models: Applications and Data Analysis Methods. 2nd edn. Thousand Oaks, CA: Sage Publications. Reitz, Jeffrey G. 1998 Warmth of the Welcome: The Social Causes of Economic Success for Immigrants in Different Nations and Cities. Boulder, CO: Westview Press. Reitz, Jeffrey G. (ed) 2003 Host Societies and the Reception of Immigrants. La Jolla, CA: Center for Comparative Immigration Studies, University of California, San Diego. Sanders, Jimy M. and Victor Nee 1987 “Limits of Ethnic Solidarity in the Enclave Economy.” American Sociological Review 52(6) (December): 745–67. Schoeni, Robert F. 1998 “Labor Market Outcomes of Immigrant Women in the United States: 1970 to 1990.” International Migration Review 32(1) (Spring): 57–77.

Occupational status of immigrants

171

Shields, Michael A. and Stephen Wheatley Price 2002 “The English Language Fluency and Occupational Success of Ethnic Minority Immigrant Men Living in English Metropolitan Areas.” Journal of Population Economics 15(1) (January): 137–60. Snijders, T. A. B. and R. J. Bosker 1999 Multilevel Analysis: An Introduction to Basic and Advanced Multilevel Modeling. London: Sage Publications. Statistics Canada 1991 “1991 Census Public Use Microdata File.” Distributed by Statistics Canada, Ottawa, Canada [data file]. Statistics Canada 1996 “1996 Census Public Use Microdata File.” Distributed by Statistics Canada, Ottawa, Canada [data file]. US Census Bureau 1980 “1980 US Census 1 Percent Public Use Microdata Sample.” Distributed by Interuniversity Consortium for Political and Social Research, Ann Arbor, Michigan [data file]. US Census Bureau 1990 “1990 US Census 1 Percent Public Use Microdata Sample.” Distributed by Interuniversity Consortium for Political and Social Research, Ann Arbor, Michigan [data file]. van Tubergen, Frank 2004 “The International File of Immigration Surveys.” Utrecht: ICS/Department of Sociology [data file]. van Tubergen, Frank, Ineke Maas and Henk Flap 2004 “The Economic Incorporation of Immigrants in 18 Western Societies: Origin, Destination, and Community Effects.” American Sociological Review 69(5) (October): 704– 27. van Tubergen, Frank and Matthijs Kalmijn 2005 “Destination-Language Proficiency in Cross-National Perspective: A Study of Immigrant Groups in Nine Western Countries.” American Journal of Sociology 110(5) (March): 1412– 57. Wanner, Richard A. 1998 “Prejudice, Profit, or Productivity: Explaining Returns to Human Capital among Male Immigrants to Canada.” Canadian Ethnic Studies 30(3): 24–55. Wilson, Kenneth L. and Alejandro Portes 1980 “Immigrant Enclaves: An Analysis of the Labor Market Experiences of Cubans in Miami.” American Journal of Sociology 86(2) (September): 295–319. World Bank 2001 World Development Indicators. Washington, DC: World Bank, CD-ROM. Zhou, Min and John R. Logan 1989 “Returns on Human Capital in Ethnic Enclaves: New York City’s Chinatown.” American Sociological Review 54(5) (October): 809–20.

7

Immigrants, unemployment, and Europe’s varying welfare regimes Ann Morissens

Introduction Both immigration and immigrants have been high on the political agenda in many European countries for a few decades. In recent years the tone of these debates has become harsher and more negative towards immigrants, and we have observed a tightening of policies related to immigration and immigrants. Arguably, the main policy consequence of these actions has been a decrease in social rights for immigrants and asylum seekers, while refugees’ social rights have remained strong in most rich nations. Prior to this period, immigration scholars (Soysal 1994; Guiraudon 2000) had observed an expansion of immigrants’ social rights. The focus of this chapter is on outcomes for immigrants in six countries that are characterized by different welfare settings: Canada, Germany, Norway, Sweden, the United Kingdom, and the United States. The inclusion of Canada and the United States in the analysis offers some broader comparative perspective on European experiences. The central theme of this chapter is the relation between welfare states and outcomes for immigrants in the socio-economic domain, with a focus on poverty and labor market participation. In the chapter, I compare outcomes for immigrants with outcomes for the non-immigrant population, and attempt to estimate the distance between the two groups. Large differences between the two groups can be considered an indicator of failing integration in the socio-economic domain (see also chapter 15 here). The second part of the chapter addresses the consequences of weak labor market attachment and examines the relationship between unemployment and poverty. I pay special attention to the role of transfers in alleviating poverty for immigrants in this group. The role of the market for poverty alleviation through market forces is also explored through a comparison of income poverty amongst native and immigrant households with different levels of labor market attachment. The chapter proceeds as follows. It briefly discusses welfare and unemployment regimes and generates hypotheses about immigrants’ outcomes 172

Europe’s varying welfare regimes

173

under different regimes. It then describes the unemployment and poverty situation in the six countries. Before turning to the empirical findings, data and concepts are discussed. The empirical part examines in more detail the role of both the labor market and welfare state for the alleviation of poverty amongst those with a tenuous labor market position. Finally, a conclusion sums up the chapter’s main findings and puts them in a policy context. Readers may also want to review these findings in light of those found in the contributions of Georg Menz (chapter 15), Frank van Tubergen (chapter 6), and Peterson et al. (chapter 3) to this volume.

Immigrants, welfare states, and unemployment regimes The aim of the chapter is to explore immigrants’ outcomes in terms of labor market participation and income. Both are interrelated with each other and, especially as employment at good wages, is often seen as a key for the alleviation of poverty. Another important actor is the welfare state. Amongst welfare scholars (Ringen 1987; Goodin et al. 1999) there seems to be agreement that poverty alleviation is one of the main objectives of welfare states and when they fail to eradicate poverty by market means, their raison d’ˆetre could be questioned. The question I am interested in is whether welfare states are equally or less equally successful in alleviating poverty amongst their immigrant population as compared to their native population. There is a vast welfare research literature, to which we can only make some reference given the limited space available in this chapter. Esping-Andersen’s typology (Esping-Andersen 1990) is probably the best known and makes a good starting point. One of the advantages of typologies is that they describe the main characteristics of welfare states in a compact way and make it easier to hypothesize about outcomes. The countries examined here represent the three welfare regimes as identified by Esping-Andersen. The United States, Canada, and the United Kingdom represent the liberal regime type, Germany serves as an example of the conservative corporatist welfare regime, while, Sweden and Norway are categorized as social democratic welfare regimes. In liberal welfare regimes, state intervention is limited and benefits are directed to those in need. Benefits are often lump-sum and meager, which makes them less effective against poverty. Conservative corporatist regimes are characterized by a strong insurance and thus contributions based on a social system that has a tendency to favor those with a strong employment record prior to benefit receipt. Those with a weak labor market attachment are often less protected against poverty. Social democratic regimes are known for their universal benefits, based on citizenship/residence rather than

174

Immigration and the Transformation of Europe

on contributions or need. Benefits are rather generous, which generally results in low poverty rates. The generosity of the welfare system is believed by some to act as a “magnet” on immigrants (Borjas 1999), though research on this topic is not conclusive. Other scholars (Faist 1996; Banting 2000) have argued that there is a relationship between the type of welfare state and the socioeconomic integration of immigrants. Banting (2000) sees better possibilities for immigrants’ social rights in corporatist and social democratic regimes, whereas with the exception of Canada, immigrants’ access to transfers is often more limited in liberal regimes. His focus is more on inclusion within the social security system and thus differs from the outcome approach that I apply. Thomas Faist (1996) notes that it is often easier for immigrants to find a job in liberal regimes, but that they at the same time face a greater poverty risk. In the social democratic and conservative regimes, access to the labor market may be more difficult, but the generosity of the system is likely to prevent poverty amongst the group of unemployed immigrants. A trade-off between work and poverty is suggested in this case. Unemployment and income situation amongst immigrants Our data indicates that immigrants’ labor market situation is better in countries that belong to the liberal cluster, which seems to confirm Faist’s hypothesis (Table 7.1). The exception is Norway, which displays an unemployment rate similar to Canada and the United States. However the gap between the immigrant and non-immigrant group is much more pronounced in Norway than in the other two countries. The liberal regimes seem the most successful in keeping the unemployment discrepancy between the two groups smallest. Unemployment rates amongst immigrants are considerably higher in Sweden and Germany and the difference in outcomes between immigrants and non-immigrants is also larger. Individual characteristics are also likely to influence the unemployment rate and here, immigration histories and policies come into play. It would lead us too far astray to discuss this issue in this chapter, but whether a country positively selects immigrants based on qualitative criteria or mainly receives refugees will affect the labor market possibilities1 of this group. 1

Canada is an example of a country that selects its immigrants; in the United States, ongoing family reunification is a source of labor; whereas the European countries in this chapter have been receiving a good share of asylum seekers during the last decade. Until the early 1970s, foreign labor was recruited to fill a shortage in the labor market.

Europe’s varying welfare regimes

175

Table 7.1 Unemployment rates amongst immigrants and non-immigrants, 2001–02 average

Germany Norway Sweden Canada US UK

Immigrants

Non-immigrants

12.9% 7.3% 10.7% 7.5% 5.9% 7.9%

7.7% 2.7% 4.6% 7.4% 5.4% 4.7%

Source: Re-calculated from SOPEMI 2003:51. For Norway, figure for immigrants is taken from SOPEMI 2003: 248, for non-immigrants the figure comes from ILO Laborsta and refers to the year 2000.

Data and definitions The results presented here are based on data from the Luxembourg Income Study (LIS).2 The Luxembourg Income Study is a cross-national data set with detailed information about sources of household income for over twenty countries. A special feature of LIS is the harmonization of the main income variables – disposable income, gross income, and market income – which creates a unique data bank for researchers to conduct cross-national analyses of poverty, income distribution and inequality. Besides detailed income variables, LIS also includes demographic and socio-economic information, as well as information about transfers. This allows us to see who is receiving transfers and how much. Immigrant population The possibilities for identifying immigrants within the LIS datasets are still limited despite recent efforts and when information about the immigrant status is available, comparability is often far from optimal.3 For this reason the analysis is limited to those countries where it was possible 2 3

See Appendix Figure 7.A.1 for a list of surveys and years. LIS is here, of course, very dependent on the information about immigrant status provided by the individual countries. Unfortunately, immigrant information is still missing for several countries but for those countries that have this information available, the efforts made by LIS to provide more detail are much appreciated. As comparative researchers, we can only hope that future surveys start to pay more attention to the topic of migration. It is important for researchers and policy-makers alike to collect key-data (i.e. such variables as country of birth, nationality, year of arrival in the country, etc.) in this area. Comparability amongst countries with regard to immigrant definition also remains a topic for improvement for the years to come.

176

Immigration and the Transformation of Europe

to identify the immigrant group in the most recent wave of LIS data,4 which is situated around the year 2000. Ideally I wanted to limit the analysis to non-European Union (EU) citizens,5 but the absence of information about nationality or country of birth for some countries made this impossible. One consequence of the non-uniformity in immigrant definitions amongst countries is that the results presented here can only be considered as exploratory and that they need to be treated with the necessary caution. The best workable definition of immigrants taking into account the problems mentioned and that serves as a common denominator throughout the countries is the following: an immigrant is someone who is foreignborn or someone born in the country with parents born abroad. I will refer to this group as immigrants or immigrant population. When I refer to the native population, I will use the concepts of natives and nonimmigrants interchangeably. It is the immigrant/ethnicity status of the head of the household that determines whether a household is counted as an immigrant/non-immigrant household.6 The LIS dataset contains two variables7 that can be used to identify the immigrant population in a country. A first variable is the ethnicity/nationality of head variable (D8) and a second variable is the immigration status of head variable (IMMIGRHD), though, there is no uniformity between countries as to the content of these variables. In what follows I will briefly describe the situation for each individual country. This will also clarify why the above immigrant definition was chosen and not a more ambitious or detailed one. For the Canadian dataset the ethnicity/nationality variable (D8) distinguishes between immigrants and non-immigrants. The immigration status of head variable (IMMIGRHD) shows how many years someone with an immigrant background has been in the country. It is not possible to have more information about the country of origin of the immigrant, 4 5

6 7

For this reason, Denmark and France are not included; they have immigrant information available but there are no data for the year 2000. European Union citizens are often temporary workers sent by their companies who are only residing a limited number of years in the new country. Besides the temporary character of their stay, within the EU they derive their social rights from EU agreements, which do not allow the individual countries to exclude them from the social security systems, as is often the case for immigrants from less developed countries. The right to reside in another EU country is also dependent on having employment, which results in a situation that most of the EU citizens in another EU country are employed. Besides this, within the EU countries, EU citizens only make for a relatively small share of the “foreign population” and can therefore not be seen as the majority immigrant group. For a fruitful discussion about this, see Saunders 1994. For some countries, there is also immigrant status information available in the pslot variable, for information see the basic descriptives by country.

Europe’s varying welfare regimes

177

nor his/her citizenship status. It is thus not possible to know whether or not someone who came to Canada as an immigrant has been naturalized or not. The naturalization problem does not occur for the United States because the variable used allows for a distinction between: native born, foreign-born naturalized citizen, foreign-born and non-citizen. In this chapter, both categories of foreign born are included in the immigrant group. Besides this distinction it is also possible to distinguish between the different ethnic groups, though we do not use this here. The data for the United Kingdom are the most problematic in terms of comparability due to the British emphasis on ethnicity. Most British data sources do not provide information about country of birth or nationality. The emphasis is on ethnicity and the following ethnic minority groups can be distinguished in the British dataset: White, Black African, Black Caribbean, Black other, Indian, Pakistani, Bangladeshi, Chinese, and other. This information does not give us information on whether someone was foreign-born or not. To allow for some sort of comparison, which remains however less than ideal, I use the group of white people as a proxy for the non-immigrant group. The group of other ethnicities is used as a proxy for the immigrant population. This data issue does of course disrupt the comparability with the other countries for which we selected the foreign-born persons to represent the immigrant group. One of the arguments in favour of using ethnicity rather than country of birth is that visible aspects do not change after having naturalized or having been in the country for a long time.8 The German dataset in LIS contains information about the nationality of the persons included in the survey. All those with non-German nationalities serve as the immigrant population in the analysis. Norwegian data has two variables that can be used for selecting the immigrant population; I use the immigrant status of head variable. This variable has the following categories: no immigrant background, first generation immigrant without Norwegian background, persons born in Norway with parents born abroad (second generation), persons born 8

Citizenship or the fact of being born in the country in which one lives is not always sufficient to be no longer considered an outsider. This is also one of the reasons why some British scholars (Modood and Berthoud 1997: 13) have pointed out the inadequacy of the concepts of nationality or citizenship and country of birth as selection criteria for the immigrant population. I believe that this is an important observation, because it is true that in their daily lives, immigrants and their children interact with people who often consider them as “strangers” and this may have an impact on their chances for incorporation. Other characteristics such as family name, skin color or different dress codes do not disappear by taking a country’s citizenship, and they can easily result in a differential or discriminating treatment.

178

Immigration and the Transformation of Europe

abroad with Norwegian-born parents, adopted persons, persons born abroad with one parent born abroad. I include both the first and second generation. In the Swedish household file, there is information on both the country of birth/nationality and immigrant status. The latter variable refers to the year of arrival in Sweden, thus implying that the person was born abroad. A cross-tabulation of immigrant status and nationality variables showed that many of those who arrived in Sweden as immigrants have acquired Swedish citizenship. As for the other countries for which I had this information, naturalized citizens are included in the immigrant group. Poverty incidence This section describes the poverty situation for the immigrant group; the non-immigrant group and the overall population for those within working age (eighteen–sixty-four). The relative poverty measure is used and I report poverty rates at 50 percent of the median of the equivalent disposable income. This threshold is calculated for the entire population. To adjust the disposable income for family size, I used the square root of family size as equivalence scale. To assess the effectiveness of unemployment and means-tested benefits, respectively, poverty rates based on equivalent disposable income minus unemployment benefits or means-tested benefits are compared with poverty rates based on disposable equivalent income. The former refers to the disposable income people have before receiving unemployment benefits/means-tested benefits. The poverty rate based on equivalent disposable income includes unemployment benefits/means-tested benefits and the reduction in poverty can thus be attributed either to unemployment or to means-tested benefits. The results are expressed as “relative poverty reduction” effectiveness scores. The poverty rates confirm findings of previous studies (Berthoud 1998; Buchel ¨ and Frick 2003; Platt 2002; Blume et al. 2003), namely that immigrants in most countries face a far larger poverty risk than their native counterparts (Figure 7.1). Poverty rates are of course only one indicator9 and it is likely that there are differences within the immigrant group as well. The data available from the Luxembourg Income Study (LIS) unfortunately does not allow for a more detailed analysis for specific immigrant groups. But, even with better comparative data, the number of 9

The head-count method of poverty has also been criticized because of methodological flaws (Ringen 1987) but because the battle to make immigrant data somewhat comparable is a big enough fight, this discussion will not be taken up here. Rather, we stick to the method of poverty measurement, which is still used in comparative research.

Europe’s varying welfare regimes

179

30

25

%

20

15

10

5

0

GE00

NW00 immigrants

SW00 CN00 country/year non-immigrants

US00

UK99

overall population

Figure 7.1. Percentage of persons under the 50 median income poverty line by immigrant status. Source: LIS; own calculations.

immigrants available in surveys is likely to be insufficiently large to allow for more sophisticated forms of analysis. Some scholars have argued that this is one of the reasons why an immigrant focus is often missing in studies related to social security (Craig 1999). We see that immigrants have on average higher poverty rates than the non-immigrant population, and therefore compared to the population as a whole. We find that immigrant poverty is highest in the United States (24.7 percent) and the United Kingdom (21.3 percent). The same pattern is found for the non-immigration group. Immigrants in Norway, Germany, and Canada have similar immigrant poverty rates (around 15 percent) and they are moderate compared to those found in the United States and the United Kingdom. Sweden, with poverty rates around 10 percent, has the lowest poverty rate amongst its immigrant population. Turning to the non-immigrant group, a familiar pattern can be identified; countries belonging to the social democratic cluster have the lowest poverty rates. Germany, as a representative of the corporatist conservative model, has a moderate poverty rate, whereas the three liberal regimes have the highest poverty rates of the six countries under investigation here. For immigrants, Norway has a poverty rate that is slightly

180

Immigration and the Transformation of Europe

Table 7.2 Poverty ratio Immigrants/Non-immigrants Germany 2000 Norway 2000 Sweden 2000 Canada 2000 US 2000 UK 1999

2.2 2.6 1.9 1.4 1.7 2.1

Source: LIS; own calculations.

higher than the Canadian one, and thus does not follow the social democratic pattern as it does for the non-immigrant group. Looking at the ratios of poverty rates for immigrants to non-immigrants (natives), we have an indication of the poverty gap between the immigrant population and natives. Two clusters, can be identified in Table 7.2. A first cluster consists of Canada, the United States, and Sweden, and has a ratio under two. The other three countries form a second cluster and have ratios of over two, with Norway having the highest ratio (2.6 percent). In Norway, immigrants are almost three times more likely to be poor than their non-immigrant counterparts. The clusters of countries, which we found for poverty rates in general and amongst natives, do not entirely hold for ratios of immigrants to native poverty rates. Unemployment and poverty The causes of poverty are without doubt multiple. As will be illustrated, lack of employment is often one of these. Welfare states have set up social transfers schemes with the aim to protect people against loss of income due to sickness, old age or market failure. Unemployment schemes can be seen as representative of the latter, but they are also among the most contested, and some countries have only implemented limited versions of protection. Accordingly, we expect to find variation in the poverty outcomes of workers due to the unemployment regime in which the immigrant lives. In addition to the social insurance schemes, there is a residual category of means-tested benefits that constitutes the ultimate safety net upon which people can rely when they are not eligible for social insurance benefits. The need to rely on means-tested benefits depends amongst other things on the social insurance eligibility conditions and benefit generosity. If these are strongly linked to employment-related contributions, those outside, or at the margins of, the labor market will find it hard to

Europe’s varying welfare regimes

181

qualify and will probably have to rely on means-tested benefits. In countries where social insurance benefits have a universal character, based on citizenship or residence, whether or not with an earnings-related part, people outside the labor market may be less excluded from access to unemployment benefits. The problem of unemployment has in recent years received the attention of both policy makers and welfare scholars (Gallie and Paugam 2000; Goul-Andersen and Jensen 2002; Sarfati and Bonoli 2002). At the same time it has launched a neo-liberal discourse in several countries, which in turn has resulted in an increasing emphasis on active measures to reintegrate the unemployed into the labor market. The other side of the coin has been a decrease in social rights for the unemployed, mainly reflected in tighter eligibility conditions, less generous benefits, and consequently an increasing risk for stigmatization and social exclusion due to unemployment. These trends have inspired scholars (Gallie 2004; McGinnity 2004) to focus on the problem of unemployment and those affected by it. However, the ways by which the lack of employment affects immigrants has not received much attention. This is somewhat surprising because the difficulties that immigrants face in finding steady jobs has been reported by immigration and labor market scholars and international institutions alike (Faist 1998; OECD 2000; Boje 2003). Attempting to fill this void, I will especially focus on immigrants and the financial implications of their unemployment. This chapter does not aim to discuss the policy responses in detail, or to evaluate its successes or failures; it is mainly exploratory in character and focuses on descriptive outcomes. Menz (chapter 15) discusses these benefits in a broader context. Unemployment regimes As have many welfare scholars (Esping-Andersen 1990; Leibfried 1992; Ferrera 1996; Korpi and Palme 1998) before them, Gallie and Paugam (2000: 3–13) have engaged in the business of “welfare state” typology.10 By focusing on a particular scheme, unemployment insurance, they have responded to the need for less abstract typologies, resulting in typologies of social programs alone rather than typologies of welfare states. Since my interest in this chapter is on unemployment, their typology is a good starting point to hypothesize about unemployed immigrants’ economic outcomes. Therefore, I will from now on refer to “unemployment 10

Peter Abrahamson (1999) used this concept to refer to the wide range of regime typologies that emerged from comparative welfare research.

182

Immigration and the Transformation of Europe

Figure 7.2. Unemployment regimes and expected poverty outcomes

Unemployment regime

Country

Expected poverty outcome for unemployed immigrants

Sub-protective Liberal-minimal Employment-centered Universalistic

US UK; Canada Germany Norway; Sweden

High High Moderate Low

Source: Based on Gallie and Paugam (2000).

regimes” as defined by Gallie and Paugam, rather than using the more abstract concept of welfare regimes. Gallie and Paugam use indicators related to a country’s effort to deal with unemployment as the basis for their typology. These indicators are: the level of coverage, the level and duration of benefits, and the existence and extent of active labor market policies. Although I recognize that the latter can positively influence job opportunities and increase immigrants’ skills for immigrants, this effect lies outside the scope of this chapter. The effects and outcomes of active labor market policies for a group of immigrants is still an under-researched topic, and an attempt to deal with it here would lead us too far from the main concern of this chapter. Still, it is a topic that definitely deserves further exploration. Based on these indicators, Gallie and Paugam distinguish four unemployment regimes: the sub-protective, the liberal-minimal, the universalistic and the employment-centered regime. They hypothesize that the outcomes for the unemployed will vary amongst these regimes and that within the group of unemployed, differences may exist amongst men and women, immigrants and citizens; it is the latter comparison that will be further explored in this chapter. Despite the documented disadvantage of immigrants in the labor market in most countries, very few comparative analyses have focused on the consequences of this typology in terms of financial deprivation. This chapter aims to remedy this shortcoming by focusing on immigrants. Both the sub-protective and liberal-minimal unemployment regimes are characterized by incomplete coverage, meager benefits and therefore a considerable risk for financial deprivation amongst the unemployed. The liberal-minimal regime should have better outcomes because it is also slightly different from the sub-protective regime in ideological terms. There is the political will to intervene to a certain extent in case of market failure, but the intervention remains limited because of the risk of disincentives to work. As a consequence, unemployment benefits are often means-tested in such a regime.

Europe’s varying welfare regimes

183

Turning to the universalistic unemployment case, this regime is considered to protect most of the unemployed from poverty through a generous benefits system, which is also extensive in time of coverage. It also has a comprehensive coverage because of its non-stringent eligibility conditions and the individualization of rights. There are nevertheless some details that may make this system more exclusionary for immigrants than one may expect at first sight. Some of the countries that can be classified under the universalistic regime have an unemployment scheme based on voluntary insurance. This requires membership of a fund to which contributions need to be paid. Those who have not joined the funds will have to rely upon means-tested benefits which form the second tier of the unemployment insurance in these countries. These benefits are less generous than the regular unemployment benefits. Means-tested unemployment insurance may also cause an additional barrier for immigrants who are unfamiliar with the system and the national language. The employment-centered unemployment regime is similar to the universalistic regime with regard to generosity and the duration of benefits, but distinguishes itself from it in terms of coverage. Access to the generous benefits is reserved to those with an extensive employment history. This is because unemployment schemes have the character of a contribution-based insurance related to employment. Those without a sufficient employment record and thus insufficient contributions to be entitled to benefits, will find themselves excluded from the unemployment scheme and may have to turn to less attractive protection schemes, in most cases social assistance. At first sight, the sub-protective and liberal-minimal regimes seem to have better inclusion chances for immigrants in terms of coverage. They put more emphasis on need and less emphasis on work history (albeit employment duration remains an eligibility criterion in the insurancebased part). The downside is that needs-based benefits are low in such regimes, which in turn increases the risk of poverty. The means-tested nature of benefits can also be a problem for immigrant women since they must take the household income and not individual job loss as a reference to assess need for benefits. Unemployed versus workless households Since my focus is on those who are unemployed, I am mainly interested in persons of working age. For the poverty analysis, I use the household as the unit of analysis and therefore only households with a head within working age (eighteen–sixty-four) are included in the analysis. A further narrowing down of this category to those aged twenty-five–sixty-five could be more appropriate and would take into account the longer education

184

Immigration and the Transformation of Europe

spells for some groups, but this approach would also result in a smaller number of observations. The immigrant sample available is already small, so a more narrow definition of working age was not desirable. One problem with identifying unemployed persons within the LIS datasets is the lack of a harmonized LIS definition for unemployment.11 LIS uses the labor force status variables as provided in the different surveys. These variables provide insufficient information to make a simple distinction between those who are in employment and those who are not. Not all countries have a straightforward definition of unemployed because they fail to distinguish between persons not in the labor market and the jobless. This is the case for Germany, Sweden, and Norway. Despite the detailed information provided by the Canadian survey, it does not allow us to identify the unemployed in a straightforward way either. Given these problems, I decided to use two alternative approaches that allow a resolution. The first approach identifies those households in which there is income from unemployment benefits. This is an indicator that at least one household member is receiving this type of benefit and can consequently be considered to have at least one unemployed member. The second approach identifies those households in which there is no earner present. I will refer to those households as workless households. After having discussed the incidence of unemployment and workless households amongst the immigrant population, the second step in the analysis will turn to the financial consequences of the lack of income from labor. The risk of poverty amongst unemployed and workless immigrant and non-immigrant households in the different countries and the importance of employment can be seen when the same exercise is repeated for households in which there is at least one earner. Based on previous research (Forster ¨ 1994) we expect a drop in poverty as soon as there is one earner in the household, this decrease will be even more outspoken when two earners are present. If we find similar results for the immigrant group, the importance of labor market participation will be clearly demonstrated. It would at the same time indicate that employment plays a key role for immigrants’ integration and that it is one of the best remedies against poverty. 11

Recently LIS has been integrating variables related to labor market participation, they are the so-called Luxembourg Employment Study (LES) variables; one of these variables refers to the ILO definition of unemployment and distinguishes the three categories, employed, unemployed, and not in the labor force. The ILO definition is also well suited for comparative studies. Unfortunately, this LES variable was unavailable for most of the countries under investigation here. For a discussion about different unemployment definitions and problems of comparability see McGinnity 2004.

Europe’s varying welfare regimes

185

Table 7.3 Poverty incidence amongst households in receipt of unemployment benefits

Germany 2000 Norway 2000 Sweden 2000 Canada 2000 US 2000 UK 2000

Immigrants

Non-immigrants

20.6 3.7 8.3 8.2 22.3 29.3

9.7 3.6 4.7 4.2 15.2 29.9

Source: LIS; own calculations.

One potential consequence of unemployment is poverty (Table 7.3). Besides the duration of the unemployment, eligibility and generosity of the existing unemployment benefits play an important role. Due to space restrictions in this chapter, it is not possible to discuss the unemployment schemes in detail so the results based on micro data will only be tested against the unemployment regimes typology. Another problem, is that these poverty rates do not include those who are unemployed but do not receive benefits, which is likely to underestimate the poverty rates. We actually find lower poverty rates amongst those who receive unemployment benefits than amongst the immigrant and non-immigrant groups in general. This can be an indication that unemployment benefits are effective in pulling people out of poverty. Testing the results against the unemployment regimes typology, we find moderate poverty rates in the employment-centered regime for both the immigrant and non-immigrant group. Sweden and Canada have slightly lower poverty rates for the immigrant-group compared to Germany, but seem to do a better job for the non-immigrant group. For this group, Canada clusters together with Sweden and not with the United Kingdom as we would have expected. Only Norway has low poverty rates for both groups. The United Kingdom and the United States have the highest poverty rates. Immigrants in the United States who receive unemployment benefits are considerably more at risk for poverty than their non-immigrant counterparts. Looking at the effectiveness of unemployment benefits for immigrants (measured as the percent reduction in the poverty rate due to unemployment benefits), we do find a wide range of effectiveness rates ranging from a minimum of 28 percent reduction in the two liberal-minimal regimes to a maximum of 81 percent reduction in Norway. The other nation representative of the universalistic regime,Sweden, also displays a good result with 75 percent reduction.

186

Immigration and the Transformation of Europe

Table 7.4 Relative poverty reduction effectiveness scores for unemployment benefits

Germany 2000 Norway 2000 Sweden 2000 Canada 2000 US 2000 UK 1999

Immigrants

Non-immigrants

47.4 81.6 74.5 28.7 28.1 53.9

63.7 62.9 78.7 55.8 24.7 46.9

Source: LIS; own calculations. Note: The table shows percent reduction in poverty due to receipt of unemployment benefits.

The United Kingdom with its score of 53 percent, comes closer to the employment-centered regime, Germany (47.4), than to its liberalminimal regime colleagues, Canada and the United States. Regarding Canada, we should also mention the low poverty rate (11 percent) prior to receiving unemployment benefits. In the United Kingdom, on the contrary this figure is much higher (63 percent). The initial low poverty rates in Canada partly explain the lower effectiveness results. The high poverty rates prior to receiving unemployment benefits in the United Kingdom already indicate that there are more factors at play than just unemployment. Other countries display much lower poverty rates prior to receiving unemployment benefits, which indicates that there are likely other earnings present in the household. Income from earnings from a partner can have a beneficial influence on a household’s financial resources. As shown in the next section, the absence of earners is a problem in the United Kingdom. As for the non-immigrant population, we find a similar poverty pattern but as for the overall poverty observations, unemployed non-immigrants are less likely to be poor, with the exception of the United Kingdom. Incidence of workless households The remainder of the chapter deals with households that are very vulnerable to poverty: those in which there is no earner present. I consider the lack of earners in a non-elderly household as a proxy for a very weak labor market attachment and define these households as workless. By taking no-earner households as the unit of analysis, we are also able to control for immigrants’ weaker labor market participation. If we compare the immigrant and non-immigrant groups’ financial situation amongst households without a single earner, immigrants’ weaker labor market

Europe’s varying welfare regimes

187

Table 7.5 Distribution of no-earner, one-earner, and two-earner households amongst immigrants and non-immigrants Immigrant households

Germany 2000 Norway 2000 Sweden 2000 Canada 2000 US 2000 UK 1999

Non-immigrant households

No-earner

TwoOne-earner earners No-earner

TwoOne-earner earners

10.4 7.5 14.3 7.5 5.8 29.6

47.6 35.2 36.2 28.5 35.9 41.7

39.4 27.1 32.6 36.1 37.9 32.6

42.1 38.6 36.2 41.7 10.1 23.4

9.1 3.4 5.8 6.5 7.1 20.6

42.2 46.1 45.7 44.5 43.2 37.8

Source: LIS; own calculations.

attachment cannot be the explanation for the difference in outcomes. Additionally, high poverty amongst workless households can indicate a less favorable treatment of immigrants by the existing social insurance schemes. This does not mean that the system is directly discriminating against the non-immigrant group, but it is possible that certain rules make access to benefits or eligibility more difficult for the group of immigrants. Other authors (Schierup 1993; Berthoud 1998) have referred to this as a form of “indirect” or “institutional” discrimination. A first section describes the incidence of work-poor households in the six countries under investigation in this chapter. This will inform us whether work-poor households are equally distributed amongst the immigrant and non-immigrant population, and whether there are significant differences amongst countries in terms of incidence of workless households. The results of this exercise can also shed light on whether the trend toward a universal “dual earner” society is taking place in all countries, and if this trend can also be observed amongst the immigrant group. Looking at the incidence of workless households amongst the immigrant population, we find again a pattern of disadvantage (Table 7.5). In these nations’ labor markets, immigrants are more likely to experience difficulties, compared to their non-immigrant counterparts. With the exception of the United States, there is a higher share of households without a single earner amongst immigrants. Yet, there is variation between countries. In Germany, Canada, and the United States, the divergence between the immigrant and non-immigrant group is only marginal. In Norway and Sweden, however, there is a pronounced difference between the two groups. Both countries have the lowest incidence of workless households amongst the non-immigrant group but the figure is more than twice as high for the immigrant group.

188

Immigration and the Transformation of Europe

From Table 7.5 it is also clear that workless households are a serious problem in the United Kingdom, and that this is even more the case for households with an ethnic background. Of the latter, almost 30 percent finds themselves in a situation where there is not a single earner. For the white ethnic minority group, this is 20 percent. Both figures largely exceed those in other countries. Turning to the other side of the continuum, dual-earner households, we find rather similar results for the non-immigrant group in all countries. With the exception of the United Kingdom, the share exceeds 40 percent and is highest in the two social democratic countries, Sweden and Norway. This is not surprising given the high female participation rates in these countries and the efforts made to reconcile work and family life by means of extensive daycare provision and generous parental leave possibilities. But in Canada and the United States, countries that are not particularly famous for their women-friendly policies, we also find a high proportion of dual-earner households. This result is probably driven by the low starting wages offered in these nations (Smeeding 2006). Only the United Kingdom has failed to reach a similar level of dual-earner households. It is not amongst the leaders in terms of single-earner households either, which once more emphasizes that the main problem is the high share of workless households. Hypothetically, we expected immigrant households to be underrepresented within the group of dual-earner households. There are several reasons for this. They have more difficulties finding a secure labor market position. They often come from countries with a different cultural background in which it may not be common for women to work.12 Higher fertility rates can also make it hard for immigrant women to work, especially in countries where there is not much publicly-funded daycare. This hypothesis was confirmed in Figures 7.3 and 7.4. Non-immigrant households are more likely to have two earners than immigrant households (Table 7.5). Availability of extensive daycare does not seem to result in more dual-earner families amongst immigrants. Whereas Norway and Sweden rank first when it comes to non-immigrant dual earners’ families, they do not achieve the same result for immigrant families. Two liberal regime representatives, Canada and the United States, have the best scores here and the difference with the non-immigrant group is also minor. As noted previously, the United Kingdom has the worst result for immigrants. 12

Looking at unemployment rates by gender for 2001–02 (SOPEMI 2004), we do however find that immigrant women have slightly lower poverty rates than their male counterparts in Germany, Sweden, and the United Kingdom. This shows that there are several factors that influence immigrant women’s labor force participation; ideally one should break down these figures by different cultural background.

Europe’s varying welfare regimes

189

100

80

%

60

40

20

0

GE00

SW00

NW00

CN00

US00

UK99

country/year no earner

one earner

two earners

Figure 7.3. Poverty rates (50% median) for different earners’ groups amongst immigrants. Source: LIS; own calculations.

These results can be combined with the earlier observed higher unemployment rates and serve as another indicator that immigrants have a weaker labor market attachment in all countries and that this is especially true for the United Kingdom, Sweden, and Germany. The United States and Canada, and to a lesser extent Norway, have been more successful with regard to immigrants’ labor market integration if we use the number of earners in the household as an indicator for the strength of labor market attachment. Workless households and poverty What are the consequences of the lack of labor market attachment for households’ economic resources? Not surprisingly, the results in Figures 7.3 and 7.4 clearly demonstrate that the absence of earners in the household dramatically increases the risk for poverty. This is even more so for the group of immigrants, which may be an indication that this group controls less savings and financial reserves than the non-immigrant group. Only in Sweden do immigrants in workless households have a similar poverty risk as their Swedish counterparts. Norway is successful in keeping the poverty rate of its workless non-immigrant households below

190

Immigration and the Transformation of Europe

80

%

60

40

20

0 GE00

SW00

NW00 CN00 country/year no earner

one earner

US00

UK99

two earners

Figure 7.4. Poverty rates (50% median) for different earners’ groups amongst non-immigrants. Source: LIS; own calculations.

30 percent, but is not able to achieve the same result for its immigrant population whose poverty rate is over 60 percent. As soon as there is one earner present in the household, the risk for poverty decreases significantly; it is at least more than halved in all of the countries. The decrease is most spectacular in Germany. The pattern is similar for the non-immigrant group. Having a second earner in the non-immigrant household almost eradicates poverty completely in Germany, Norway, and Sweden. For these three countries we also find good results for the immigrant households, bringing the poverty rates below the overall level that was discussed previously. This is a clear indication that employment offers a good protection against income poverty. The United Kingdom also has good results. The United States, on the contrary, continues to stand out with a poverty rate that remains high in a comparative perspective. The poverty rate for two earner immigrant households remains three times as high as the poverty rate for their non-immigrant peers. This is a clear indicator that economic activity alone is not always a sufficient strategy to reduce poverty; it only works when proper wages and salaries accompany it (Smeeding 2006). These results suggest the importance of labor market

Europe’s varying welfare regimes

191

Table 7.6 Recipience of means-tested benefits amongst workless households Immigrants

Non-immigrants

54.5 54.5 62.1 48.0 43.7 84.2

35.8 39.1 46.8 57.0 44.5 72.1

Germany 2000 Norway 2000 Sweden 2000 Canada 2000 US 2000 UK 1999 Source: LIS; own calculations.

participation for the alleviation of poverty, but also that, in countries like the United States, the wage level also is a decisive factor in poverty alleviation. The role of means-tested benefits It is clear that if at least one member of the households has a job, the risk of poverty decreases. But what if employment is not an immediate solution? It is not hard to imagine that a newly arrived immigrant family does not have the required skills to find a job right away. In such a situation, the immigrant family will have to rely on benefits provided by the state or use their own capital and savings (if the latter are available). The question then is: does the state guarantee these households a living standard that is in line with the common practice in a country? From the poverty levels amongst workless households we have seen that most countries fail to do this. For work-poor households we assume that means-tested benefits are the last safety net. The question then is are they effective in pulling work-poor households out of poverty? More and more countries have made access to social assistance for newly arriving immigrants much more restrictive and also require sufficient personal financial means in case of family reunification. This can have serious consequences for the immigrants’ financial situation, which may hamper their further integration process. First, I turn to the share of workless households that receive some sort of means-tested benefit. The LIS variable MEANSI was used to calculate the recipience rate. This variable includes means-tested cash benefits and near cash benefits (more details are provided in the Figure 7.A.1). As expected, I find rather high recipience rates of means-tested benefits amongst this group (Table 7.6). The highest rate is found in the United Kingdom where 84.2 percent of immigrant workless households receive

192

Immigration and the Transformation of Europe

100%

80%

60%

40%

20%

0% GE00

SW00 poverty level

NW00

CN00

means-tested transfers

US00

UK99

other transfers

Figure 7.5. Poverty reduction effectiveness by benefit type for workless immigrant households. Source: LIS; own calculations. Note: Percent reduction in poverty rate due to means-tested and other transfers, where 100 percent is the poverty rate before transfers of any type and poverty level is the rate of poverty after receipt of benefits for the group.

at least one form of means-tested benefits. The lowest rate was found in the United States with 43 percent. In both the United States and Canada, non-immigrants were more likely to receive means-tested benefits. Poverty reduction: The effects of benefits on poverty rates for workless households To answer the question whether means-tested benefits are effective in serving as a safety net for this particularly vulnerable group, I now turn to their ability to lift workless households above the poverty threshold. This factor is expressed by a “poverty reduction effectiveness ratio.” I compare the poverty rates of workless households prior to having received mean-tested benefits with the situation after having received them. If the poverty rate is lower, it means that the benefits have been effective and that the household was lifted above the poverty line. Effectiveness rates range from a low of 3.3 percent in the United States to a high of

Europe’s varying welfare regimes

193

100%

80%

60%

40%

20%

0% GE00

SW00 poverty level

NW00

CN00

means-tested transfers

US00

UK99

other transfers

Figure 7.6. Poverty reduction effectiveness by benefit type for workless non-immigrant households. Source: LIS; own calculations. Note: Percent reduction in poverty rate due to means-tested and other transfers, where 100 percent is the poverty rate before transfers of any type and poverty level is the rate of poverty after receipt of benefits for the group.

57 percent in Sweden. Put differently, means-tested benefits lift 3.3 percent of households that were poor prior to receipt of these benefits above the poverty threshold after they have received them. In Sweden this is much higher at 57.7 percent (Figure 7.5). As discussed earlier, poverty was lowest amongst workless immigrant households in Sweden and much higher in the United States (Figure 7.5). Canada with the highest poverty rate amongst the workless also has a low effectiveness result (7.2 percent). The same is true for Germany (15.2 percent) that nonetheless has a slightly lower poverty rate than Canada and the United States. Norway occupies an intermediate position both in terms of effectiveness and poverty. The two countries with the lowest poverty rates are also those with the highest effectiveness rates. Sweden and the United Kingdom have the best results here. These estimates illustrate the effectiveness of means-tested and other benefits for workless households. From these results it is clear that despite

194

Immigration and the Transformation of Europe

the similar effectiveness of means-tested benefits in the United Kingdom and Sweden, other social transfers play an important role as well in Sweden, which also explains Sweden’s good result in terms of poverty amongst workless households. Turning to non-immigrants (Figure 7.6), there is some variation in the pattern. Poverty rates are generally slightly lower and I also find that transfers other than means-tested benefits are an important source for reducing poverty in workless households. This is the case in all countries, but it is most visible in Germany and Sweden and to a lesser extent, Norway. Whereas means-tested benefits were the most important tool for reducing poverty amongst immigrant workless households, we see that other transfers have taken over this role in the case of non-immigrant households. In Norway there is an increased role for both means-tested and other transfers. Conclusion This chapter has explored immigrants’ labor market and income situations in different welfare regimes. It has focused on the financial consequences of immigrants’ weaker labor market attachment. Given the problems faced in terms of comparability and small sample sizes, the results should be interpreted as suggestive. In terms of labor market participation we find the best results for immigrants in the United States, Canada, and Norway. Earlier research has indicated that it is probably easier for immigrants to gain access to the labor market in countries with less regulated labor markets (Faist 1998). The results for the United States and Canada confirm this finding. The result for Norway is, however, surprising, as an example of the social democratic regime, it is also characterized by a highly-regulated labor market and one would expect a weaker labor market attachment for immigrants. Sweden, on the contrary, confirms the hypothesis concerning more difficult access to highly-regulated labor markets. Based on the deregulation thesis, we would also have expected better outcomes for immigrants in the United Kingdom. But we found bad scores on both indicators for employment: the level of unemployment and the incidence of work-poor households. Though it has to be noted that the latter is not a problem specific to the immigrant group, but rather a general problem in the United Kingdom. Germany, which is also characterized by a regulated labor market, scores moderately in terms of integration of immigrants in the labor market. In a next step I examined the consequences of immigrants’ precarious labor market situation. The poverty rates calculated for households with

Europe’s varying welfare regimes

195

different numbers of earners showed that the presence of one earner resulted in a significant poverty decrease. For dual earner households, poverty was almost eradicated. This shows that efforts to make access to the labor market easier for immigrants should be encouraged and are likely to have good results in combating poverty amongst the immigrant group. Still, policy makers cannot always expect newly arrived immigrants and asylum seekers to find a job overnight, and several factors may complicate their integration in the labor market (economic situation, language difficulties, psychological problems, discrimination). For this reason it is important to see if countries are able to guarantee a decent level of living in the meantime. Here the welfare state comes into the picture. Turning to the role of the welfare state for the group of workless households, I expected the lowest poverty rates in countries with the most extensive and generous welfare provisions. We did indeed find the lowest poverty rates in one of the social democratic regimes, Sweden. Whereas Sweden had difficulties integrating immigrants in the labor market, it is successful in keeping the poverty rates under 30 percent, and at the same level as the non-immigrant group. We also see that the United Kingdom, despite the high incidence of workless households, is relatively effective in combating poverty amongst this group. A plausible explanation is that the meanstested benefits do an effective job for those with a weak employment record regardless of citizenship status. Germany and Norway occupy a medium position in terms of poverty amongst workless households. Despite the good outcomes for Canada and the United States in terms of labor market participation, poverty amongst those outside the labor market is a serious problem. The transfers are largely ineffective to pull workless households above the poverty line. This is not only a problem for the immigrant group, but for the nonimmigrant group as well. Also at the overall level, immigrants’ poverty rates tend to be higher in these two countries compared to Germany and Sweden. Norway once more deviates from the expected outcome and comes closer to the poverty levels of the liberal countries. Despite the relatively good results in some countries and the small differences between immigrants and non-immigrants, poverty levels for the workless immigrant households are alarmingly high. Further cutbacks in benefits for newly arrived immigrants or exclusion from certain benefits are unlikely to improve their financial situation. Policy makers seem to think that taking away or lowering benefits is a sufficient tool to force immigrants into the labor market (or to prevent immigrants from coming). If this approach remains unsuccessful, they are left with a group of

196

Immigration and the Transformation of Europe

immigrants that is extremely poor. This poverty is likely to affect their chances of integration in other areas as well and will also deprive their children from life chances in line with those of non-immigrant children. Instead of taking away financial means from immigrant families, more effort should be made to remove barriers in the labor market and to find jobs with decent wages. Such a policy is likely to be beneficial for all parties involved. Increased employment will have a positive impact on immigrants’ financial situation and can also facilitate their integration in other areas. At the same time they will be less dependent on benefits, which is likely to be welcomed by the policy makers.  Abrahamson, Peter 1999 “The Welfare Modelling Business.” Social Policy and Administration 33 (4) (December): 394–415. Banting, Keith 2000 “Looking in Three Directions: Migration and the European Welfare State in Comparative Perspective,” in Immigration and Welfare: Challenging the Borders of the Welfare State, Michael Bommes and Andrew Geddes (eds). London: Routledge, 13–33. Berthoud, Richard 1998 “The Incomes of Ethnic Minorities.” ISER Report 98– 1. Colchester: Institute for Social and Economic Research, University of Essex. November. Blume, Kræn, Bjorn ¨ Gustafsson, Peder J. Pedersen and Mette Verner 2003 “A Tale of Two Countries: Poverty among Immigrants in Denmark and Sweden since 1984.” WIDER Discussion Paper No 2003/36. Helsinki: United Nations University, World Institute for Development Economics Research, April. www.wider.unu.edu/publications/dps/dps2003/dp2003–36.pdf Boje, Thomas P. 2003 “Age and Gender Differences in Labor-Force Participation and Employment,” in Post-Industrial Labour Markets: Profiles of North America and Scandinavia, Thomas P. Boje and Bengt Fur˚aker (eds). London: Routledge. Boje, Thomas P. and Bengt Fur˚aker 2003 Post-Industrial Labour Markets: Profiles of North America and Scandinavia. London: Routledge. Bommes, Michael and Andrew Geddes 2000 Immigration and Welfare: Challenging the Borders of the Welfare State. London: Routledge. Borjas, George J. 1999 “Immigration and Welfare Magnets.” Journal of Labor Economics 17 (4, Part 1) (October): 607–37. Buchel, ¨ Felix and Joachim R. Frick 2003 “Immigrants’ Economic Performance across Europe: Does Immigration Policy Matter?” EPAG Working Paper 2003–42. Colchester: European Panel Analysis Group, University of Essex. www.iser.essex.ac.uk/epag/pubs/workpaps/pdf/2003-42.pdf Craig, Gary 1999 “‘Race,’ Social Security and Poverty,” in Introduction to Social Security: Policies, Benefits, and Poverty, John Ditch (ed). London: Routledge, 206–26. Ditch, John (ed) 1999 Introduction to Social Security: Policies, Benefits, and Poverty. London: Routledge.

Europe’s varying welfare regimes

197

Esping-Andersen, Gøsta 1990 The Three Worlds of Welfare Capitalism. Cambridge: Polity Press. Faist, Thomas 1996 “Immigration, Integration and the Welfare State,” in The Challenge of Diversity: Integration and Pluralism in Societies of Immigration, Rainer Baubock, ¨ Agnes Heller, and Aristide R. Zolberg (eds). Aldershot: Avebury, 227–50. Faist, Thomas 1998 “Immigration, Integration und Wohlfartsstaaten. Die Bundesrepublik Deutschland in vergleichender Perspektive,” in Migration in nationalen Wohlfahrtsstaaten: theoretische und vergleichende Untersuchungen, Michael Bommes and Jost Halfmann (eds). Osnabruck: ¨ Universit¨atsverlag Rasch. Ferrera, Maurizio 1996 “The ‘Southern Model’ of Welfare in Social Europe.” Journal of European Social Policy 6(1): 17–37. Forster, ¨ Michael 1994 “Family Poverty and the Labor Market.” Luxembourg Income Study Working Paper No 114. Syracuse, NY: Center for Policy Research, Syracuse University, July. Gallie, Duncan (ed) 2004 Resisting Marginalization: Unemployment Experience and Social Policy in the European Union. Oxford: Oxford University Press. Gallie, Duncan and Serge Paugam (eds) 2000 Welfare Regimes and the Experience of Unemployment in Europe. Oxford: Oxford University Press. Goodin, Robert E., Bruce Headey, Ruud Muffels and Henk-Jan Dirven 1999 The Real Worlds of Welfare Capitalism. Cambridge: Cambridge University Press. Goul Andersen, Jørgen and Per H. Jensen (eds) 2002 Changing Labour Markets, Welfare Policies and Citizenship. Bristol: Policy Press. Guiraudon, Virginie 2000 Les politiques d’immigration en Europe. Paris: L’Harmattan. Korpi, Walter and Joakim Palme 1998 “The Paradox of Redistribution and Strategies of Equality: Welfare State Institutions, Inequality, and Poverty in the Western Countries.” American Sociological Review 63(5): 661–87. Leibfried, Stephan 1992 “Towards a European Welfare State? On Integrating Poverty Regimes into the European Community,” in Social Policy in a Changing Europe, Zsuzsa Ferge and Jon Eivind Kolberg (eds). Frankfurt-am-Main: Campus Verlag, 245–79. McGinnity, Frances 2004 Welfare for the Unemployed in Britain and Germany: Who Benefits? Cheltenham: Edward Elgar. Modood, Tariq and Richard Berthoud (eds) 1997 Ethnic Minorities in Britain: Diversity and Disadvantage. London: Policy Studies Institute. OECD 2000 Employment Outlook 2000. Paris: OECD. Platt, Lucinda 2002 Parallel Lives? Poverty among Ethnic Minority Groups in Britain. London: Child Poverty Action Group (CPAG). Ringen, Stein 1987 The Possibility of Politics: A Study in the Political Economy of the Welfare State. Oxford: Clarenden Press. Sarfati, Hedva and Giuliano Bonoli (eds) 2002 Labour Market and Social Protection Reforms in International Perspective: Parallel or Converging Tracks? Aldershot: Ashgate. Saunders, Peter 1994 “Immigrants and the Distribution of Income: National and International Comparisons.” Luxembourg Income Study Working Paper

198

Immigration and the Transformation of Europe

No 123. Syracuse, NY: Center for Policy Research, Syracuse University, December. Schierup, Carl-Ulrik 1993 P˚akulturens slagmark: mindretal og størretal om Danmark. Esbjerg: Sydjysk Universitetsforlag. Smeeding, T. M. 2006 “Government Programs and Social Outcomes: The United States in Comparative Perspective,” in Poverty, Public Policy, and the Distribution of Income, A. Auerbach, D. Card and J. Quigley (eds). New York: Russell Sage Foundation, 149–218. Soysal, Yasemin Nuhoglu 1994 Limits of Citizenship: Migrants and Postnational Membership in Europe. Chicago, IL: University of Chicago Press. Systeme d’Observation Permanente sur les Migrations (SOPEMI) 2004 Trends in International Migration: SOPEMI – 2003 Edition. Annual report, 2003 edn, OECD Emerging Economies. Paris: OECD.

Survey year

2000

2000

2000

2000

1999

2000

Country

Canada

Germany

Norway

Sweden

United Kingdom

United States

Current population survey

Family resources survey

Survey of consumer finances German socioeconomic panel Income and property distribution survey Income distribution survey

Survey name

Head immigration status

Ethnicity/nationality of head; Immigration status of head

D8; Immigrhd

Immigrhd

Ethnicity/nationality of head

D8

Ethnicity/nationality of head

Ethnicity/nationality of head

D8

D8

Head immigration status

LIS variable description

Immigrhd

LIS variable

Figure 7.A.1 Luxembourg Income Study data

Appendix

N=

5580

1094

1567

774

836

1331

Nationality; German nationals; All those with other nationality than German First generation immigrant without Norwegian background; Secondgeneration immigrant; Non-immigrant background Date of immigration; Earliest date is 1945, latest date is 1999; Naturalized immigrants are included in the immigrant group; Non-immigrants White; Black Caribbean; Black African; Black Other; Bangladeshi; Pakistani; Chinese; Other Native-born in the US, US territory, foreign-born to US parent; Naturalized citizen foreign born; Non-citizen

Immigrant background; Non-immigrant

Possible responses/Notes

Content of means-tested benefits

Earned Income Tax (EITC); Temporary assistance for needy families (TANF); Refugee assistance programs; Emergency assistance programs; Low income home energy assistance program (LIHEA); Food stamps; Supplementary security income

Income support; Family credit; Housing benefits; Value of free school meals and milk/welfare milk

Social assistance (socialbidrag); Housing benefits (bostadsbidrag); Introductionary compensation for refugees

Subsistence allowance; Arbeitslosenhilfe; Hilfe in Besonderem Lebenslagen; Wohngeld; Education assistance Social assistance; Housing benefits

Social assistance and Provincial income supplements

8

How different are immigrants? A cross-country and cross-survey analysis of educational achievement1 Sylke Viola Schnepf Introduction

Migration plays an important and growing role in contributing to the population growth in most OECD (Organization for Economic Cooperation and Development) countries (see editors’ introduction and chapter 1 in this volume). This chapter looks at an under-examined and crucial factor for immigrants’ successful integration into the labour market and society: their education. The first aim of this chapter is to analyze how immigrants differ from natives regarding educational outcomes and its determinants. Ten countries with a share of the foreign-born in the total population similar or greater than 10 percent (SOPEMI 2004) are examined: Australia, Canada, France, Germany, The Netherlands, New Zealand, Sweden, Switzerland, the United Kingdom, and the United States. Literature investigating educational disadvantages of immigrants focus generally on educational attainment data that capture progression up national educational systems (e.g. van Ours and Veenman 2003; Riphahn 2003; and Crul, chapter 9 in this volume). This chapter examines educational achievement data that refer to educational outcomes like ability or ‘functional literacy’ (the ability to function in modern society). Educational achievement can be compared more easily across countries than educational attainment. First, educational attainment partly reflects countries’ institutional differences in how education is organized. Second, the educational attainment data of immigrants collected by different countries are not directly comparable, since the definition of what an immigrant is often differs between countries. It is also notable that a specific degree used as a measure of educational attainment might 1

This chapter first appeared as a paper written for the conference on “Immigration in a Cross-National Context: What are the Implications for Europe?” Luxembourg, June 19–23, 2004. I am grateful to the conference organizers and participants for comments and especially to Endre Sik and my discussant, Vincent Tinto. Many thanks are due to John Micklewright for numerous ideas and suggestions deriving from joint work and comments on the chapter.

200

How different are immigrants?

201

mirror different ability for immigrants and natives even in the same country. In general, immigrants have less information about their host countries’ educational system. Hence, their chances to reach a specific educational attainment are not only dependent on their general ability, but also on their integration into the host country. The focus on educational achievement overcomes these problems and is only seldom applied in the literature.2 There are several recent international surveys of learning achievement available for examining what pupils actually know or can do. Which achievement survey should be used for this analysis? Typically, each survey is analysed in isolation with no consideration as to whether its results support or contradict those from another. But there is ample reason to assume that the choice of the survey impacts upon observed results (Brown et al. 2005) since – as will be discussed below – surveys differ considerably in their design. As a consequence, the second aim of this chapter is to pull together the evidence from three different surveys to see if a robust picture exists on immigrants’ educational disadvantage in the ten high immigration countries. There are several factors that impact upon differences in educational achievement between immigrants and natives. These factors motivate the research interest of this chapter, but also show its limitations. First, educational achievement per se is greatly determined by pupils’ socio-economic background. Hence, this chapter examines how far differences between immigrants and natives in terms of their family background determine achievement differences between both groups of children. Second, educational achievement differences can also derive from immigrants’ problems of integration into the host country. Since these problems are likely to be related to the time spent in the country, this analysis focuses on two different kinds of immigrant pupils: (a) first-generation immigrants who were born abroad, and (b) second-generation immigrants who were born in the host countries. The ability of the immigrant pupil to communicate in the language of the host country is a further crucial aspect of achievement and is consequently taken into account in the 2

One recent exemption is Entorf and Minoiu (2005) using exclusively PISA data for comparing achievement differences between immigrants and natives across nine countries. Compared to this work, some contributions of my chapter are to test the robustness of survey results by using PISA, TIMSS and PIRLS data, the application of a wide range of transparent pupils’ background factors for estimating immigrants’ educational disadvantage and the investigation of the impact of school segregation.

202

Immigration and the Transformation of Europe

investigation of immigrants’ educational disadvantage. Another important factor might be school segregation – the uneven distribution of immigrants across schools. Pupils are likely to be influenced by their peers’ school ambitions and abilities and these might differ between schools with high and low shares of immigrants. This provides the motivation to estimate additionally the impact of school segregation on educational achievement gaps between immigrants and natives in this chapter. Finally, even if the socioeconomic background between immigrants and natives were equal, and immigrants were integrated successfully into the society, the process of selection impacting upon the characteristics of migrants living in a host country might still mould achievement differences between natives and immigrants. On the one hand, a selection through immigration control takes place. In traditional countries of immigration like Australia, New Zealand, and Canada, immigration policies try to attract highly-skilled immigrants (Inglis 2004; Ray 2002; Bedford 2003). These immigrants are very likely to differ in their motivations and expectations from immigrants in the former ‘guest worker’ countries like Switzerland and Germany (Castles and Miller 2003). On the other hand, a self-selection of immigrants takes place since immigrants might have very different characteristics depending on the host country they decide to live in and their country of origin. Turkish immigrants deciding to live in Canada might be very different from Turkish immigrants living in Germany. Asian immigrants differ greatly from Mexican immigrants in the United States (Schmid 2001; Glick and White 2003). It is possible to take these selection effects at least partly into account by focusing on the issue of socio-economic background discussed above. However, to some degree selection is related to unobservable characteristics of immigrants that cannot be addressed in this chapter, since information on immigrants’ country of origin is not available in the data sources used. Nevertheless, a focus on educational dispersion between immigrants can give an impression of how far immigrants differ from each other regarding their educational achievement. Chapter 9, by Maurice Crul, offers an alternative design that does capture some variation with the country of origin. This chapter is structured as follows. The next section introduces the three surveys the analysis draws on: the Trends in International Math and Science Study (TIMSS), the Programme of International Student Assessment (PISA) and the Programme of International Reading Literacy Study (PIRLS). Then the following section on educational achievement gaps examines how different immigrants are from natives in terms of their educational outcome and compares educational

How different are immigrants?

203

dispersion between immigrants and natives across countries and surveys. The following section will then apply a regression framework to estimate how far observed immigrants’ educational disadvantage can be explained by their socioeconomic background and school segregation. The final section concludes. Data Table 8.1 lists the educational achievement surveys3 and their sample size of ten countries selected for this analysis due to their high share of immigrant pupils and their participation in at least two of the surveys. All three data sources relate to children in compulsory schooling and are recent pertaining to 1995, 1999, 2000, and 2001. While PIRLS focused on primary school children aged nine to ten years, TIMSS and PISA covered children in secondary school. PISA data examined Table 8.1 Educational achievement surveys and sample sizes of immigration countries covered TIMSS 1999/1995

PISA 2000

PIRLS 2001

Testing of

8th graders

15 year-olds

4th graders

Subjects covered

Mathematics, Science

Reading, Mathematics, Science

Reading

Australia Canada France Germany Netherlands New Zealand Sweden Switzerland UK USA

12,852 16,581 5,763 4,084 6,867 8,855 11,722 9,091 10,973

5,176 29,687 4,673 5,073 2,503 3,667 4,416 6,100 9,340 3,846

8,253 3,538 7,633 4,112 2,488 6,044 5,873 3,763

Source: TIMSS 1999, 1995; PISA 2000; PIRLS 2001. Note: TIMSS and PIRLS are organized by the International Study Center, Boston College, USA and PISA by the OECD. TIMSS data for USA refer to 1999, for all other countries to 1995. Data for UK in TIMSS and PIRLS refer to England and Scotland only. Data on The Netherlands for PISA might be biased due to low response rates. PISA 2000 sample sizes refer to the subject reading. 3

Details on the surveys can be found in their reports: Mullis et al. (2000); Mullis et al. (2003); and OECD (2001).

204

Immigration and the Transformation of Europe

pupils who are fifteen years old, TIMSS covered seventh and eighth graders (the following analysis uses eighth grader data only). The sample designs involve the selection of a sample of schools and then a single class (TIMSS and PIRLS) or a random sample of pupils within each school (PISA). PISA, TIMSS, and PIRLS differ considerably in terms of their design, so the choice of the survey for examining the achievement gap of natives and immigrants might impact upon the results. The three surveys assessed different types of achievement, covered diverse subjects and collected information differently. PISA assessed ability in reading, science and math with the aim to measure broad skills, trying to look at how students would be able to use what they have learned in “real-life situations.” PIRLS measured the ability of primary school children to read and understand written texts. In contrast to PISA and PIRLS, TIMSS focused on assessing a mastery of internationally agreed curricula in the subjects of math and science. PISA and PIRLS assessed achievement predominantly by using openended questions. In contrast, about two-thirds of the TIMSS questions were multiple choice in 1999. Since on average immigrants have lower language skills than their native counterparts, achievement gaps are likely to be lower between native and immigrant pupils with the ability measure of TIMSS which uses multiple choice questions on technical curriculumbased subjects like mathematics and science. The surveys also differ regarding their application of item response models that are used by organizers to aggregate respondents’ answers into achievement scores. Survey organizers do not report the sensitivity of results to the choice of item response model, but Brown et al. (2005) show with TIMSS data that the use of different item response models has an impact on the extent of educational inequalities observed in countries. Other differences between surveys can be cited, including response rates. Even the basic premise that culturally-neutral questions can be successfully designed and translated into different languages can be debated, with the problems in this area probably varying from survey to survey. (Given that survey organizers claim their assessment measure is not culturally biased, it is assumed in the following that achievement differences observed between immigrants and natives do not derive from misunderstandings of cultural contents.) In short, there seems ample reason for comparing results across the different surveys rather than relying on a single source. Table 8.1 shows that PISA covered all ten countries with a high immigration background, while Australia and Switzerland did not participate

How different are immigrants?

205

in PIRLS. TIMSS covered all ten countries, but the variable used for the calculation of pupils’ immigration status was not administered in France. TIMSS data of 1999 are used only for the United States, while data for all other countries refer to 1995, as countries were either not covered in the 1999 round, or variables used for the analysis were only available for the 1995 round. In all three surveys, information on immigration status was collected in the pupils’ questionnaire with identical questions about whether pupils, their mothers and fathers were born in the test country or abroad, and how often the language of the test country is spoken at home. For the purpose of this chapter, immigrants are defined as pupils with both parents born abroad. First-generation immigrants are pupils who were born abroad, while second-generation immigrants were born in the test country. Children who are not immigrants and who have consequently at least one parent born in the test country are referred to as natives. In general, estimates of percentages of pupils with immigration background are relatively similar between surveys (see summary statistics in Tables 8.A.3–8.A.5 in the Appendix). An exception is the United States with a share of immigrant pupils of 12 and 14 percent for PISA and TIMSS (pupils in secondary school) and 20 percent for PIRLS (primary school pupils). Educational achievement gaps between immigrants and natives Is immigrants’ educational achievement lower than that of natives and how do different countries compare? Figure 8.1 presents the percentage of natives (x-axis) and immigrants (y-axis) who do not reach the international TIMSS math median. These children are defined as “unable to apply basic mathematic knowledge in straightforward situations” by the organizers. (Obviously these classifications are open to debate.) For each country two values are given on the y-axis, one value for first-generation and one for second-generation immigrants. In Switzerland, Sweden, Germany, and the United States, the share of first-generation immigrants lacking basic math skills is significantly higher than that of second-generation immigrants, indicating unsurprisingly that pupils who grew up in the host country are achieving better than those who arrived later. However, in The Netherlands and New Zealand the effect is reversed: a lower share of newly-arrived immigrants than those born in the country cannot solve basic mathematic tasks. PISA math achievement scores show the same counter-intuitive result, but only for New Zealand. This result for the Netherlands and New Zealand might

206

Immigration and the Transformation of Europe

0.6

USA

0.5

DEU

Percentage of immigrants

NZL CHE

0.4

SWE

GBR

NLD 0.3 AUS

CAN

0.2

0.1

Second-generation immigrants First-generation immigrants

0 0

0.1

0.2

0.3

0.4

0.5

0.6

Percentage of natives

Figure 8.1. Percentage of natives and immigrants not able to solve basic math tasks in TIMSS. Source: TIMSS 1999, 1995.

be explained by the much higher level of parental education for firstgeneration than for second-generation immigrants.4 Immigrants in Canada, Australia and the United Kingdom are situated around the 45-degree line showing that the percentage of immigrants and natives achieving below TIMSS math international median are similar in these countries. However, the most surprising result of Figure 8.1 is the very large difference between natives and immigrants not able to solve basic math tasks in some of the countries examined. In Germany 4

Given TIMSS data, 46 percent of the parents of second-generation immigrants in New Zealand have completed secondary education compared to 65 percent of parents of firstgeneration immigrants. The trend is smaller in The Netherlands with 21 percent of second- and 29 percent of first-generation immigrants’ parents who completed secondary schooling. Similar results for The Netherlands are found by van Ours and Veenman (2003). However, also in Australia and Germany, the parental education of newly-arrived immigrants is better than that of immigrant pupils born in those countries.

How different are immigrants?

207

and Sweden 20 and in Switzerland even 30 percent points more firstgeneration immigrants than natives fall below the international median of TIMSS math achievement. Expressed differently, one-and-a-half times more first-generation immigrants than natives in the United States, Germany, and The Netherlands, about twice more in Sweden and three times more in Switzerland are unable to solve basic math tasks. In PISA, children who score beneath a critical benchmark level of competence (the reading literacy level 2) are viewed as “unable to solve basic reading tasks such as locating straightforward information.” Applying this survey, organizers’ benchmark to pupils with different migration background (without differentiation between first- and second-generation immigrants) shows that the group of immigrants unable to solve basic reading tasks is about 10 percent points higher than the group of natives in the United Kingdom, France, and United States, 20 percent points higher in the Netherlands, Switzerland and Sweden and even 30 percent points higher in Germany. In line with TIMSS results, the percentage of pupils with low PISA reading ability is similar for natives and immigrants in Canada, Australia, and New Zealand. Up to now percentages of immigrant and native pupils falling below an absolute educational benchmark were compared. The disadvantage of this approach is that pupils are divided solely in two groups: those above and those below the international benchmark. The advantage is that it is possible to attribute a meaning in the form of a statement about ability to students falling below the benchmark. Another way for reporting educational achievement results is to focus on average achievement scores. However, educational achievement scores lack a natural metric and are therefore difficult to interpret. In general, pupils in all participating countries taken together have a mean score set to 500 and a standard deviation of 100 in TIMSS, PISA, and PIRLS. But scores cannot be directly compared between surveys since each survey covers a different set of countries. Nevertheless, TIMSS data can give us a reasonable tool at hand for interpreting differences in mean achievement scores between immigrants and natives. TIMSS tested seventh and eighth graders in 1995. On average across the ten countries examined, eighth graders show roughly a 30 point higher achievement in TIMSS math and fare 40 points better in TIMSS science than seventh graders. Hence, the native-immigrant achievement gap can be expressed in years of grade progression indicating how many years of schooling immigrants lack compared to natives. Table 8.2 displays the point differences in average achievement scores between immigrants and natives for the three surveys and six ability measures. Countries are ordered by achievement gaps in PISA reading.

208

Immigration and the Transformation of Europe

Table 8.2 Differences in average scores between native and immigrant pupils by country and measure PISA

Switzerland Germany Netherlands Sweden France USA UK New Zealand Canada Australia

TIMSS

Read

Math

Science

Math

Science

−83.6 −82.3 −77.5 −57.8 −46.9 −37.7 −33.6 −27.4 −10.9 −9.2

−83.8 −80.0 −89.7 −62.9 −43.8 −38.4 −36.7 −13.2 −9.3 −3.9

−83.2 −91.3 −99.9 −58.1 −65.4 −38.8 −35.0 −24.2 −21.3 −10.2

−57.4 −39.8 −32.0 −33.9

−85.0 −78.8 −49.2 −61.7

−30.7 −0.7 −10.9 −14.0 −3.0

−52.2 −19.1 −34.6 −35.0 −17.0

PIRLS Read −51.4 −42.4 −43.7 −30.4 −31.0 −33.3 −4.7 −17.8

Source: PISA 2000; TIMSS 1995, 1999; PIRLS 2001, author’s calculation. Note: Countries are ordered by mean achievement differences in PISA reading; bold printed figures mean that achievement differences between natives and migrants are significant at the 1 percent level. Immigrant pupils are pupils whose parents were both born in a foreign country; native pupils refer to all other children. Results between TIMSS data of 1995 and 1999 are relatively similar (TIMSS 1999 data could not be used for many countries since the newer data source lacks information on student background).

Negative figures printed bold indicate that immigrants’ achievement is significantly worse (1 percent level) than that of natives. In almost all countries and surveys, immigrants achieve significantly lower test scores than natives. Switzerland, Germany, and The Netherlands appear to have the greatest differences between both groups of pupils throughout most of the surveys’ measures. In these three countries, TIMSS results show that pupils whose parents were born abroad lack about one year of schooling in math and almost two in science compared to natives. Immigrants in Sweden fare better regarding their educational disadvantage in PISA than these three countries, but are similarly bad in TIMSS and PIRLS. France, the United States, and the United Kingdom have lower but still moderate achievement differences between immigrants and natives. But in Canada, Australia, and New Zealand, the immigrant population faces only low or no significant educational disadvantages compared to the native population. Survey and subject results show a relatively consistent ranking of countries on achievement differences between immigrants and natives. The correlation coefficients between all surveys given in Table 8.A.1 in the Appendix indicate the high agreement between surveys on immigrants’

How different are immigrants?

209

disadvantage in countries. The lowest correlation coefficient between surveys of 0.67 is for achievement gaps in PIRLS reading and TIMSS science. We might expect that immigrants’ educational disadvantage is smaller in subjects where language skills are generally of a lower importance like in math. However, achievement in math in PISA is measured by applying the “life-skill” approach related to open-ended questions on wordy descriptions of “real life” situations. Hence, it is not necessarily surprising that immigrants do not fare better in PISA math than in PISA reading. This result stands in contrast to TIMSS: for all countries immigrants’ educational achievement gap is significantly lower (1 percent level) in TIMSS math than in TIMSS science. Immigrants’ educational disadvantage seems to tend to be smaller in technical subjects as long as achievement is assessed in a more curriculum-based approach by using predominantly multiple-choice questions (as is done in TIMSS). To this point the discussion of immigrants’ educational disadvantage has treated pupils with parents born abroad as a more or less homogenous group. But immigrant pupils’ achievement can differ greatly depending on their parents’ country of origin, ethnicity, and socioeconomic background (among other things). Some literature shows that Asian pupils outperform natives in the United States (Glick and White 2003), and Chinese and Indian pupils are better achievers than their English counterparts (Demie 2001). Even though the survey data do not allow for this specification, it is possible to examine differences between immigrants by investigating the variation in their achievement. A focus on immigrants’ achievement distribution compared to that of natives can also shed light on where achievement differences between natives and immigrants derive from in different countries. Is immigrants’ average achievement low because immigrants generally perform worse than natives? Or can immigrants’ educational disadvantage be explained by a great share of low paired with a small share of high achieving immigrants? It is interesting to note that in all ten countries for PISA reading and in seven countries (of nine) for TIMSS math immigrants’ dispersion in terms of their educational achievement is higher than that of natives (author’s calculation based on the standard deviation of mean achievement scores, results not shown). In order to compare the achievement distribution of immigrants and natives, scores for each ventile of the distribution were calculated separately for both groups in the ten countries. Figure 8.2 presents the ratio of the PISA reading scores (immigrants’ score divided by natives’ score) for each percentile and country.

210

Immigration and the Transformation of Europe 1

Ratio achievement natives/immigrants

0.95

0.9

0.85

0.8

0.75

0.7 0

10

20

30

40

50

60

70

80

90

100

Percentile AUS

CAN

NZL

GBR

SWE

NLD

DEU

CHE

FRA

USA

Figure 8.2. Ratio of achievement scores of immigrants to natives by percentile for PISA reading. Source: PISA 2000, author’s calculation. Note: A ratio of 0.75 in the 5th percentile for Germany indicates that immigrants in the 5th percentile achieve a PISA reading score that is 25 percent lower than that of natives in the 5th percentile.

Throughout all countries and percentiles the ratio is smaller than 1, indicating that immigrants’ achievement is worse than that of natives in all percentiles. Furthermore, for all countries, immigrants’ achievement disadvantage is always more pronounced at the bottom than at the top of achievement distributions. However, countries differ greatly regarding immigrants’ achievement gaps across the distribution. Not surprisingly, countries where immigrants’ educational achievement is similar to that of natives (Australia, Canada, and New Zealand) appear to have an achievement ratio closer to 1 across all percentiles than countries with greater average achievement gaps like The Netherlands, Germany, and Switzerland (see country ranking in Table 8.2). In Australia and Canada, immigrants at the bottom of the achievement

How different are immigrants?

211

distribution (5th percentile) show only about 5 percent lower reading achievement scores than natives in the same percentile. On the other hand, in Switzerland, Germany, and The Netherlands immigrants score about 25 percent lower than natives at the left tail of the achievement distribution. Notable is the steep decline in achievement differences between natives and immigrants between the 5th and 15th percentile in New Zealand, the United States, and United Kingdom. While in these countries immigrants in the 95th percentile do not differ greatly from their native counterparts, pupils with parents born abroad show almost a 15 percent lower achievement than natives once we focus on achievement ratios at the bottom of the distribution. On the other hand, the achievement ratio shrinks only marginally for increasing percentiles in France. There are also great differences at the top of the distribution. The best performing immigrants achieve similarly to best ranked natives in Australia, Canada, New Zealand, the United Kingdom, and the United States. This stands in contrast to The Netherlands, Germany, and Switzerland, where even immigrants with the best abilities reach scores that are still about 15 percent lower than those of natives at the top of the distribution. Taken together, the surveys show relatively consistent results on educational achievement gaps between immigrants and natives. In traditional countries of immigration like Australia, Canada, and New Zealand, immigrants do not appear to be greatly different from natives. On the other hand, in European countries like The Netherlands, Germany, and Switzerland immigrants fare significantly worse than natives. These average achievement differences are great in magnitude as a comparison to achievement differences between two adjacent grades and absolute disadvantage showed. PISA results suggest that in countries where immigrants’ educational achievement is similar to that of natives, immigrants at the bottom of the achievement distribution show only slightly worse performance than their native counterparts. In countries with high immigrants’ educational disadvantage, achievement differences between immigrants and natives are large – especially at the bottom of the achievement distribution. Determinants of immigrants’ achievement Can a socioeconomic status (SES) gap between immigrants and natives explain the large educational disadvantage immigrants appear to face in some countries? This question is of considerable policy importance: if immigrants’ achievement is lower only due to their lower average SES, educational policies would just need to provide additional

212

Immigration and the Transformation of Europe

support for children with a disadvantaged family background. However, if immigrants face additional barriers besides SES, countries’ educational policies need to adapt to immigrants’ special needs to decrease their educational disadvantage. One great concern regarding educational policies might also be school segregation as a measure of countries’ capacity to integrate their immigrant population equally and successfully into the schooling system. To what degree is school segregation important for explaining educational achievement? Table 8.3 investigates these issues separately for the ten high immigration countries and presents OLS (Ordinary Least Squares) regression results for three different models for PISA math, TIMSS math and PIRLS reading. The focus on math has the advantage of standardizing for subjects between the surveys with children of similar ages in secondary school. In all regression models the achievement score is the dependent variable. The table reports only the estimated effects of those explanatory variables related to immigration background. Native children are the base category. The symbol “o” denotes that the coefficient of the variable is not significant at the 10 percent level, coefficients without asterisk are significant at the 1, with one asterisk at the 5 and with two at the 10 percent level. Coding of variables used in the models, summary statistics of these variables by country and the whole regression results for Model 2 for all three surveys are given in the Appendix (Tables 8.A.2–8.A.8). First-generation and second-generation immigrants and language skills (Model 1) The aim of Model 1 is to add to the unconditional results presented above by examining educational disadvantage for different types of migrant pupils. This regression model uses only three dummies on immigrant status as independent variables: first-generation immigrants, secondgeneration immigrants and pupils living in a home where the language spoken differs from that of the host country. Coefficients of the three explanatory variables are presented in Model 1 of Table 8.3. The base pupil is native and speaks the test countries’ language at home. Language spoken at home does not only identify immigrants but also other children from different ethnicities.5 An interaction variable capturing immigrants who speak another language at home was added to the model. It was generally not significant. This indicates that there are 5

For PISA math in general more than 75 percent of children who speak another language at home than that in the test country are immigrants. However, in Canada and The Netherlands the share is much smaller (58 percent and 27 percent).

How different are immigrants?

213

usually no differences between children with immigration and ethnic backgrounds regarding the impact of the language spoken at home on educational achievement. In all surveys and countries, children who speak a foreign language at home receive lower achievement scores than other children. This effect is consistently greatest in Germany with as many as 76 points in PISA and 44 points in TIMSS. However, foreign language appears to impact less on achievement in Canada, Sweden, and France and only marginally in The Netherlands where the coefficients are insignificant (TIMSS) or below 20 points (PISA and PIRLS). To help interpret the coefficients, it is worth bearing in mind that the standard deviation in mean achievement is close to 100 for countries and that progression in average scores between seventh and eighth grades in TIMSS math is on average 30 points. In general, PISA and TIMSS results are quite consistent for secondary school pupils’ math achievement, while PIRLS results on immigrants’ reading achievement in primary school partly stand out. Where immigration status is concerned, the regression results show substantial variation between countries, as the unconditional analysis revealed before. In PISA and TIMSS, immigration status does not have a significant negative impact on achievement in Australia and Canada. However, as unconditional results discussed above revealed, immigrants face an educational disadvantage in the United Kingdom and the United States. Nevertheless, once language is controlled for, immigration no longer matters in both countries (with the exception of first-generation immigrants in the United States in TIMSS). This indicates that in the United Kingdom and United States, the language skills of immigrants might be the greatest barrier to immigrants reaching the achievement scores of their native counterparts. A similar ‘effect’ appears in Germany, where immigrants’ educational disadvantage was large given unconditional results presented above. But once language is held constant, second-generation immigrants (for PISA and TIMSS) do not differ any more from natives regarding their achievement skills.6 On the other hand, in countries like France, Switzerland, Sweden, and The Netherlands, where the impact of language spoken at home is low, first- and second-generation immigrants still face a significant educational disadvantage (similar to unconditional results). Throughout all surveys and compared to other countries, both types of immigrant pupils face the greatest educational disadvantage in The Netherlands. What conclusions can we draw about the differences between achievement gaps for first- and second-generation immigrants? In general, 6

Based on educational attainment data, Riphahn (2003) receives a contradictory result.

1

PISA

PIRLS

TIMSS

PISA

PIRLS

TIMSS

PISA

Survey

Second-generation First-generation Language Schools’ ISEI Share immigrants

Second-generation First-generation Language Second-generation First-generation Language Second-generation First-generation Language

Second-generation First-generation Language Second-generation First-generation Language Second-generation First-generation Language

Migration

O O −14∗ 2.6 0.30∗∗

O O −14∗∗ O +16 −29

O O −19 O +17∗ −39

AUS

O O O 2.0 −0.31

O O O O O −15∗ +11 −18 −39

O O −10∗ O O −18 +9 −21 −41

CAN

O −31 O 3.0 −1.14

−7∗∗ −18 −25

−16∗ −46 −26

O O −45 4.5 −1.15

O −23∗ −56 O O −39 −13 −32 −31

O −36 −76 O −21∗ −44 −21 −42 −40

−23 −63 −32

−17 −28 −31

DEU

FRA

O +18∗ −43 −21 +23 −24 +17 O −50 O +19∗ −40 2.5 −0.45

−18∗ −49 O 5.2 O

O +25 −51 −24 +24∗ −36 +19 +14∗ −69

NZL

−47 −78 −15∗∗ O −39∗ O −35 −22 −14

−70 −102 −14∗∗ −22∗∗ −53 O −43 −30 −17

NLD

−26 −48 −32 −19 −26 −34

−35 −71 −42 −26 −41 −41

CHE

O O −36 +13∗∗ +19∗∗ −26 +15 −32 −40

O O −48 O O −36 +11∗∗ −45 −47

GBR

O O O O −26 O −29 −33 −25 2.3 2.6 3.6 −0.33∗∗ −0.99 −0.33∗∗

−28∗ −28∗ −31∗ O −28 O −8∗ −29 −23

−37 −45 −31 −12∗∗ −37 −17∗ −16 −43 −28

SWE

+21∗ +27∗ −41 3.7 −0.43∗

O +23∗∗ −57 O O −30

O O −81 O −21 −43 O −39 −48

USA

Source: PISA 2000; TIMSS 1995, 1999; PIRLS 2001; author’s calculations. Note: O denotes that the variable is not significant at the 10 percent level, ∗∗ means significance at 10 and ∗ significance at 5 percent level, numbers without asterisk are significant at 1 percent level. Results of Model 2 are given in Appendix for all three surveys. In PIRLS the variable on education was not administered in the US. The clustering of the students within schools is taken into account for estimating standard errors.

3 Controlled for SES, family structure and immigrants’ distribution in schools

2 Controlled for SES and family structure

Model

Table 8.3 Selection of OLS regression results showing differences in achievement between migrants and natives unconditional (Model 1) and conditional on parental background (Model 2) and on parental background and school segregation (Model 3) for PISA math, TIMSS math (8th graders) and PIRLS; dependent variable pupils’ achievement score

How different are immigrants?

215

when we control for language, second-generation immigrants greatly outperform first-generation immigrants. In only four of the twenty-seven regression results presented in Model 1 did more recently-arrived immigrants perform better than immigrants born in the country. Again New Zealand stands out. First-generation immigrants in this country achieve even better than native students (once language is controlled for) and this result is consistent for TIMSS, PISA, and PIRLS. Socioeconomic background (Model 2) Can the observed achievement differences between immigrants and natives be explained by socioeconomic background? At the country level, compositional differences between immigrants and natives seem to have considerable explanatory power for the countries’ magnitude of immigrants’ educational disadvantage. This is shown in Figure 8.3, which

Differences in mean PISA reading score between natives and immigrants

120 BEL 100 CHE LUX 80

DEU DNK

NLD

AUT

GRC NOR

60

SWE FRA

40

GBR

ITA

ESP

CZE

USA NZL 20 CAN

AUS

0 0

5

10

15

20

25

30

35

40

Percent point difference between natives and immigrants with more than 100 books at home

Figure 8.3. Differences in achievement and socioeconomic background between natives and immigrants in PISA reading. Source: PISA 2000; author’s calculations.

216

Immigration and the Transformation of Europe

shows the strong association between immigrants’ achievement gap and a measure capturing socioeconomic background differences between immigrants and natives: pupils’ estimation of the number of books at home. This measure of SES has the great advantage of being comparable across countries and immigrant populations. (It might be assumed that children’s estimates of books at home are quite unreliable, but the correlation coefficient of 0.93 between the countries’ percentage of pupils with more than 100 books, calculated separately with TIMSS and PISA, shows a very high level agreement between both surveys.) A straightforward way for estimating socioeconomic background differences between immigrants and natives is to calculate the percent point difference between natives and immigrants with more than 100 books at home. Figure 8.3 gives this difference on the x-axis. Obviously, differences between natives and immigrants vary greatly across countries. In Canada and New Zealand, the percentage of native pupils having more than 100 books at home is only 10 percent points higher than that of immigrants. The figure rises to more than thirty-five in Sweden and The Netherlands and is similar to thirty in Switzerland and Germany. The y-axis gives mean educational achievement differences in PISA reading that were reported in Table 8.2 for the ten high immigration countries. Higher SES differences between natives and immigrants are positively related to a higher immigrants’ achievement gap with a correlation coefficient of 0.82 for the ten countries. This result refers only to PISA reading, but other surveys’ outcomes are similar: percent point differences between natives and immigrants with more than 100 books at home are correlated positively with immigrants’ achievement gaps in TIMSS math with 0.85, in TIMSS science with 0.82, and in PIRLS reading with 0.70 for the ten high immigration countries. This result is also robust if parental education instead of books at home is used as an indicator for SES. Nevertheless, regarding this indicator it is noteworthy that an immigrant’s mother who has completed upper secondary education in Mexico and a native’s mother who completed the same degree in the United States probably have a very different quality of education. Thus parental education is not easily comparable between immigrants and natives. Nevertheless, the correlation between parental background (the difference between the share of immigrants and natives whose mothers completed secondary education) and immigrants’ achievement gap is also high for the ten countries: the correlation coefficients are 0.76 for TIMSS math, 0.67 for PISA reading, 0.65 for PIRLS, and 0.62 for TIMSS science. In summary, on the country level, results indicate consistency across surveys and indicators of SES that those countries that display the highest

How different are immigrants?

217

educational disadvantage for immigrants also have a migrant composition with the lowest SES. Does this result imply that countries’ differences in immigrants’ achievement gaps are a mere reflection of immigrants’ socioeconomic status in these countries? Or does immigrant status matter beyond SES? The regression Model 2 of Table 8.3 aims to answer these questions at the individual level, comparing immigrants’ disadvantage across the ten countries once controlled for socioeconomic background. In order to compare immigrants’ achievement gap unconditionally (Model 1) and conditional on SES, Model 2 expands Model 1 by adding control variables for SES. There are a number of variables that could be used to measure socioeconomic background. The main PISA report (OECD 2001) placed considerable emphasis on the association of scores with indices constructed from principal components analysis of a range of parental characteristics, including occupational indices based on the work of Ganzeboom et al. (1992). Obviously these indices cannot be replicated for other surveys that have collected different family background data, so this chapter takes a much simpler approach to make comparisons between the different data sources. This also has the merit of greater transparency with variables that are relatively easy to comprehend. Consistently across surveys, socioeconomic background is taken into account with dummy variables for mothers’ education at the secondary and tertiary level as well as a dummy for children with more than 100 books at home. Furthermore, dummies for children with siblings, children from singleparent families and from “mixed” family structures are included. In addition, I control for the gender of the pupils and the area of the school (rural or urban). Table 8.A.2 in the Appendix summarizes the explanatory variables and their coding. The base group child in Model 2 is a male native pupil without siblings, who speaks the host country’s language at home, whose mother has not completed (upper) secondary education and who lives with both parents in a home with less than 100 books in an urban area. Results in Table 8.3 present only the significant coefficients on the three immigration variables. (See Tables 8.A.6 to 8.A.8 in the Appendix on full results for the three surveys.) Hence, the information in the Table allows to examine how far immigrants’ educational disadvantage changes between Model 1 (unconditional) and Model 2 (conditional on socioeconomic background) and how far countries differ regarding the extent of immigrants’ disadvantage. Unsurprisingly, regression coefficients for Australia, Canada and New Zealand – the three countries with no or only marginal educational achievement gaps of immigrants – do not differ greatly from

218

Immigration and the Transformation of Europe

unconditional results or those presented in Model 1 once we control for socioeconomic status and family structure. Since immigrants’ background is similar to that of natives, immigrants’ achievement also appears to be similar in these countries once parental background, area, and family structure are held constant. In the United Kingdom and United States we see a slight effect given SES controls, with immigrants faring even better than natives. These positive coefficients are always smaller than the coefficient for “language spoken at home” so that only immigrant pupils who speak the language of the host country at home have a marginal educational advantage compared to their native counterparts in both countries. The most interesting results regard countries where immigrants were still different from natives in Model 1. In France, The Netherlands, Sweden, and Switzerland, the impact of immigrant status shrinks greatly once SES is controlled for. Nevertheless, in all four countries first-generation as well as second-generation immigrants still fare significantly worse in PISA math achievement than their native counterparts with similar SES and who speak the test language at home. This result appears to be consistent with outcomes found for PIRLS and predominantly also for TIMSS. In Germany, differences between immigrants and natives in terms of language spoken at home and SES largely explain educational achievement gaps at least for PISA and TIMSS (see also Schnepf 2003). By holding the language variable constant, first-generation immigrants fare still worse in Germany for TIMSS and PISA, but the coefficient of those immigrants shrinks by about one-third in PISA and appears to be insignificant in TIMSS once we control for SES. However, in PIRLS the results indicate that primary school immigrants still face a significant educational disadvantage in Germany, even given controls for language and SES. Taken together, immigrants’ achievement does not differ greatly from that of natives in Australia, Canada, and New Zealand, across surveys and regardless of whether the focus is on unconditional or conditional results controlling for language spoken at home, SES, family structure, and area. A foreign language spoken at home seems to explain a large amount of immigrants’ achievement differences in the United Kingdom and the United States. The picture is different in The Netherlands, France, Sweden, and Switzerland. In these countries, neither SES nor foreign language can explain the total amount of immigrants’ educational disadvantage across all the surveys. Immigrants who do not differ from natives in terms of socioeconomic background and language spoken at home still have poorer educational achievement results than their native

How different are immigrants?

219

counterparts. For all four countries, about 60 percent7 of the achievement gap between immigrants and natives remains unexplained by Model 2 in PISA, while it is about 40 percent for the countries of The Netherlands, Sweden, and Switzerland covered by TIMSS.8 Socioeconomic background and school segregation (Model 3) What might be the reason for immigrants’ relatively high educational disadvantage that persists in some countries even if language skills and socioeconomic background are held constant? Immigrants are not only different from natives regarding their socioeconomic background, but also in terms of their cultural attitudes, social contacts, their schooling ambitions, career planning, and orientation on return migration. Borjas (1999) argues that it is not only parental influences that determine the child’s socioeconomic development, but also “ethnic capital,” with children in particular ethnic groups being exposed to a whole set of important ethnic characteristics. Even though immigrants belong to different ethnicities, as a group they might share a similar “immigrant capital” that is not necessarily uniform, but generally different from that of natives (at least in some countries). Furthermore, if we apply Borjas’ argument on immigrants’ school achievement, it is the exposure to this “immigrant capital” which is likely to impact schooling outcomes. The exposure might be transmitted by “neighborhood” in terms of residential segregation of ethnicities as found in Borjas (1995) or the closely-related factor of school segregation (Rivkin 1994). Here, the term “school segregation” refers to differences in the distribution of immigrants and natives in schools and is conceptually related to the impact of peers’ achievement and learning attitudes on general educational outcome at the school level. An immigrant child in a highlysegregated school with a high percentage of low-performing immigrants is very likely to be pulled to the average of immigrants’ achievement, while the same child (with the same socioeconomic background) integrated in a diverse school with a high percentage of well-achieving natives is less likely to fall behind in educational achievement. 7

8

To receive this result, achievement differences between natives and immigrants were decomposed into an explained and residual component (Oaxaca 1973). The percentage gives the unexplained component as a share of the total difference in achievement between immigrants and natives. While in PISA and TIMSS countries do rather not differ regarding the unexplained share in the achievement gap this is different for PIRLS. With this survey the unexplained achievement gap between natives and immigrants is 35 percent for Sweden, 45 percent for France, 55 percent in Germany, and 69 percent in The Netherlands.

220

Immigration and the Transformation of Europe

School segregation as a measure of conglomeration of immigrant peers might therefore be central for explaining the immigrants’ educational achievement gap. It is also an important factor in broader terms, since it shows how successful countries are in integrating immigrant minorities, and in providing opportunities for them to take advantage of a fruitful school environment equal to that of their native counterparts. How can school segregation be measured? In order to make regression results easily comprehensible, this chapter uses a simple approach. The school’s segregation variable used in the regression framework gives the percentage of immigrants in the school minus the average percentage of immigrants across all schools in the country. Hence, the variable indicates the percentage of immigrants that are over- (positive value) or under-represented (negative value) in a school given the average share of immigrants across schools. This estimation of school segregation is not unproblematic given the sample design of the data available. First, the data should provide representative samples of data on pupils’ immigration status within each school. This criterion is met by PISA, since a sample of about thirtyfive students is randomly drawn from the fifteen-year-olds attending the school. However, this contrasts with the sample design of TIMSS and PIRLS. These surveys select randomly a single whole class within each school. This is a procedure that is likely to result in the data providing a biased estimate of immigration background at the individual school level if, as is common, there is “setting” for the subjects that are tested in the survey (i.e. children separated into different classes according to their ability levels). In this case, the ability levels of the selected class will be more homogenous than among all children of that age in the school. Given the correlation between ability and immigration status, the same can be expected to be true of the latter. Thus I restrict the estimation of the impact of segregation to PISA data. Second, with PISA data the estimation of the share of immigrants in a school is problematic due to the relatively small sample size of pupils (thirty-five) drawn in a PISA school. It can be assumed that any sampling error will be unbiased. But it is noteworthy that the resultant share of overor under-represented immigrants in a school might capture considerable “noise” given small sample sizes. Adding only a variable on immigrants’ representation in schools to Model 2 could lead to an over-estimation of its impact, since schools with high shares of immigrants might be attended by pupils with lower average socioeconomic background than schools with an over-representation of natives. In order to control for this socioeconomic composition, a further variable is added to the model that captures the schools’ average

How different are immigrants?

221

“International Socio-Economic Index of Occupational Status” (ISEI)9 of pupils’ parents. Table 8.3 shows the results of Model 3 giving coefficient for immigrant variables now conditional on pupils’ and schools’ SES and school segregation. For all countries, a higher “socioeconomic position” of the school (variable school ISEI) is also related to a significantly (1 percent level) higher achievement of the pupil attending this school even given controls for pupils’ socioeconomic background. What is the impact of highly segregated schools on educational achievement? In six of the ten countries covered by PISA, pupils in schools with an over-representation of immigrants fare worse even if controlled for pupils’ and schools’ socioeconomic background (at least 5 percent level). Only in one country – Australia – does a higher share of immigrants in the school have a positive impact on achievement results, but the significance level is low (10 percent). In the United States, New Zealand, and Canada children in schools with 50 percent more immigrants than on average across schools (which represents schools with a share of immigrants at about the 95th percentile, see Table 8.A.3 in the Appendix) achieve about 20 points less than their counterparts in schools with a representative distribution of immigrants. The impact of school segregation is two or three times greater in Switzerland, Germany, and France, where achievement decreases by 1 point for each percent point of immigrants who are over-represented in the school. For example, children in schools at the 95th percentile regarding their share of immigrants (65 percent) in Switzerland perform about 45 points worse than pupils in schools with an average share of immigrants of 21 percent. Nevertheless, despite controls for schools’ and pupils’ SES, these point differences need to be interpreted carefully. The educational system in Germany and Switzerland is shaped by a hierarchical school structure. Immigrant students are likely to attend the less prestigious school tracks, where educational achievement in general is lower than in schools with higher prestige. Hence, in both countries the schools’ share of immigrants might be greatly correlated with the school track pupils attend and so we may simply be seeing lower achievement levels in schools at the bottom end of the school hierarchy. Nevertheless, the result on school segregation of Model 3 indicates that both natives and immigrants fare worse if immigrants are overrepresented in the school. To test for this, an interaction variable between the schools’ share of immigrants and migration status was added to 9

The ISEI is based on education, income, and age of parental occupational groups (see Ganzeboom et al. 1992). For summary statistics see Table 8.A.3 in the Appendix.

222

Immigration and the Transformation of Europe

Model 3 (results not shown). This variable was insignificant in all but one case (United States), suggesting that generally the schools’ segregation “effect” impacts negatively on both kinds of pupils, immigrants and natives. The results of Model 3 indicate that immigrants’ lower integration in the host countries’ school environment gains additional explanatory power besides SES for explaining educational outcomes. Nevertheless, the less favorable capital of natives who attend highly-segregated schools might also lower schools’ achievement outcomes. How far can school segregation explain the great share of immigrant educational disadvantage that remained unexplained in some countries in Model 2? In contrast to results for PISA in Model 2, once we control for school segregation and schools’ SES, in three of the four countries with high immigrants’ disadvantage (France, Sweden, and Switzerland), second-generation immigrants no longer fare significantly worse than natives. In addition, point differences for first-generation immigrants shrink by a third (France, Switzerland, and The Netherlands) or fall into insignificance (Sweden). This result indicates that school segregation does matter for explaining immigrants’ lower educational achievement in these countries. Taken together, the regression analysis shed light on variation in achievement between different groups of immigrants and natives (Model 1), the impact of SES on immigrants’ achievement gap (Model 2), and the influence of school segregation (Model 3). TIMSS and PISA results for about fourteen- to fifteen-year-olds appear to be relatively consistent, while PIRLS results on reading achievement of primary school children differ compared to the other two surveys. For Australia, Canada, and New Zealand, once again, the results suggest that immigrants do not differ greatly from natives in educational achievement. Achievement gaps between natives and migrants in the United Kingdom and the United States diminish greatly once we control for the language spoken at home. In Germany, immigrants seem to differ from natives mainly due to their language spoken at home and their lower socioeconomic status. In France, The Netherlands, Sweden, and Switzerland, by contrast, migrant status still matters even with controls for language skills and parental background. Adding a variable to the model that captures the dissimilarity in immigrants’ distribution between schools greatly decreases the significance of the migration variables in these countries. In these last four countries, then, it is socioeconomic background and the higher clustering of immigrants in schools that explains most of the relatively large educational disadvantage of pupils whose parents were born abroad.

How different are immigrants?

223

Conclusion How do immigrants differ from natives regarding educational achievement and are results robust across surveys? Here I reorganize the findings into the three different reasons that the introduction proposed for why we might expect immigrants’ achievement to be different from that of natives. First, immigrants’ socioeconomic background might be different from that of natives. At the country level of analysis, results indicated consistently across surveys that the lower the relative socioeconomic background of immigrants, the bigger the immigrants’ relative educational disadvantage. In Australia, Canada, and New Zealand immigrants’ SES is similar to that of natives, and surveys consistently show no or only marginal educational disadvantage of immigrants for these three countries. In contrast, in The Netherlands, Germany, and Switzerland immigrants’ SES composition is much less favorable than that of natives, and in these countries immigrants fare much worse than natives regarding educational achievement. But in some countries SES factors are insufficient to explain the achievement gaps. In France, The Netherlands, Sweden, and Switzerland immigration still matters even if pupils’ socioeconomic background is held constant. Second, immigrants’ educational disadvantage might derive from their problems of integration into the host country. One aspect of the integration issue is the pupils’ capacity to communicate in the language of the host country. Regression analysis showed consistently across surveys that speaking a foreign language at home greatly decreases pupils’ achievement in all countries compared. Another aspect of integration is the time immigrants live in the host country. Pupils who were born and grew up in the host country (secondgeneration immigrants) are likely to be better integrated into their host society than newly-arrived immigrants (first-generation immigrants). On the aggregated country level, but also by applying the OLS regression framework, results showed that second-generation immigrants fared much better than first-generation immigrants in all countries with the exception of New Zealand. Again, this result was greatly consistent across surveys. The last aspect regarding immigrants’ integration examined was school segregation as a measure of concentration of immigrant peers and as an indicator for how successful countries are in integrating immigrant minorities in the school environment. In Canada, France, Germany, New Zealand, Switzerland, and the United States, pupils in schools with

224

Immigration and the Transformation of Europe

an over-representation of immigrants (compared to the average share of immigrants in schools) fared worse than pupils in other schools, even if pupils’ and schools’ socioeconomic backgrounds were held constant. Obviously, high clustering of immigrants in some schools is neither favorable for the educational achievement of immigrants nor for that of natives attending these schools. (This result, however, is only based on PISA data, since the sample design of the other two surveys did not allow for measuring immigrants’ distribution across schools.) Controlling for school segregation decreased immigrants’ educational disadvantage in France, The Netherlands, Sweden, and Switzerland, filling in much of the variation in educational achievement that socioeconomic background could not explain. Third, the process of selection of immigrants is likely to impact on their achievement results. Immigration control, but also the self-selection of immigrants, determines the characteristics and motivations of immigrants. In this chapter I was unable to examine this selection issue. These data sources lack even the most basic variable for doing so, the immigrants’ country of origin. (Again, see chapter 9, by Maurice Crul for a study that does incorporate this variation.) However, achievement data allow for examining educational dispersion of immigrants and natives separately, indicating how different immigrants are compared to natives. This perspective also emphasizes that within each country “immigrant” contains many different kinds of people. Immigrants generally differ more in their educational outcomes than natives. Immigrants’ achievement at the bottom of their achievement distribution is much lower than that of lowest-performing natives in those countries where immigrants face considerable educational disadvantages. Hence, the worst-achieving immigrants fall even further behind the worst-achieving natives in Switzerland, Germany, and The Netherlands. This result is particularly troubling, suggesting an especially bleak outlook for these poorly-performing immigrants. The results of the analysis lead to some clear policy implications for fostering immigrants’ educational achievement. The promotion of language skills for immigrant students speaking a foreign language is important in all the countries studied here. In countries where socioeconomic background differences are great between immigrants and natives (like Switzerland, The Netherlands, Germany, France, and Sweden), promoting all pupils with unfavorable family background would also greatly benefit immigrants. In addition, decreasing school segregation is likely to have a positive outcome on pupils’ achievement in general, and for some countries it might additionally decrease immigrants’ educational achievement gap. Furthermore, in The Netherlands, Germany, and Switzerland, a

How different are immigrants?

225

general promotion of the lowest low-performing students would greatly help those immigrants who are gathered around the bottom end of the achievement distribution.  Bedford, Richard 2003 “New Zealand: The Politicization of Immigration.” Washington, DC. Accessed 14 June, 2005. Last updated in January 2003. Available from www.migrationinformation.org/Profiles/display.cfm?id=86. Borjas, George J. 1995 “Ethnicity, Neighborhoods, and Human-Capital Externalities.” American Economic Review 85(3) (June): 365–90. Borjas, George J. 1999 “Ethnic Capital.” Chapter 8 in Heaven’s Door: Immigration Policy and the American Economy. Princeton, NJ: Princeton University Press, 146–60. Brown, Giorgina, John Micklewright, Sylke V. Schnepf and Robert Waldmann 2005 “Cross-National Surveys of Learning Achievement: How Robust Are the Findings?” S3RI Applications and Policy Working Paper. Southampton: Southampton Statistical Sciences Research Institute. http:// eprints.soton.ac.uk/16250/01/s3ri-workingpaper-a05-05.pdf. Castles, Stephen and Mark J. Miller 2003 “New Ethnic Minorities and Society.” Chapter 10 in The Age of Migration: International Population Movements in the Modern world. New York: Guilford Press, 220–54. Demie, Feyisa 2001 “Ethnic and Gender Differences in Educational Achievement and Implications for School Improvement Strategies.” Educational Research 43(1) (Spring): 91–106. Entorf, Horst and Nicoleta Minoiu 2005 “‘What a Difference Immigration Policy Makes:’ A Comparison of PISA scores in Europe and Traditional Countries of Immigration.” German Economic Review 6(3) (August): 355–76. ftp://ftp.iza.org/dps/dp1021.pdf. Ganzeboom, Harry B. G., Paul M. De Graaf and Donald J. Treiman 1992 “A Standard International Socio-Economic Index of Occupational Status.” Social Science Research 21(1) (March): 1–56. Glick, Jennifer E. and Michael J. White 2003 “The Academic Trajectories of Immigrant Youths: Analysis within and across Cohorts.” Demography 40(4) (November): 759–83. Inglis, Christine 2004 “Australia’s Continuing Transformation.” Washington, DC. Accessed January 25, 2006. Last updated in August. Available from www.migrationinformation.org/Profiles/display.cfm?id=242. Mullis, Ina V. S., Michael O. Martin, Eugenio J. Gonzalez, Kelvin D. Gregory, Robert A. Garden, Kathleen M. O’Connor, Steven J. Chrostowski and Teresa A. Smith 2000 TIMSS 1999 International Mathematics Report: Findings from IEA’s Repeat of the Third International Mathematics and Science Study at the Eighth Grade. Chestnut Hill, MA: International Study Center, Lynch School of Education, Boston College. http://timss.bc.edu/timss1999i/ math achievement report.html. Mullis, Ina V. S., Michael O. Martin, Eugenio J. Gonzalez and Ann M. Kennedy 2003 PIRLS 2001 International Report: IEA’s Study of Reading

226

Immigration and the Transformation of Europe

Literacy Achievement in Primary School in 35 Countries. Chestnut Hill, MA: International Study Center, Lynch School of Ecuation, Boston College. http://timss.bc.edu/pirls2001i/PIRLS2001 Pubs IR.html. Oaxaca, Ronald 1973 “Male–Female Wage Differentials in Urban Labor Markets.” International Economic Review 14(3) (October): 693–709. OECD 2001 Knowledge and Skills for Life–First Results from PISA 2000. Paris: OECD. www.pisa.oecd.org/dataoecd/44/53/33691596.pdf. Ray, Brian 2002 “Canada: Policy Legacies, New Directions, and Future Challenges.” Washington, DC. Accessed June 14, 2005. Last updated in May. Available from www.migrationinformation.org/Profiles/display.cfm?ID=20. Riphahn, Regina T. 2003 “Cohort Effects in the Educational Attainment of Second Generation Immigrants in Germany: An Analysis of Census Data.” Journal of Population Economics 16(4) (Winter): 711–37. Rivkin, Steven G. 1994 “Residential Segregation and School Integration.” Sociology of Education 67(4) (October): 279–92. Schmid, Carol L. 2001 “Educational Achievement, Language-Minority Students, and the New Second Generation.” Sociology of Education 74: 71–87. Schnepf, Sylke V. 2003 “Inequalities in Secondary School Attendance in Germany.” S3RI Applications and Policy Working Paper No A03/16. Southampton: Southampton Statistical Sciences Research Institute. http://eprints.soton.ac.uk/212/. Systeme d’Observation Permanente sur les Migrations (SOPEMI) 2004 Trends in International Migration: SOPEMI – 2003 Edition. Annual report, 2003 edn. OECD Emerging Economies. Paris: OECD. van Ours, Jan C. and Justus Veenman 2003 “The Educational Attainment of Second Generation Immigrants in The Netherlands.” Journal of Population Economics 16(4) (Winter): 739–54 www.uvt.nl/osa/producten/ wop/oswp01 11.pdf.

Appendix Table 8.A.1 Correlation coefficients of differences in achievement scores between natives and immigrants between surveys PISA

PISA

TIMSS PIRLS

Reading Math Science Math Science Reading

TIMSS

Reading

Math

Science

Math

Science

1 0.98 0.96 0.85 0.85 0.84

1 0.97 0.82 0.80 0.89

1 0.80 0.78 0.80

1 0.97 0.69

1 0.67

PIRLS Reading

1

Source: PISA 2000; TIMSS 1995, 1999; PIRLS 2001; author’s calculations. Note: Correlation coefficient is based on all observations that are not missing in the two surveys correlated.

FT (Family type)

Area

School ISEI

Percent of immigrants in school minus average percent of immigrants in countries’ schools

2, 3

2, 3

3

3

(Control group: mother did not complete secondary education) 1 = mother completed at least upper secondary education, 0 = rest 1= mother completed tertiary education, 0 = rest 0 = data available, 1 = data missing 0 = child without siblings, 1 = other 1 = child raised in single parent family, 0 = rest 1 = other family type than single or nuclear family (control group), 0 = rest 0 = urban or suburban, 1 = rural 0 = data available, 1 = data missing Continuous, schools’ average International Socio-Economic Index of Occupational Status Continuous, negative values give percent of immigrants under-represented and positive values give percent of immigrants over-represented in school given national average of immigrants in school

(Mother has education below upper secondary) Mother above (upper) secondary education

Share of immigrants in school

Area RegMis, Location missing Pupils’ average ISEI in school

Mother tertiary education Education missing (edumis) Sibling Single parent Other family type

Books in household

1=parents and pupil born abroad, 0=rest 1=parents born abroad, pupil born in test country, 0=rest 1=(almost) always foreign language, 0=rest Boys = 0, girls = 1 0 = 0–100 books, 1 = more than 100 books

First-generation immigrant Second-generation immigrant Language spoken at home

Math test score Math test score Reading test score

Coding of variable

Note: “Model” indicates in which regression model variables were used (see Table 8.3). In PIRLS and PISA mothers’ education refers to the completion of upper secondary education, in TIMSS to the completion of secondary education. In PISA and TIMSS single parent families refer to children who live with only one of the following guardians: mother, father, male guardian or female guardian. In PIRLS single parent family refer to families where children live solely with one adult. In this survey information on other family types is not available.

Gender SES (Parental socioeconomic status)

2, 3 2, 3

Independent variables 1, 2, 3 Migration

PISA TIMSS PIRLS

1, 2, 3

Dependent variables

Variable

Model

Table 8.A.2 Variables and coding for regression analysis

533.32 533.00 517.15 489.80 563.82 536.87 509.77 529.34 529.20 493.15

0.46 0.50 0.52 0.51 0.49 0.49 0.50 0.50 0.50 0.51

0.96 0.94 0.92 0.88 0.96 0.95 0.96 0.93 0.94 0.95

0.16 0.15 0.14 0.14 0.10 0.21 0.16 0.13 0.20 0.20

0.11 0.13 0.12 0.11 0.07 0.14 0.14 0.10 0.13 0.25

0.11 0.10 0.09 0.05 0.08 0.06 0.04 0.08 0.06 0.07

0.12 0.10 0.02 0.10 0.04 0.13 0.06 0.12 0.02 0.05

0.18 0.12 0.05 0.07 0.19 0.11 0.07 0.19 0.04 0.11

0.68 0.82 0.61 0.64 0.39 0.65 0.76 0.51 0.72 0.77

0.29 0.48 0.28 0.12 0.15 0.39 0.43 0.14 0.33 0.30

0.04 0.03 0.07 0.14 0.06 0.21 0.06 0.07 0.10 0.12

0.60 0.56 0.44 0.50 0.41 0.56 0.62 0.48 0.49 0.46

0.14 0.00 0.26 0.31 0.11 0.24 0.48 0.56 0.27 0.27

0.00 1.00 0.11 0.11 0.06 0.00 0.02 0.03 0.08 0.23

45.4 45.7 43.4 43.9 47.0 45.2 45.1 45.1 45.9 46.1

20.3 13.1 11.8 15.0 11.0 17.8 11.0 20.9 5.6 16.3

0 0 0 0 0 0 0 0 0 0

15.8 3.4 6.7 9.1 5.9 11.1 6.1 15.9 3.2 4.4

63.6 64.7 40.7 47.8 45.0 57.9 61.3 64.5 23.3 72.4

Share immigrants in school Area School Mis ISEI Average 5th 50th 95th

Source: PISA 2000, author’s calculations. Note: The share of immigrants in schools refers to the school average. In addition, the table gives the share of immigrants in the 5th, 50th, and 95th percentile of schools by country.

Australia Canada France Germany Netherlands New Zealand Sweden Switzerland UK USA

Math

Second First Second Tertiary Edu Gender Sibling Single Other Gen Gen Language Edu Edu Miss Books Area

Table 8.A.3 PISA summary statistics

518.87 520.54 502.31 528.84 500.94 513.38 533.69 496.02 501.63

0.50 0.50 0.51 0.50 0.48 0.49 0.48 0.49 0.50

0.89 0.83 0.76 0.90 0.88 0.84 0.88 0.88 0.84

Gender Sibling 0.15 0.18 0.14 0.07 0.17 0.12 0.14 0.17 0.19

Single

Source: TIMSS 1995, 1999; author’s calculations.

Australia Canada Germany Netherlands New Zealand Sweden Switzerland UK USA

Math

Table 8.A.4 TIMSS summary statistics

0.10 0.13 0.09 0.06 0.13 0.12 0.07 0.14 0.20

Other 0.10 0.11 0.05 0.04 0.07 0.04 0.11 0.08 0.08

0.10 0.08 0.08 0.03 0.08 0.05 0.10 0.03 0.06

Second First Gen Gen 0.09 0.10 0.13 0.09 0.09 0.09 0.19 0.05 0.10

Language 0.47 0.64 0.27 0.50 0.52 0.46 0.51 0.04 0.73

0.22 0.35 0.06 0.07 0.21 0.20 0.03 0.01 0.46

0.20 0.19 0.30 0.34 0.29 0.42 0.29 0.95 0.19

Second Tertiary Edu Edu Edu Miss

0.67 0.58 0.52 0.42 0.66 0.65 0.46 0.48 0.50

Books

0.15 0.15 0.12 0.00 0.15 0.17 0.00 0.12 0.20

Area

0.15 0.15 0.41 1.00 0.06 0.08 1.00 0.18 0.18

Area Mis

544.15 525.17 539.09 554.21 528.82 561.01 550.46 542.15

0.50 0.48 0.50 0.50 0.49 0.49 0.52 0.51

0.88 0.88 0.82 0.91 0.90 0.91 0.87 0.86

0.10 0.10 0.08 0.07 0.11 0.11 0.13 0.11

Gender Sibling Single

Source: PIRLS 2001, author’s calculations.

Canada France Germany Netherlands New Zealand Sweden UK USA

Read

Table 8.A.5 PIRLS summary statistics

0.09 0.10 0.06 0.05 0.08 0.06 0.06 0.07

0.14 0.06 0.08 0.05 0.12 0.06 0.08 0.13

Second First Gen Gen 0.16 0.13 0.10 0.14 0.16 0.10 0.11 0.15

Language 0.70 0.39 0.49 0.25 0.68 0.77 0.23 0.00

0.17 0.12 0.10 0.03 0.21 0.25 0.11 0.00

0.23 0.27 0.39 0.38 0.23 0.12 0.51 1.00

Second Tertiary Edu Edu Edu Miss

0.44 0.37 0.32 0.28 0.45 0.57 0.42 0.42

Books

0.20 0.33 0.42 0.40 0.21 0.18 0.20 0.32

Area

0.04 0.04 0.06 0.14 0.05 0.03 0.05 0.04

Area Mis

529.42 (9.30)∗ ∗ ∗ 2768 0.13

522.81 (4.28)∗ ∗∗ 15712 0.10

−17.30 (3.50)∗ ∗ ∗ −3.23 (5.45) −18.53 (4.51)∗ ∗ ∗ −16.37 (5.30)∗ ∗ ∗ 26.64 (4.13)∗ ∗ ∗ 1.83 (4.32) −39.91 (8.07)∗ ∗ ∗ 34.43 (3.84)∗ ∗ ∗ −32.27 (8.60)∗ ∗ ∗ 3.99 (10.30) 520.41 (7.91)∗ ∗ ∗ 2356 0.21

−23.09 (4.12)∗ ∗ ∗ 2.21 (4.50) −15.08 (4.96)∗ ∗ ∗ −33.07 (6.39)∗ ∗ ∗ 32.64 (5.22)∗ ∗ ∗ 35.69 (5.90)∗ ∗ ∗ 9.18 (8.45) 50.58 (4.45)∗ ∗ ∗ −13.70 (7.98)∗ −15.39 (12.21) 475.82 (8.29)∗ ∗ ∗ 2442 0.26

−23.47 (11.72)∗ ∗ −5.34 (9.32) −56.42 (11.89)∗ ∗ ∗

DEU

−9.08 (5.20)∗ 0.56 (10.46) −27.38 (8.28)∗ ∗ ∗ −12.46 (6.89)∗ 12.50 (5.76)∗ ∗ 5.32 (7.70) −56.38 (11.49)∗ ∗ ∗ 41.25 (6.20)∗ ∗ ∗ 4.07 (16.13) 24.89 (21.50) 560.54 (12.29)∗ ∗ ∗ 1314 0.25

−78.24 (17.17)∗ ∗ ∗ −46.88 (14.65)∗ ∗ ∗ −14.88 (8.08)∗

NLD

538.76 (11.79)∗ ∗ ∗ 1873 0.19

−3.66 (4.70) −23.97 (9.38)∗ ∗ −18.90 (5.03)∗ ∗ ∗ −33.45 (5.18)∗ ∗ ∗ 20.05 (5.61)∗ ∗ ∗ 20.21 (4.15)∗ ∗ ∗ −17.36 (6.33)∗ ∗ ∗ 44.59 (4.51)∗ ∗ ∗ −11.11 (6.19)∗

17.52 (7.52)∗ ∗ −7.50 (10.68) −42.65 (9.44)∗ ∗ ∗

NZL

−12.51 (3.16)∗ ∗ ∗ 1.07 (8.08) −17.94 (5.28)∗ ∗ ∗ −23.31 (4.80)∗ ∗ ∗ 19.95 (5.46)∗ ∗ ∗ −0.10 (3.67) −35.03 (9.50)∗ ∗ ∗ 36.68 (4.09)∗ ∗ ∗ −8.65 (4.22)∗ ∗ 6.90 (15.45) 496.65 (9.28)∗ ∗ ∗ 2362 0.14

−28.60 (12.80)∗ ∗ −27.97 (11.04)∗ ∗ −30.53 (12.68)∗ ∗

SWE

Source: PISA 2000; author’s calculations. Note: standard errors in parenthesis; ∗ denotes significance at 10, ∗ ∗ significance at 5, and ∗ ∗ ∗ significance at 1 percent level.

Observations R-squared

Constant

Area missing

Area

Books at home

Education missing

Mother tertiary

Mother secondary

Other

Single

Sibling

−11.41 (1.81)∗ ∗∗ −14.46 (3.29)∗ ∗∗ −14.34 (2.45)∗ ∗∗ −22.38 (2.49)∗ ∗∗ 18.88 (2.78)∗ ∗∗ 18.05 (2.22)∗ ∗∗ −30.75 (7.59)∗ ∗∗ 26.10 (1.96)∗ ∗∗

−14.97 (4.93)∗ ∗ ∗ −20.18 (7.97)∗ ∗ −5.91 (5.23) −9.91 (5.80)∗ 14.27 (5.30)∗ ∗ ∗ 29.64 (4.92)∗ ∗ ∗ −23.61 (8.73)∗ ∗ ∗ 34.14 (4.19)∗ ∗ ∗ −18.72 (5.97)∗ ∗ ∗

Gender

−45.69 (15.06)∗ ∗ ∗ −16.25 (7.59)∗ ∗ −26.14 (8.18)∗ ∗ ∗

−8.10 (5.23) −1.44 (3.76) −4.83 (3.59)

0.02 (7.28) 6.38 (6.36) −13.64 (6.92)∗

FRA

CAN

First-generation immigrant Second-generation immigrant Language

AUS

Table 8.A.6 OLS regression results for PISA math (Model 2)

−19.71 (3.89)∗ ∗ ∗ 11.76 (7.66) −11.34 (4.75)∗ ∗ −27.25 (6.33)∗ ∗ ∗ 38.20 (4.39)∗ ∗ ∗ −1.46 (4.91) −28.89 (7.95)∗ ∗ ∗ 41.85 (3.98)∗ ∗ ∗ −15.29 (7.84)∗ 0.72 (24.32) 518.89 (10.75)∗ ∗ ∗ 3173 0.29

−47.88 (7.30)∗ ∗ ∗ −25.78 (8.42)∗ ∗ ∗ −32.27 (6.28)∗ ∗ ∗

CHE

−13.60 (4.63)∗ ∗ ∗ −23.60 (6.35)∗ ∗ ∗ −22.90 (3.66)∗ ∗ ∗ −18.24 (5.40)∗ ∗ ∗ 18.23 (4.47)∗ ∗ ∗ 12.41 (4.24)∗ ∗ ∗ −26.02 (7.19)∗ ∗ ∗ 46.16 (4.24)∗ ∗ ∗ 1.09 (5.51) −4.72 (8.71) 533.11 (7.74)∗ ∗ ∗ 4927 0.17

−4.28 (20.48) −7.17 (9.17) −36.29 (9.21)∗ ∗ ∗

GBR

−10.99 (4.70)∗ ∗ 1.73 (8.01) −32.92 (6.05)∗ ∗ ∗ −46.19 (4.81)∗ ∗ ∗ 35.61 (7.75)∗ ∗ ∗ 23.96 (5.32)∗ ∗ ∗ −12.34 (9.80) 44.55 (4.58)∗ ∗ ∗ −18.94 (7.59)∗ ∗ 7.83 (10.71) 471.43 (10.79)∗ ∗ ∗ 1878 0.28

22.87 (12.85)∗ 14.22 (11.84) −56.95 (15.34)∗ ∗ ∗

USA

0.42 (3.63) 3.01 (3.96) −7.64 (2.98)∗∗ −23.43 (3.66)∗∗∗ 5.89 (3.06)∗ 26.90 (3.32)∗∗∗ −21.67 (3.70)∗∗∗ 38.12 (3.00)∗∗∗ −13.52 (11.30) 10.65 (9.54) 494.34 (6.81)∗∗∗ 6532 0.15

Gender

524.26 (8.94)∗ ∗ ∗ 1769 0.13

−13.32 (4.02)∗ ∗ ∗ −6.35 (8.83) −4.72 (6.51) −47.64 (15.83)∗ ∗ ∗ 16.28 (5.72)∗ ∗ ∗ −4.48 (12.60) −2.64 (6.27) 37.20 (6.49)∗ ∗ ∗

−7.57 (3.92)∗ 1.28 (3.55) −11.65 (5.42)∗ ∗ −21.53 (5.08)∗ ∗ ∗ 11.04 (4.27)∗ ∗ 18.96 (8.08)∗ ∗ −17.16 (4.40)∗ ∗ ∗ 48.29 (4.25)∗ ∗ ∗ −12.40 (12.62) −14.15 (8.87) 499.12 (7.32)∗ ∗ ∗ 2321 0.23

1.45 (3.05) 1.50 (4.68) −19.54 (3.40)∗∗ ∗ −10.95 (3.67)∗∗ ∗ 13.53 (3.24)∗∗ ∗ 10.85 (3.20)∗∗ ∗ −3.59 (4.53) 6.97 (2.50)∗∗ ∗ 8.91 (7.73) 9.96 (14.55) 509.19 (7.11)∗∗ ∗ 7883 0.06

−38.76 (17.33)∗ ∗ −8.20 (11.78) 8.59 (7.66)

−10.15 (8.37) 5.75 (10.70) −38.54 (7.81)∗ ∗ ∗

−5.63 (7.81) −5.85 (4.95) −15.05 (5.86)∗∗

NLD

DEU

CAN

−8.34 (5.23) 7.57 (3.45)∗ ∗ −17.28 (3.70)∗ ∗ ∗ −31.12 (4.35)∗ ∗ ∗ 3.75 (4.31) 27.25 (4.48)∗ ∗ ∗ −9.59 (3.86)∗ ∗ 39.06 (3.64)∗ ∗ ∗ −7.64 (8.97) −21.39 (12.68)∗ 480.24 (6.91)∗ ∗ ∗ 3523 0.17

22.87 (7.78)∗ ∗ ∗ −21.01 (6.52)∗ ∗ ∗ −24.43 (6.66)∗ ∗ ∗

NZL

−6.37 (2.91)∗ ∗ 7.77 (3.35)∗ ∗ −3.23 (3.96) −16.71 (3.63)∗ ∗ ∗ 20.91 (3.66)∗ ∗ ∗ 6.68 (2.89)∗ ∗ 6.31 (3.78)∗ 34.11 (3.30)∗ ∗ ∗ −11.05 (6.45)∗ −18.66 (6.86)∗ ∗ ∗ 486.80 (5.88)∗ ∗ ∗ 3260 0.15

−27.83 (7.05)∗ ∗ ∗ −7.09 (6.35) −11.20 (7.58)

SWE

Source: TIMSS 1995, 1999; author’s calculations. Note: standard errors in parenthesis; ∗ denotes significance at 10, ∗ ∗ significance at 5, and ∗ ∗ ∗ significance at 1 percent level.

Observations R-squared

Constant

Area missing

Area

Books at home

Education missing

Mother tertiary

Mother secondary

Other

Single

Sibling

15.55 (5.67)∗∗ ∗ −2.44 (5.50) −29.03 (6.10)∗∗ ∗

First-generation immigrant Second-generation immigrant Language

AUS

Table 8.A.7 OLS regression results for TIMSS math (Model 2)

526.08 (5.88)∗ ∗ ∗ 4645 0.26

−8.28 (2.71)∗ ∗ ∗ 8.16 (3.44)∗ ∗ −10.75 (3.76)∗ ∗ ∗ −9.16 (5.12)∗ 12.15 (3.99)∗ ∗ ∗ 30.43 (8.30)∗ ∗ ∗ −7.44 (3.78)∗ 31.42 (3.15)∗ ∗ ∗

−25.61 (7.75)∗ ∗ ∗ −19.10 (6.85)∗ ∗ ∗ −33.82 (5.81)∗ ∗ ∗

CHE

−17.87 (5.57)∗ ∗ ∗ −3.86 (4.72) −16.30 (3.56)∗ ∗ ∗ −28.19 (4.26)∗ ∗ ∗ 5.50 (4.46) 30.33 (9.32)∗ ∗ ∗ −0.38 (4.92) 46.18 (4.70)∗ ∗ ∗ 20.24 (11.08)∗ −10.39 (9.75) 494.53 (7.17)∗ ∗ ∗ 5222 0.15

18.60 (10.58)∗ 13.46 (7.01)∗ −26.22 (8.48)∗ ∗ ∗

GBR

−11.00 (2.11)∗ ∗ ∗ 0.56 (2.52) −24.93 (3.02)∗ ∗ ∗ −31.06 (3.34)∗ ∗ ∗ 19.13 (3.84)∗ ∗ ∗ 21.79 (3.18)∗ ∗ ∗ 1.69 (3.96) 36.82 (2.35)∗ ∗ ∗ −10.88 (5.90)∗ −0.77 (8.11) 485.39 (6.04)∗ ∗ ∗ 8115 0.20

−11.26 (7.22) −1.41 (5.77) −30.19 (3.99)∗ ∗ ∗

USA

14.1 (1.4)∗ ∗ ∗ −3.2 (2.2) −11.2 (2.5)∗ ∗ ∗ 19.1 (2.9)∗ ∗ ∗ 27.8 (1.9)∗ ∗ ∗ 3.3 (3.1) 18.6 (1.5)∗ ∗ ∗ −9.5 (2.0)∗ ∗ ∗ 8.3 (2.8)∗ ∗ ∗ 520.4 (3.5)∗ ∗ ∗ 7431 0.18

9.6 (2.0)∗ ∗ ∗ −4.4 (3.1) −3.8 (3.2) 28.9 (2.6)∗ ∗ ∗ 27.4 (3.2)∗ ∗ ∗ 1.5 (2.8) 22.1 (2.2)∗ ∗ ∗ 2.2 (2.4) 2.7 (5.3) 509.8 (3.5)∗ ∗ ∗ 3211 0.21

−17.7 (4.3)∗ ∗ ∗ −6.6 (3.7)∗ −25.3 (3.3)∗ ∗ ∗

−17.5 (2.3)∗∗∗ 10.9 (2.6)∗∗∗ −38.9 (1.7)∗∗∗ 9.8 (1.3)∗ ∗ ∗ −4.5 (1.6)∗ ∗ ∗ −3.3 (2.3) 33.6 (2.3)∗ ∗ ∗ 20.9 (2.1)∗ ∗ ∗ 15.3 (2.4)∗ ∗ ∗ 26.2 (1.5)∗ ∗ ∗ 8.9 (1.4)∗ ∗ ∗ 15.5 (2.8)∗ ∗ ∗ 509.1 (2.7)∗ ∗ ∗ 6607 0.24

−31.9 (2.7)∗ ∗ ∗ −13.2 (3.3)∗ ∗ ∗ −31.1 (2.7)∗ ∗ ∗

DEU

13.2 (1.5)∗ ∗ ∗ 3.3 (2.7) −2.7 (2.8) 26.7 (2.0)∗ ∗ ∗ 12.2 (4.7)∗ ∗ −6.0 (1.8)∗ ∗ ∗ 11.9 (1.7)∗ ∗ ∗ 4.9 (1.7)∗ ∗ ∗ 3.7 (2.5) 542.9 (3.1)∗ ∗ ∗ 3818 0.19

−22.4 (3.7)∗ ∗ ∗ −35.0 (3.5)∗ ∗ ∗ −13.6 (2.3)∗ ∗ ∗

NLD

25.3 (3.3)∗ ∗ ∗ −1.4 (5.4) 1.1 (5.2) 31.1 (6.3)∗ ∗ ∗ 32.8 (3.9)∗ ∗ ∗ 1.1 (7.1) 31.4 (3.3)∗ ∗ ∗ 8.6 (4.3)∗ ∗ 9.5 (7.4) 485.4 (8.4)∗ ∗ ∗ 2283 0.20

7.7 (5.2) 16.5 (6.3)∗ ∗ ∗ −50.1 (5.1)∗ ∗ ∗

NZL

20.5 (1.5)∗ ∗ ∗ 0.3 (2.6) 0.4 (2.6) 22.1 (2.7)∗ ∗ ∗ 22.5 (1.7)∗ ∗ ∗ 8.0 (3.3)∗ ∗ 18.1 (1.6)∗ ∗ ∗ −4.3 (2.3)∗ −8.1 (5.1) 525.2 (3.5)∗ ∗ ∗ 5701 0.19

−29.0 (3.9)∗ ∗ ∗ −8.1 (3.5)∗ ∗ −22.5 (3.1)∗ ∗ ∗

SWE

Source: PIRLS 2001; author’s calculations. Note: standard errors in parenthesis; ∗ denotes significance at 10, ∗ ∗ significance at 5, and ∗ ∗ ∗ significance at 1 percent level.

Observations R-squared

Constant

Area missing

Area

Books at home

Education missing

Mother tertiary

Mother secondary

Single

Sibling

Gender

Language

Second-generation immigrant

First-generation immigrant

FRA

CAN

Table 8.A.8 OLS regression results for PIRLS reading (Model 2)

17.9 (2.6)∗ ∗ ∗ −6.6 (4.0) −4.8 (4.1) 21.8 (4.6)∗ ∗ ∗ 25.6 (5.2)∗ ∗ ∗ −16.1 (3.2)∗ ∗ ∗ 30.7 (2.7)∗ ∗ ∗ 15.3 (3.1)∗ ∗ ∗ −6.1 (6.8) 543.4 (4.9)∗ ∗ ∗ 3047 0.21

−31.9 (4.9)∗ ∗ ∗ 14.9 (5.5)∗ ∗ ∗ −40.3 (4.3)∗ ∗ ∗

GBR

33.6 (2.5)∗ ∗ ∗ −14.1 (3.0)∗ ∗ ∗ −38.7 (7.0)∗ ∗ ∗ 550.7 (3.8)∗ ∗ ∗ 3575 0.17

14.6 (2.4)∗ ∗ ∗ −6.9 (3.4)∗ ∗ −12.9 (4.2)∗ ∗ ∗

−36.2 (4.0)∗ ∗ ∗ 4.1 (4.3) −45.3 (3.8)∗ ∗ ∗

USA

9

Immigration, education, and the Turkish second generation in five European nations: A comparative study1 Maurice Crul and Hans Vermeulen Introduction

Research on the second generation of postwar immigrants is a relatively new phenomenon. Only in the past decade has it become a central focus in the study of immigrant integration. The postwar second generation in Europe came of age at roughly the same time as the American one – and that was when researchers began exploring it more systematically. Examples of early studies in various European countries are Seifert (1992), Crul (1994), Tribalat (1995), Veenman (1996), and Lesthaeghe (1997). International comparative research on the second generation is still scant. In fact, only one such venture has been undertaken – the EFFNATIS project, conducted from 1998 to 2000 by researchers in eight European countries (EFFNATIS, 2001). Because EFFNATIS focused on different ethnic groups in different countries, no comparisons could be made of how the same ethnic group had fared in different settings. Cross-national comparisons were, therefore, awkward to make (Crul and Vermeulen 2003). The primary focus of this chapter is the comparison of integration processes in different countries. This is an aspect that has received far more attention in European than in American research. More specifically, we compare the integration of second-generation Turkish immigrants in five European countries: Germany, The Netherlands, Belgium, France, and Austria. This focus on immigrants from a single source country offers a complementary contrast to chapter 8 by Schnepf, as does a focus on educational attainment rather than achievement.

1

An earlier version of this chapter was published as an introduction to the special issue, “The Future of the Second Generation: The Integration of Migrant Youth in Six European Countries,” 2003 International Migration Review Winter 2003.

235

236

Immigration and the Transformation of Europe

International comparative research on integration The past decade has seen a growing number of publications on immigration and integration in different European countries. Most such studies are policy-oriented, and they have paid little systematic attention to actual integration processes. Although many confine themselves to Europe (Vermeulen 1997; Mahnig 1998), some also include one or more non-European countries (e.g. Castles and Miller 1993; Faist 1995; Joppke 2000; Mollenkopf 2000). Many such writings have argued that national immigration and integration policies are pervaded by notions of nationality and perceptions of the nation. Many authors also seem to assume that government integration policies constitute one of the major determinants of the integration process itself. A well-known example of the viewpoint that sharply differing ideas of nationhood have a strong influence on citizenship policies is Brubaker’s Citizenship and Nationhood in France and Germany (1992). This type of research has been criticized on a number of grounds. Several authors have warned that the influence of targeted government policies on integration outcomes should not be exaggerated (e.g. Vermeulen and Penninx 2001: 228–9; Muus 2003). Banton (2001) recently argued that the actual integration processes in different European countries may be far more similar than much of the research on national models has suggested. Indeed, Banton’s critique lays bare one of the major shortcomings of international comparative research so far: it has focused very little attention on the integration process itself, or it does so by merely citing a selection of existing studies.

The Turkish second generation in Europe We have chosen to compare the Turkish second generation because they are the largest immigrant group in Europe, numbering up to 4 million, and because they reside in a large number of European countries. The focus is on five countries with sizable Turkish communities – Germany, The Netherlands, Belgium, France, and Austria. To compare “Turks” in different countries does not necessarily mean one is comparing the “same” group. An adequate comparison must also take account of the internal differences within the Turkish emigrant populations, based on characteristics like ethnicity, first-generation education levels, and religion. The socioeconomic backgrounds of Turkish labor migrants turn out to be fairly similar in all receiving countries, with some variations (e.g. in Germany). Similarities across countries are partly explained by the fact that many migrants in different countries

Education and the Turkish second generation

237

Table 9.1 Population of Turkish descent in Germany, The Netherlands, France, Austria, and Belgium Country

Year

Population of Turkish descent

Germany1 Netherlands2 France3 Austria4 Belgium5 Total

2002 2002 1999 2001 2001

2.470.000 299.662 218.360 182.000 109.000 3.279.022

Notes: 1 The Turkish population count in Germany is an estimation because those who are naturalized have to be estimated based on the naturalization registers (Worbs 2003). 2 The Turkish population count in The Netherlands both includes naturalized and non-naturalized (Crul and Doomernik 2003). 3 The Turkish population count includes those who are born in France with one or two parents who are born in Turkey and those with Turkish nationality (Simon 2003). 4 The Turkish population count in Austria excludes those who are born in Austria and who have one or two parents born in Turkey (Herzog 2003). 5 The Turkish population count in Belgium both includes nonnaturalized and an estimate of the naturalized citizens from Turkish descent (Crul and Doomernik 2003).

originate from the same regions in Turkey, or even the same villages. Labor migrants form the vast majority of the Turkish migrants in Europe. However, there are also significant groups of refugees who fled political persecution in Turkey or the armed conflict between Kurds and Turks. Most of them arrived in Europe later than the labor migrants, and their children are still young. One reason why we confine our study to secondgeneration young people who are older than fifteen is that they are almost exclusively children of labor migrants, thereby facilitating the comparison across countries. The second, more important reason, is that our focus is on educational attainment and labor market transition, and thus involves the secondgeneration immigrants who have already ended their school careers and are now entering the job market. Second-generation young people still in school are therefore not included in the comparison, since their eventual status is not yet clear. Since the initial wave of second-generation Turks has only recently entered the labor market, not all countries can supply detailed information on their labor-market status. This is why our

238

Immigration and the Transformation of Europe

emphasis is on the transition from school to labor market, and not on their employment status as such. Our comparison is devoted mainly to what is often called the structural dimension of integration, as reflected more specifically in education and the transition from school to work. One should be aware that we have chosen the Turkish immigrant group partly because of the sharp contrasts between it and the native populations of the Western European countries. Its socioeconomic background is extremely low (unlike the Turkish-American population, who are generally better educated [Karpat 1995], and it is a group with a traditional Muslim background. Turkish immigrants are widely considered to be one of the toughest groups to integrate, and they thereby put to the test the wide panoply of European national policies aiming at the integration of newcomers. We hope that analysis of the difficulties they encounter will accentuate the differences between countries. Turkish labor migration followed comparable patterns everywhere. Beginning with Germany in 1961 and ending with Sweden in 1967, European countries signed official agreements on labor migration with Turkey. Spontaneous migration through relatives and co-villagers then also ensued, later even surpassing the scale of official immigration. The peak of labor migration was between 1971 and 1973, years in which more than half a million Turkish workers came to work in Western Europe, ¨ uekren 90 percent of them recruited by German industry (Oz ¨ and van Kempen 1997:5). From 1973, the economic recession following the oil crisis slowed the demand for labor and prompted an official immigration stop in 1974. Unemployment forced many immigrants to return to their home countries, but many men who had remained began sending for their wives and children. Migration took a new upturn in the 1980s and 1990s, when the in-between generation reached marriage age and began choosing spouses from Turkey. The Turkish population in Europe now totals almost 4 million, including naturalized and second-generation Turks. More than 60 percent of them live in Germany. Most Turkish migrants came from small villages in central Turkey or along the Black Sea coast; those from large cities (Istanbul, Izmir, and Ankara) are in the minority. Some districts in central Turkey delivered inordinate numbers of migrants over the years, often dispersed over various European countries. People from the Afyon district, for instance, now reside in Sweden, Germany, The Netherlands, and France. Some villages have more offspring living in Europe than in the village itself. European industry was in need of low-skilled labor at the time, and indeed the majority of these first-generation Turkish “guest workers” were

Education and the Turkish second generation

239

recruited from the lowest socioeconomic strata in their home countries and had very little education. In the rural areas where most of them grew up, educational opportunities were limited to the primary school level. Generally speaking, first-generation men had only finished primary school and most women had just a few years of schooling. Because smallscale subsistence farming was the primary activity in this peasant way of life, school played no particular role. Sending children to school brought no advantage in the struggle for existence; having them help on the farm had a high priority. Another reason that education seemed to hold little promise was the nature of the schooling on offer. Education in Turkey was not primarily geared to conveying knowledge that would aid people in their peasant existence, or in breaking away from it. Its main aim was to transmit the Turkish national ideology and promote the cultural integration of the country.2 The first generation made few advances in the European labor market – in fact, it was the opposite. Economic crises and industrial restructuring put many Turkish immigrants out of work. Despite this, a substantial group of first-generation men did manage to start their own business or to help their children to do so. Most second-generation children – those born in the country of immigration or (more broadly) those who arrived before primary school age – grew up in unfavorable circumstances. Family income was often very low by European standards, and most families lived in substandard and cramped accommodation. In many neighborhood schools, children from a mix of migrant backgrounds were in the majority. The Turkish second generation compared in five countries We collected data on the second generation in five countries within roughly the same time period and for approximately the same age group (Crul and Vermeulen 2003). Ideally, educational status and labor market status would each be described by several identical indicators in each country. Unfortunately, not all the pertinent categories of data are known for all countries. Relevant indicators for educational status are: school attendance rates; educational performance of school pupils and students; highest educational attainments of graduates and dropouts; dropout percentages and repeater rates. In this comparison we can only make use of the indicators for highest educational attainment and dropout percentages because these are the most relevant, and are easily compared across countries (see the Appendix to this chapter). The data in the 2

On the role of education in rural Turkey at the time of mass emigration, see Coenen 2001: 56–73.

240

Immigration and the Transformation of Europe

tables per country compiled in the Appendix are not gathered at the same moment in time, are gathered for different age groups and different educational categories are used in each country. Due to this incomplete evidence, we will make comparisons only with great caution, and as a rule we will only describe situations for which differences are clear and substantial. As a result, we do not describe percentages of school attainment or dropout percentages of each individual country, but we will refer to the tables in the Appendix. In the tables, the different criteria of the data collection and the data source is described for each country. Integration policies A common feature of educational policy in European immigration countries in the 1980s was their lagging adaptation to the changing realities of education for immigrant pupils. Piecemeal engineering was the rule, and coherent sets of policy responses materialized only slowly (Fase 1994). Initially, the persisting belief that immigrants would go home someday was the main underlying reason for the lack of policies. Policy differences between countries were greatest in the early 1980s, but the policies increasingly converged (Vermeulen 1997; Doomernik 1998; EFFNATIS 2001). In day-to-day practice, the sharply contrasting national integration policies (which have often been depicted as codifications of national traditions and culture) were confronted everywhere with the same basic needs of migrant children. In essence, this meant that all countries ultimately launched first- and second-language programs, compensatory programs, and preparatory programs for both primary and secondary school. The differences between countries, according to Fase’s observations, lay more in the specific methods the authorities applied to deal with these issues and in the relative priority assigned to various facets. In some countries, for example, compensatory programs were integrated into school curricula, while other countries opted for separate programs and classes for migrant children. There were also wide variations in the dates of inception. Nonetheless, Fase concludes that such differences had few consequences for the educational status of immigrant children in the various countries (Fase 1994: 164). Policy convergence in labor market issues has been less pronounced than in education. In some countries like The Netherlands and Sweden, targeted policies have been developed to help migrants integrate into the private and public sectors of the labor market. In other countries, on the contrary, access to government jobs is restricted to national citizens. In all countries, immigrant youth unemployment is a policy concern and

Education and the Turkish second generation

241

labor market discrimination is officially acknowledged, though policies to combat that discrimination vary widely. Differences in outcomes The school careers of second-generation children exhibit startling differences across Europe. The greatest distinctions can be seen in the percentages of young people entering lower vocational tracks – the lowest secondary school type in all countries. In France, Belgium, and The Netherlands, between one-third and one-half of the second-generation Turks begin their secondary school careers in lower vocational school, whereas in Germany and Austria the figure is between two-thirds and three-quarters (see Tables 9.A.1–5 in the Appendix). At the top end of the ladder, we see that far more second-generation Turks in France and Belgium enter preparatory tracks for higher education than in other countries (see Tables 9.A.3 and 9.A.4 in the Appendix). National contexts vary widely in the types of opportunities they offer to second-generation Turks. Although one might now be tempted to conclude that France and Belgium, and to a lesser extent The Netherlands, provide the best institutional contexts for migrants, that is not the whole story. A comprehensive assessment also requires knowledge of how children perform in vocational or in academic tracks. In France, for instance, we see that as well as having higher rates of entry into preparatory tracks for higher education, the Turkish second generation also has higher school dropout rates than in other countries (see Tables 9.A.1–6 in the Appendix). Of the secondgeneration Turkish young people in France who have already ended their school careers, almost half have gained no secondary school diploma at all, compared to only one-third in The Netherlands and substantially fewer in Germany and Austria (see Tables 9.A.1–6 in the Appendix). In the latter two countries, the majority of Turkish second-generation children enter an apprenticeship system that enables them to work and study while gaining job qualifications and experience. Transition to the labor market likewise differs sharply between countries. The apprenticeship system in the German-speaking countries ensures a relatively smooth transition. Unemployment among secondgeneration Turks in countries with apprentice systems is three to four times lower than in France, Belgium or The Netherlands. Such systems apparently give young people with low vocational diplomas a start on the job market – a step much harder to accomplish in countries without such a system. Worbs (2003) shows how Turkish second-generation youth in Germany have acquired a relatively secure labor market position through the apprenticeship system. In France and Belgium, and to a

242

Immigration and the Transformation of Europe

lesser extent in The Netherlands, we see more polarization. While a substantial group of second-generation Turks are reaching white-collar or professional positions, many qualified and unqualified workers are suffering serious unemployment as a result of their difficult transition to the labor market. In terms of higher education, the French educational system appears more effective at guiding Turkish young people towards university than the German system, while the German system is better at incorporating young adults into the labor market. Thus, although the findings presented here do not point to any clear effect of integration policies that specifically target migrants, some of the generic national institutional arrangements do seem to matter. Belgium represents an excellent case in point. The Walloon region is strongly focused on the French republican model of integration, while the Flemish region is more inspired by the Dutch multicultural model. At the same time, the education system is the same all over Belgium, and so are the school achievements of the Turkish second generation! (See Timmerman et al. 2003.) This remarkable outcome is perhaps the most compelling evidence yet that integration models do not have the impact they are often purported to have. Explaining differences in outcomes On the basis of the studies mentioned before, we are now in a position to identify the factors in national education systems that either hinder or facilitate the school careers of second-generation Turks. National educational systems may differ in school duration, face-to-face contact hours with teachers, selectivity, and amounts of supplementary help and support available to children inside and outside school. In the transitions from school to work, the presence or absence of an apprenticeship system forms a major point of contrast. The only feature of the national systems that specifically relates to migrant children is the second-language training. One significant disparity between countries lies in the age at which education begins. In France and Belgium, Turkish second-generation children start school at the age of two-and-a-half, in Germany and Austria at age six, and in The Netherlands at age five. Thus, immigrant children in France and Belgium on average have about three more years of education in that crucial developmental phase in which they begin learning the majority language. In France and Belgium, very young Turkish children thus find themselves in situations almost every day where they have to speak French or Dutch with their peers and learn these languages in an educational environment.

Education and the Turkish second generation

243

Striking differences also appear in the number of face-to-face contact hours with teachers during the years of compulsory schooling. Here, once again, these are below average for Turkish pupils in the Germanspeaking countries, especially during the first part of their educational careers. Nine-year-olds in German schools have a total of 661 contact hours, as compared to 1,019 hours in The Netherlands, because children in Germany and Austria only attend school on a half-day basis. Turkish children in Germany thus receive about ten hours less tuition per week than those in The Netherlands. Although children in Germany and Austria are assigned more homework, help with homework is a scant resource in Turkish families. This may be a source of serious disadvantage. A third distinction, which in combination with the first two can culminate in serious disparities, lies in school selection mechanisms. Germany and Austria select at the age of ten. In Germany the selection mechanism channels the children into three school levels and in Austria into two. Coupled with the late start in education and the below-average contact hours, Turkish second-generation pupils in Germany and Austria are thus given little time to pull themselves out of their disadvantaged starting position. In this respect, Turkish children in Germany and Austria are in the worst possible situation. Selection in The Netherlands occurs two to four years later. Belgium selects at age fourteen and France at fifteen. In countries with a relatively early selection, most pupils end up in short vocational streams – Hauptschule in Germany and Austria, and lower vocational school in The Netherlands. The higher selection ages in France and Belgium result in higher percentages of Turkish children moving into more prestigious streams. As suggested above, this has its benefits and drawbacks. Though it may offer more opportunities to Turkish children than they receive in other countries, many of them falter in the higher tracks and end up with no diploma at all. A fourth area in which major differences are evident between countries involves the amount of assistance and support made available to migrant children inside and outside school. All countries have a host of educational priority projects aimed either specifically at migrant youth or more broadly at youth with learning problems. Some programs are national, others are regional or municipal initiatives. Though cross-national comparisons are difficult to make, the most reliable data so far derive from an international study known as PISA 2000 (www.pisa.oecd.org). It questioned fifteen-year-olds about the supplementary assistance and support they received inside and outside school. Although it only distinguishes between youth with migrant backgrounds and native youth, it gives a good indication of the extra support received by Turkish children in each country. It delivers a clear ranking. Migrant children in France, and to a

244

Immigration and the Transformation of Europe

lesser extent in Belgium and The Netherlands, receive the most support, and Germany performs worst of all. Since basically the same differences between countries apply to native youth, we may simply speak of divergent general education practices in different countries. Less meaningful for explaining differences between countries, but not altogether insignificant, are the second-language programs. The options and practices of second-language education are many and varied, and there is still considerable debate about the best method for improving proficiency in official national languages. This has yielded a multitude of programs and methods, ranging from transitional bilingual programs to intensive instruction exclusively in the second language. No country appears to have clear-cut guidelines in place for the provision of second-language teaching. Normally it is a part of primary school curricula, but it may be integrated there into mainstream language programs, given as supplementary instruction to migrant children during school hours, or provided outside of school hours. In comparing this we focus on two issues: when the countries introduced second-language programs on any substantial scale, and how the quality and results of the programs roughly compare. Virtually all countries now have well-established programs. The biggest distinction lies in when they were introduced. France began implementing orientation classes back in the early 1970s. Other countries started much later. In the late 1980s, Belgium had only one project of any real significance, and The Netherlands did not introduce programs on a larger scale until the early 1990s. In Germany, some federal states opted for intensive second-language programs, while others provided tuition in migrant languages, creating separate classes for the children. Even into the 1990s, however, methods of learning German other than the traditional approaches were still rare. Overall, then, the group we are focusing on here – the second generation above age fifteen who attended primary school in the 1980s or early 1990s – did not profit from special language programs to any reasonable degree. In most countries, the programs reached only limited numbers of children and their quality was questionable. In some cases, as in Belgium and France, secondgeneration children born in the country itself were mostly excluded from the programs, because they were thought to have no language problems. If we view all five of these factors together, it seems no wonder that second-generation Turks in France enter preparatory schools for higher education at higher rates than elsewhere in Europe. Children start school early in France, have more hours of face-to-face tuition, have the most supplementary help and support available inside and outside school, and do not undergo educational selection until a fairly late age. At the other extreme are Austria and Germany, where children enter school late, are

Education and the Turkish second generation

245

selected early, and in the former two countries also have fewer contact hours and receive less supplementary support. In France and Belgium, twice as many Turkish second-generation young people reach middle or higher educational levels than in Germany, Austria or The Netherlands. France’s relatively open system does carry the disadvantage that many of the children wind up leaving school without a diploma. The countries that perform rather poorly in the first stages of the educational track, Germany and Austria, do a relatively good job in the later stages, and they also deliver more successful transitions to the job market. This can be credited in large part to their apprenticeship systems. Although the distribution of study hours in relation to working hours varies between these countries, the core principle is to introduce youngsters to working life at an early age. Turkish young people thereby get a chance to test their skills early, and once they have served in a firm as an apprentice, they are often hired to work there. In countries where no apprentice system exists, such as France, Belgium, and The Netherlands, young people must venture onto the labor market alone. Many employers consider them not yet suitable for jobs, and favor applicants with work experience. Getting a foothold in the job market is difficult, and especially for the children of migrant parents, who consequently suffer higher rates of unemployment. Where selection is tougher, discrimination seems to hit harder. In countries without apprenticeship systems, secondgeneration Turks with low vocational qualifications are widely shut out of the labor market. Young people without diplomas are at risk in any system, and some countries produce far more of them than others. France and The Netherlands are extreme examples in terms of Turkish school dropouts. Conclusion The comparisons we make in this issue are not comprehensive. The data needed for adequate comparisons are missing in some countries, and research on the second generation is still scant everywhere. Although we need to be cautious in drawing conclusions, we do seem to have identified some interesting leads that we hope to build on in our future research. We now want to propose some hypotheses to guide that research. The position of second-generation Turks varies widely between the different countries in Europe, and it is not easy to make an overall assessment of trends. The picture is further complicated by the polarizations within ethnic groups that exist in some countries. The debate about integration seems to have had a persistent blind spot for the importance of the national context in which the second generation

246

Immigration and the Transformation of Europe

is trying to move forward. Although rigorous study is still urgently needed, we believe we can already conclude from the material presented here that national contexts have a considerable impact on the paths of integration that the second-generation Turks are following in the various countries. In some countries, a sizable number of second-generation Turks have fallen behind to the point where they now seriously risk becoming an underclass. In other countries, the Turks seem to be performing well enough, either through education or the apprenticeship system, to be able to integrate smoothly. The cross-national comparisons do not yet permit unequivocal conclusions about the desirability of one system or the other. In countries where second-generation Turks have experienced a smooth transition to the labor market, only small numbers have managed to enter higher education. This seems the price that Turks have to pay. In countries with more open educational systems, the second generation can reach higher, but fall deeper. The Turkish communities there are already characterized by a degree of polarization – as the first Turkish elites are emerging, another subgroup may be spiraling towards the bottom of the social hierarchy. Interestingly, the differential outcomes reported here do not seem to be attributable to arrangements specifically targeted at migrant youth, but more to the generic policies prevailing in each particular country. A further conclusion, then, which we again draw with some caution, is that the probability of underclass formation may be linked to the opportunities that national, generic institutional arrangements for education and labor market transition offer to the second generation. This means that a debate on the differential effectiveness of national institutional arrangements is just as urgently needed as the discussions on distinctions between ethnic groups.  Banton, Michael 2001 “National Integration in France and Britain.” Journal of Ethnic and Migration Studies 27(1) (January, 1): 151–68. Brubaker, Rogers 1992 Citizenship and Nationhood in France and Germany. Cambridge, MA: Harvard University Press. Castles, Stephen and Mark J. Miller 1993 The Age of Migration: International Population Movements in the Modern World. New York: Guilford Publications, Inc. Coenen, Elisabeth M. 2001 ‘Word niet zoals wij!’: de veranderende betekenis van onderwijs bij Turkse gezinnen in Nederland. Amsterdam: Het Spinhuis. Crul, Maurice 1994 “Springen over je eigen schaduw. De onderwijsprestaties van Marokkanen en Turken van de tweede generatie.” Migrantenstudies 10(3): 168–86. Crul, Maurice and Jeroen Doomernik 2003 “The Turkish and Moroccan Second Generation in the Netherlands: Divergent Trends between and

Education and the Turkish second generation

247

Polarization within the Two Groups.” International Migration Review 37 (4(144)) (Winter): 1039–64. Crul, Maurice and Hans Vermeulen 2003 “The Second Generation in Europe.” International Migration Review 37 (4(144)) (Winter): 965–86. Doomernik, Jeroen 1998 The Effectiveness of Integration Policies towards Immigrants and their Descendants in France, Germany and the Netherlands. Geneva: Conditions of Work Branch, International Labour Office. EFFNATIS 2001 “Effectiveness of National Integration Strategies towards Second Generation Migrant Youth in a Comparative European Perspective.” Bamberg: European Forum for Migration Studies. 29 March. Faist, Thomas 1995 Social Citizenship for Whom? Aldershot: Ashgate Publishing Ltd. Fase, Willem 1994 Ethnic Divisions in Western European Education. New York: Waxmann. Herzog-Punzensberger, Barbara 2003 “Ethnic Segmentation in School and Labor Market – 40 Year Legacy of Austrian Guestworker Policy.” International Migration Review 37 (4(144)) (Winter): 1120–44. Joppke, Christian 2000 Immigration and the Nation-State: The United States, Germany and Great Britain. Oxford: Oxford University Press, Inc. Karpat, Kemal 1995 “The Turks in America,” in Turcs d’Europe – et d’ailleurs, 3, St´ephane de Tapia (ed). Paris: ERISM: Institut national des langues et civilisations orientales, 231–52. Lesthaeghe, Ron J. 1997 Diversiteit in sociale verandering: Turkse en Marokkaanse vrouwen in Belgie. Brussels: Vubpress. Mahnig, Hans 1998 Integrationspolitik in Grossbritannien, Frankreich, Deutschland und den Niederlanded. Vol Forschungsbericht No 10. Neuenburg: Schweizerisches Forum fur ¨ Migrationsstudien an der Universit¨at Neuenburg. Mollenkopf, John H. 2000 “Assimilating Immigrants in Amsterdam: A Perspective from New York.” The Netherlands’ Journal of Social Sciences 36(2): 12– 145. Muus, Philip 2003 “An International Comparison of Migration and Immigrant Policy with respect to Immigrants from Turkey and their Participation in the Labour Market,” in Integrating Immigrants in the Netherlands: Cultural versus Socio-Economic Integration, L. Hagendoorn, J. Veenman and W. Vollenbergh (eds). Aldershot: Ashgate Publishing Ltd 17–40. ¨ uekren, Oz ¨ Sule and Ronald van Kempen (eds) 1997 “Turks in European Cities: Housing and Urban Segregation.” Research in Migration and Ethnic Relations. Aldershot: Ashgate Publishing Ltd. Seifert, Wolfgang 1992 “Die zweite Auslandergeneration in der Bundesrepublik. Langsschnittbeobachtungen in der Berufseinstiegsphase.” Kolner Zeitschrift fur Soziologie und Sozialpsychologie 44(4) (December): 677–96. Simon, Patrick 2003 “France and the Unknown Second Generation: Preliminary Results on Social Mobility.” International Migration Review 37 (4[144]) (Winter): 1091–1119. Timmerman, Christiane, Els Vanderwaeren and Maurice Crul 2003 “The Second Generation in Belgium.” International Migration Review 37 (4[144]) (Winter): 1065–90. Tribalat, Mich`ele 1995 Faire France: Une Grande Enquˆete sur les Immigr´es et leurs Enfants. Paris: Editions La D´ecouverte.

248

Immigration and the Transformation of Europe

Veenman, J. 1996 Keren de Kansen? de Tweede-Generatie Allochtonen in Nederland. Assen: Van Gorcum. Vermeulen, Hans 1997 Immigrant Policy for a Multicultural Society: A Comparative Study of Integration, Language and Religious Policy in Five Western European countries. Brussels: Migration Policy. Vermeulen, Hans and Rinus Penninx (eds). 2001 Immigrant Integration: The Dutch Case. Amsterdam: Aksant Academic Publishers. Worbs, Suzanne 2003 “The Second Generation in Germany: Between School and Labor Market.” International Migration Review 37(4[144]) (Winter): 1001– 38.

Education and the Turkish second generation

249

Appendix Table 9.A.1 Highest completed level of education of Turkish second-generation youngsters in the age category 16–25, according to gender (in percent)in Germany, 1995

Turkish Second Generation, Male (N = 590) Turkish Second Generation, Female (N = 488)

Still at school/ no degree

Hauptschule

Realschule

Gymnasium

38.1

45.9

10.7

5.3

45.1

32.0

15.0

8.0

Source: Micro census data, 1995, own calculations (Worbs 2003).

Table 9.A.2 Highest completed level of education of Turkish second-generation youngsters in The Netherlands aged 15–35 who had ended their educational careers by 1998 (percentages rounded) Highest educational attainment

Turkish

Primary school at the most Lower vocational education (Vbo), lower general secondary education (Mavo) Senior secondary vocational education (Mbo), senior general secondary education (Havo), preparatory university education (Vwo) Higher professional education (Hbo), university (WO)

34% 37% 23% 5%

∗

The top row of the table represents individuals who finished primary school only (normal age 12). Mbo is equivalent to Havo, because both diplomas give access to higher vocational school (Hbo). Source: SPVA 1998, ISEO/EUR.

Table 9.A.3 Highest completed level of education of Turkish second-generation youngsters aged 18–40, according to sex, France, 1999 Drop-out(a)

Vocational school

Baccalaureate

University

Males Females

40.1 51.6

27.8 15.3

19.1 23.3

13.1 7.7

Total

46.4

22.1

21.3

10.1

(a)

By “drop-out,” we mean people leaving school with no diploma, mainly after the coll`ege. Source: INSEE, EHF, 1999, personal calculation (Simon 2003).

250

Immigration and the Transformation of Europe

Table 9.A.4 Highest completed level of education of Turkish second-generation men aged 18 years and older in Belgium, 1996 Observed school levels

Turkish

Unqualified Primary Lower secondary (BSO) Higher secondary (ASO; TSO) Higher education N

0% 9% 37% 50% 4% 272

Source: HMSM survey Data, 1994–1996.

Table 9.A.5 Highest completed level of education of Austrian residents born in Turkey or having Turkish citizenship according to the age category 15–35 (N = 590), 2001 Highest educational degree of Austrian residents with Turkish background∗ No diploma Compulsory Apprentice Middle sec Higher sec University Second Generation

3%

62%

26%

4%

4%

0%

∗ With the exception of children (of Turkish immigrants) born in Austria, who were either Austrian citizens from birth or were naturalized at a later age (but before this count). Source: Own compilations on the basis of Microcensus 2001 (Herzog 2003).

Table 9.A.6 Highest completed level of education of Turkish second-generation youth in the age category 16–25, except pupils in the general education system or in night schools (in percent) in Germany, 1999 No degree or degree of a special needs school Hauptschule Realschule∗ Gymnasium∗ Turkish Second 6.7 Generation ( N= 239)

64.4

15.9

13.0

∗ The educational degree that pupils can achieve in the Realschule is called Mittlere Reife, the

degree at the Gymnasium is called Abitur (Hochschulreife). All degrees can also be obtained in night schools or some special school types like Fachoberschulen, which are subsumed here under the respective “normal” school type. Source: EFFNATIS field study data, own calculations (Worbs 2003).

10

Managing transnational Islam: Muslims and the state in Western Europe1 Jonathan Laurence

Introduction Contemporary scholarly accounts of Islam in Europe have tended to portray one of two extreme visions: heaven on earth or hell in a hand basket. In the pessimistic version, foreign policy analyses and the scholarly literature are in rare concurrence about the meager chances for either interreligious dialog or Muslim integration. These accounts bear witness to a showdown between intransigently secular states and an ambitious, fundamentalist religion whose followers aim to transform the continent into “Eurabia” (Ye’or 2005; Ferguson 2004; Savage 2004). To justify their gloom, these authors cite Islam’s un-hierarchical nature and the impossibility of establishing legitimate representatives – or “one phone number” – for Muslim communities in Europe (R´emond 1999; Warner and Wenner 2002). Compounding this difficulty, these scholars emphasize, is the inadequacy of Europe’s nineteenth-century State–Church institutions, which stumble from crisis to crisis with this new and agile religious challenger (Fetzer and Soper 2005; Shore 2004). The optimistic voices, on the other hand, come from the camps of post-nationalist theorists and from proponents of a reformed “Euro-Islam” that is divorced from overseers and financiers in the Muslim world (Soysal 1994). But those authors’ cheerfulness is founded, respectively, on two formidable hypotheses: the diminishing importance of “host” state institutions for immigrant integration; and the “sending” states’ renunciation of religious influence over the Muslim Diaspora (AlSayyad and Castells 2002). A decade after they were first expounded, neither of these scenarios is in view. Surprisingly, few studies (and virtually none in English) systematically compare how different national interior ministries have used 1

The author is grateful to Craig Parsons for helpful comments on an earlier draft; Stanley Hoffmann, Peter Hall, Yoshiko Herrera, Sven Steinmo, Noah Dauber, and Hillel Soifer provided useful feedback on this material. Finally, the author would like to acknowledge the support of the Hauser Center at the Kennedy School of Government and the Center on the US and Europe at the Brookings Institution.

251

252

Immigration and the Transformation of Europe

political and institutional processes to organize Islam for State–religion relations (Dassetto et al. 2001). This chapter aims to sidestep the warring bands of pessimists and optimists, and instead sketches the character of State–Islam interaction in the first thirty years (1974–2004) of government consultations with foreign and native-born Muslim representatives in Europe. Based on thirty months of fieldwork and more than 150 interviews with religious leaders and policy-makers in France, Germany, Italy, Belgium, and the United Kingdom, I argue that European nationstates have reasserted their sovereign prerogative to manage the transnational threats associated with their citizens’ religious membership (Grillo 2004). My research examines the theoretical and practical significance of the contemporary efforts to nationalize Muslim religious communities: these are political and institutional processes whose goal is to provide a more amenable context for the socio-political integration of all Muslims, religious or not. Three decades of increasingly assertive policies towards organized Islam in Europe militate against the image of states being overrun by the unplanned or undesired mass settlement of Muslims. This finding is in line with Krasner’s contrarian view that “globalization and state activity have moved in tandem,” and that this occurs across institutional models that might be expected to have dramatically distinct policy outcomes (1999: 223). A portrait emerges of the contemporary European nation-state not as a “weathervane” or a neutral broker among competing interests, but rather as an actor in its own right that structures the nature of group–State relations in crucial ways (Bentley 1949; Evans, Rueschemeyer, and Skocpol 1985). The challenge of integration There is no shortage of “bad news” to support the pessimists. A sense of foreboding and an atmosphere of failed integration and social conflict hang like a cloud over the 15 million or so Muslims living in Western Europe. As the preceding chapters have documented in part, second- and third-generation immigrants suffer disproportionately high unemployment, widespread social discrimination, and feeble political representation in local and national institutions. This is coupled with the impression of imported threats from abroad, taking the form of cultural clashes over headscarves worn by schoolgirls or civil servants, or imams preaching hatred or violence, or violence and vandalism targeting the Jewish community. At best, most observers see a grave crisis of previous models of immigrant integration (Gresh and Ramadan 2000). At worst, they see evidence of a “reverse colonization” that will lead to the “Islamicization” of Europe (Caldwell 2004). Gilles Kepel, an authority on the Muslim

Managing transnational Islam

253

world, has commented that Europe now faces a grave choice: “either we train our Muslims to become global citizens, who live in a democratic, pluralist society, or on the contrary, the Islamists win, and take over those Muslim European constituencies. Then we’re in serious trouble” (Wright 2004). To use an analogy from the first era of modern state building, how do today’s governments attempt to make Frenchmen, Germans, or Italians out of Muslims – as authors have described the state-building process of turning “peasants into Frenchmen,” “juifs” into “Isra¨elites,” or Christians into “good citizens” (Weber 1976)? There are different kinds of integration – socioeconomic indicators tell one story, for example – but political integration is the underlying impulse of this historical process. To discover the true nature of European governments’ strategies towards organized Islam begs a look beyond the repressive measures that states can (and do) take, such as deporting extremist prayer leaders or arresting individuals who threaten public order. Instead, what are the constructive steps that make up the path to citizenship and a measure of integration for this new minority? In twentieth-century Western Europe, three arenas assured the gradual political integration of immigrants and new citizens from southern and eastern Europe: traditional participation through elections and civil society associations; civic inculcation through the national education system; and identification with the nation-state through service in the armed forces (Noiriel 1988). But today, party systems have largely failed to transmit social diversity into parliaments, public schools are in budgetary crisis and mandatory conscription is a thing of the (recent) past. How have national governments granted some degree of representation to these sizable minority populations when traditional civic institutions failed to do so? The answer lies in the unexpected revival of a fourth arena: religious community. In an era of advanced secularization in Western Europe, ironically, governments there have fallen back on religion policy – via national state-church institutions – as a central tool of immigrant integration. The attempts to nationalize Muslim organizations The most striking evidence of this development is a Europe-wide move towards “nationalization” of Islam through the development of national consultations with Muslim civil society. The thousands of mosques, prayer rooms and religious associations that have popped up in the last few decades are nearly all under de jure or de facto foreign influence – as is characteristic of major world religions. The local and national Islam Councils that have emerged to streamline these prayer rooms – of greater

254

Immigration and the Transformation of Europe

or lesser efficacy, from the French Council for the Muslim Religion, the Spanish Islamic Commission, the Belgian Muslim Executive, the Italian Consultation, to the Bavarian Islamic Community, etc. – are the culmination of a fifteen-year push by Muslims and public authorities alike for the legal recognition of Islam and the protection of freedom of worship on par with other major religions. European interior ministries do not presume the existence of some essential “Muslim” to be trained into a mythical “citizen.” But regardless of Muslims’ diversity of national origin, piety and religious affiliation, policy-makers have nonetheless come to see “their” Muslims as a community, a collectivity, and the object of public policy-making. These administrations are not engaged in the special accommodation of Muslims; they are incorporating Islam into pre-existing institutional State–Church relations. It is in this domain that European governments are trying to create the institutional conditions for the emergence of an Italian or German Islam, e.g., rather than just tolerating Islam “in” Italy or Germany. The religion bureaus of European interior ministries structure and mediate the activities of religious organizations. As Baker writes in reference to another part of the world, such efforts aim to “ensure that the centrifugal push of religious loyalties that transnational religious regimes foster [ . . . ] does not overcome the centripetal pull toward national unity that the state must nurture” (Baker 1997). The institutionalized relations between state and religion are predicated on the prioritization of national laws over religious texts, and aspire to steep religious leaders in the secular precepts of a society in which church and state are separate. In practical terms, national interior ministries accomplish these lofty goals by overseeing and helping coordinate the financing and construction of mosques, the training of imams, the appointment of Muslim chaplains in prisons and hospitals, the setting of religious curriculum in publicly-funded schools, and the celebration of major holidays and religious events – from orderly lamb slaughter for Eid al-Adha, to the orderly departure on pilgrimage to Mecca. This guarantees equal access to religious exercise at the same time that it favors the transparency of community ties with foreign governments and international NGOs (non-governmental organizations). The state’s challenge has been to establish these nascent councils as legitimate interlocutors for public authorities. It is important to note that these are not “Muslim” councils with broad connections to these communities, and as such cannot claim to represent Muslims any more than the French bishops’ conference can claim to represent Catholics on nonreligious matters or Jewish central councils or grand rabbis can speak for all Jews. Nor does this stem from a desire to impose a Catholic-style

Managing transnational Islam

255

hierarchy on Islam: after all, Protestants and Jews do not naturally gravitate towards centralized representation, either, but they were required to reorganize to obtain full legal recognition. The creation of Islamic Councils reflects governments’ desire to mold Islam into an organizationally homologous shape, as the modern secular state has asked of all major religions. Governments behave with Islam as they have with previous transnational religious challenges: they seek to weaken ties abroad and strengthen institutional connections at home, in the hope of enhancing the authority of the nation-state over competing demands on citizens’ socio-political loyalties. Religious belonging has persistently posed challenges to the meaning of territory and citizenship by setting constitutions in competition with a higher law. Rosenblum calls this the “painful conflict between the obligations of citizenship and the demands of faith” (Rosenblum 2000). This is remarkably similar to Locke’s concerns over the suitability of Christians as republican citizens, as well as the enlightenment-era logic of Mirabeau or Von Dohm, who argued that emancipated and domestically-oriented Jews would make for more useful members of society as full citizens (Parekh 2000; Mendes-Flohr and Reinharz 1995). In McConnell’s apt paraphrasing of Rousseau: “your citizenship can be in Heaven or in France, but not in both” (2000: 92). Faced with a dual system, priest and prince, Rousseau doubted that believers could be trusted which to choose: “The sacred cult of Christianity aimed to become independent of the sovereign, and had no natural or necessary bond with the body of the state [ . . . ]. Far from attaching citizens’ hearts to the State, it detaches them from it as from all worldly things” (Rousseau, Masters, and Masters 1978: 126). Following the liberal democratic revolutions of eighteenth- and nineteenth-century Europe, the consolidating nation-states often tried to weaken – and domesticate – the powerful transnational Catholic network: new State– Church frameworks eliminated the Church’s monopolistic or dominant position, the clergy was subjected to a civil code, and church property and wealth was taxed or seized. The development of national institutions to regulate religious life sought to reduce the risks of radical anti-state influences and, indirectly, to make better citizens of church faithful. The results of my empirical research suggest that the same kind of effort is the major background to Muslim integration in Europe today. Rather than reflecting the post-national dynamics that some scholars claim to observe, the most significant measures of Muslim accommodation have occurred in reaction to organized Islam’s challenge to the authority of modern, secular states. What emerges as the major plot line of this story are the European states’ parallel efforts to confront the transnational nature of Muslim organizations operating under their national public law.

256

Immigration and the Transformation of Europe

These attempts to assert authority over transnational Islam look very similar to earlier moves to create centralized interlocutors for religious affairs. Public recognition has provided a tool-kit to modify or sever transnational ties; as Markell writes about the 1812 Prussian Emancipation legislation, the recognition of difference can be seen as “an instrument of, rather than a threat to, [state] sovereignty” (Markell 2003: 31). Markell views institutionalization as a “double bind” – a trade-off at the highest theoretical level of state building: “[Emancipation] was not conceived merely as the fulfillment of liberal principles of fairness or equality, nor was it simply the gift of an indifferent king who expected nothing in return. It secured recognition for the Jews, yet it also secured recognition for Prussia by placing Jews into a new relationship with the state” (Markell 2003: 141). Charting attempts at nationalization Scholarly attempts to explain and typologize these governmental actions have not closely examined the links between national State–Church organization and the Muslim world’s state-led and civil society networks that make up global Islam. Instead, they have focused on points of conflict and coercion. The existing literature offers little guidance to understand these governments’ efforts to grapple with the new social and political reality. Given my emphasis of the nation-state’s enduring importance to the integration of Islam in Europe, surprisingly little of my account flows from the particular national arrangements for the organization of religion. Instead, we are witnessing a broad, uncoordinated effort of institution building. I argue against three prevailing (and competing) misconceptions in the literature on the state accommodation of Islam in the West: the determinism of political opportunity structures; the impossibility of reconciling Islam and democracy; and the relevance of post-national dynamics. First, one tendency has been to characterize European nation-states’ policy response as falling into national models in a predictable fashion. This school of thought holds that the best predictor of Islam’s integration is national institutional or ideational trajectories (Minkenberg 2003; Mar´echal 2001). These authors argue that policies can be explained according to resource mobilization and opportunity structures: Cesari (1994), for example, writes that “different institutional arrangements tend to shape the agendas of Islamic mobilization and claims in different countries.” Fetzer and Soper (2005) suggest that conflict over the headscarf and school curriculum in France is merely the product of that country’s “long and contentious State–Church history”; similarly, they find that “inherited State–Church institutions best explain how Germany has accommodated Muslims” (2005: 94, 126). Using a similar logic,

Managing transnational Islam

257

Long and Zolberg conclude that any status improvements for Islam in Europe follow a routine trajectory of pluralist minority incorporation in different national settings (Long and Zolberg 1999). Second, other authors have advanced the notion that Islam is doctrinally unsuited for State–Church separation in contemporary western democracies and take the pessimistic view that Islam may simply be incompatible with Western principles (R´emond 1999; Sartori 2000). They argue it is impossible to politically integrate Islam within secular Western nation-states, and point to the novelty of the challenges posed by Islamic doctrine: notably, the reluctance to accept basic liberal precepts that distinguish spiritual rule from temporal authority (Modood 1998). In this view, Muslim leaders’ refusal to allow Islam to be relegated to the private sphere is compounded by their systemic incapacity to produce centralized representatives for the government to address – that is, a single organization that can “communicate and negotiate” on behalf of Muslims (R´emond 1999; Laurence 2001). “In contrast to Catholicism,” Warner and Wenner write, “the Islamic religion is not conducive to large-scale collective action [. . . . It is a] decentralized, non-hierarchical religions with multiple, competing schools [ . . . and has] no central authority to enforce cooperation or structure activity” (2002: 5–6). Thus, any state concessions to recognize Islam or to create “multicultural” policies benefiting Muslims’ religious identity are seen to weaken the foundations of liberal democracy because of Islam’s tendency to fuse religion and politics (Sartori 2000). This perspective predicts that governments will be hostile to demands made by Muslim communities and will avoid integrating Islam. Muslims, in turn, will be unable to take advantage of opportunities presented to them because of internal disunity. These authors predict continuous conflict between Islam and the state and a reluctance to make religious accommodations. Third is the argument that transnational forces are overcoming old national institutions, and that accommodation of Islam would occur only over the dead body of the nation-state. Soysal’s 1994 study of the incorporation of guest workers into West European social systems in the late 1980s and early 1990s – why states “extended rights and privileges of their citizens to migrant workers” – led to her theory of “post-national” rights acquisition. She contrasted this development with Hammar’s conception of “denizenship,” (1990) which focuses on the changes in citizenship on a territorial, nation-state basis – which Soysal called the “mere expansion of the scope of national citizenship”; instead, she claimed, the state is “no longer an autonomous and independent organization closed over a nationally defined population” (1994: 139, 169). Thus, immigrants turn to international actors and acquire rights only thanks to the

258

Immigration and the Transformation of Europe

increasing irrelevance of national citizenship regimes. The erosion of national institutions’ pertinence occurs by way of “constitutionally interconnected states with a multiplicity of memberships,” e.g., state membership in supranational organizations like the European Union and ECJ (European Court of Justice), or by way of international human rights norms established via the Universal Declaration and European Convention on human rights. According to this view, the integration of Islam takes place when national governments lose power and institutions associated with post-national processes gain authority. But Soysal’s framework ignores the cultural conflict and demands for religious accommodation that cause controversy at local and national levels. Her theory does not adequately address how states manage the nitty-gritty of family law and State–Church laws in religiously plural societies, domains that are still solidly under the competence of national governments. Her evocation of new post-national rights “to express and develop their cultural heritage” avoids the sticky question of the applicability of fatwas across borders, the propriety of wearing a headscarf in public institutions, the civil recognition of religious marriage and divorce, the obligation to wage holy war, the source of financing for prayer space and the training of imams, etc. These are all transnational challenges, but the basis for policy responses to them cannot be found in international institutions or human rights norms. After all, the ECHR (European Courts of Human Rights) respects national states’ rights to legislate the details of religious expression if public order is deemed to be threatened: the court upheld Turkish headscarf bans in public universities and its precedent was cited by French officials who drafted a law on religious symbols in public schools. Similarly, the French government felt secure enough in its position to promote the 2005 Treaty for a European Constitution, even though the document included a clause on freedom of religious expression “in private and in public.” My view is that states have not treated Islam as doctrinally impossible to integrate, and that they also regularly “jump” the purported categories of inherited institutional patterns. This goes against Soysal’s prediction for French State–Islam relations, for example, where she argues that the “unit of incorporation” is the individual citizen, and that “statist” strictures prevent “systematic representation or consultation with immigrant groups such that would promote a unified structure” (1994: 87). Indeed, I find that the central question is not, as Soysal suggests, about how Muslim groups adapt to state institutions within the overarching incorporative pressures of postnational rights. She claims that Muslim groups “adopt predominant national organizational models” and that a more centrally organized Islam will emerge in countries where the state actively

Managing transnational Islam

259

incorporates or defines Islam as a functional group. The more appropriate question is rather how states actively encourage such centralized structures to emerge as a reaction to the challenge of transnational religion. My argument also contradicts Fetzer and Soper’s perspective, in which the institutional opportunity structure alone determines the shape of Muslim accommodation. Governments have recognized that many of their integration challenges with regard to the population of Muslim origin requires reigning in certain transnational characteristics of religious communities. It is precisely in reaction to transnational forces that state action is framed. Where regime types matter is in the institutional opportunity structure: whether public recognition entails the creation of a council, a corporation of public law, or a treaty between religious community and state. The importance of national structures and historical statechurch settlements does determine which institutional actors will lead the process of nationalization, for example, the national interior ministry in France; or courts of law in Germany. And the specific narrative of national accommodation of Islam is of course particular to the historical circumstances of the given country’s interaction with the main Muslim sending states; French colonialism left a different legacy than Germany’s gastarbeiter program. But I find the similarity in the patterns of institutional outcomes to be striking, given distinctive national approaches to citizenship, divergent State–religion regimes and the political traditions of State–society relations. Of course the outcomes are far from identical, but governments whose doctrines would normally steer them away from religious affairs have all recognized the need to confront the transnational nature of this neglected religious community. By resorting to the use of corporatist-style mechanisms of political integration, governments have shown that the state is alive and well. This chapter must forego the rich detail of the options for public legal status that influence the precise path of Islam’s national institutionalization, in favor of identifying the commonalities and drawing attention to the broader themes at play and the mechanisms at work in State–Islam relations. Charting transnational Muslim organization These governments’ pursuit of state-centric agendas has required managing the competing interests and formal demands of Muslim organizations at home and abroad. It is helpful to view this story in terms of an evolving “supply and demand” for religious leadership for the activity of State–Church relations. For reasons related to the nature of civil society in the Arab-Muslim world, Roy argues, European states naturally looked to mosques and prayer rooms – not trade unions or political

260

Immigration and the Transformation of Europe

parties – as the “spontaneous form of organization” of Muslims in the West (Roy 2002: 19). The mirror image of European governments’ strategy with regard to Islam is the organized Muslim world’s pursuit of its own interests. Since the early 1970s, governments have been faced with a shifting supply of foreign and native-born Muslim religious representatives who operate through embassies or local organizational nodes of transnational movements. The first category of Muslim organizations, broadly speaking, represent the “official Islam” of “sending” states in the Muslim world; these organizations are concerned with protecting spheres of influence among e´ migr´e populations and with neutralizing threats to their own sovereignty from the growing Muslim Diaspora. The second category of organizations can be classified as “political Islam”: the dissident movements that sought refuge from repressive regimes and staked out operational bases in Western Europe. Official Islam – also known as the “Islam of the Embassies” – is characterized by the “sending” states’ foreign policies of ensuring “a cleric in every consulate catchment” (Zaptcoglu 2004). The enduring examples of such organizations are the Muslim World League, the Algerian and Moroccan consular services, and the Turkish Directorate for Religious Affairs. While Saudi Arabia did not send labor migrants to Europe, its government has sought to expand the kingdom’s religious influence outside the Arab world in part through the Muslim World League (MWL), a Mecca-based NGO founded in 1962. King Faisal bin Abdul Aziz (1964– 75) sought to enable independent Islamic diplomacy through commissioners in Saudi embassies, who would serve as MWL representatives; the MWL Paris bureau chief, for example, simultaneously served as an ambassador to UNESCO. Its charter called for making “direct contact with Muslim minorities and communities wherever they are [ . . . ] to close ranks and encourage them to speak with a single voice in defense of Muslims and Islam.” The MWL receives dozens of annual funding requests across Europe and has provided major financing for mosques across the continent, from Mantes-la-Jolie, Evry, and Lyon to Madrid, Rome, Copenhagen, and Kensington. The boom in Saudi proselytizing around the world – through the construction of grand mosques, the circulation of free Wahhabi prayer books, and the dispatching of missionaries and imams – was funded by petrodollars, at an estimated expense of more than $85 billion between 1975–2005, reflecting a determined effort to establish spiritual and political hegemony over Muslim practice (Hunter 1998: 158, 2002). In addition to this kind of grand strategy, there is also the everyday engagement of embassies and consulates that try to maintain control over e´ migr´e populations abroad. Algerian, Moroccan, and Turkish consulates,

Managing transnational Islam

261

for example, keep tabs on associations founded by their nationals abroad; they compile lists of friendly and unfriendly prayer associations, and either offer their support or report potential troublemakers to authorities of the home country or the host government. This Islam of the embassies seeks to retain a guardian status over the religious practice in the Diaspora. The emergence of Turkish, Algerian, and Moroccan federations in Europe that remain loyal to the official Islam of the homeland was no spontaneous event. Rather, consulates and embassies have encouraged – and even helped to administer – organizational structures that bring together the existing prayer associations founded in different European countries under a homeland banner. Here I will briefly discuss the official exportation of Algerian Islam in France and Turkish Islam in Germany. Shortly after World War I, the French president inaugurated the Grande Mosqu´ee de Paris and an association in charge of holy sites (GMP) in French Algeria. Though its first board of directors included Algerians, Moroccans, Tunisians, and Senegalese, the GMP gradually came under Algerian domination in the two decades that followed that country’s independence. In 1982, the Algerian government took over responsibility for the GMP’s finances and began using the mosque as a conduit for spreading its official state Islam, creating prayer spaces and attempting to co-opt existing ones following the post-1981 boom in prayer associations. The GMP is organized as a federation with five regional Muftis, and it currently controls 250 prayer spaces and associations around France. The GMP’s rector has authority over 150 imams (just over ten percent of all imams in France), most of whom are imported from Algeria. Alongside the GMP, one can look to the Turkish directorate for religious affairs (DIB) for a quintessential model of exported “official” Islam. Founded in 1950, this special administration in the prime minister’s office is responsible for the construction, administration, and staffing of mosques; its 60,000 clerics-civil servants help organize Qur’an courses and publication and censorship of liturgy. The DIB’s mission statement is to “instill love of fatherland, flag and religion,” and a portrait of Ataturk ¨ hangs in the front offices and foyers of DIB prayer spaces (Amiraux 2001). Like Algerian state Islam, the DIB lays claim on all Turkish citizens living abroad; it underwrites prayer space and religious education for Turks living abroad through local offices (DITIB), often staffing them with diplomats from Turkish consulates. The president of DIB in Turkey is the honorary chairman of every DITIB, and he may participate in DITIB membership and executive meetings (Lemmen 2000). Its prayer spaces in Europe are considered sovereign Turkish territory; when they join the national DITIB umbrella organization, the property is transferred to DIB

262

Immigration and the Transformation of Europe

and comes under the control of the Turkish interior ministry. The organization’s first German office was founded in 1982 in West Berlin, home to a large Turkish guest-worker population. Within two years, the federation had assembled 250 associations under its umbrella. In 1995, Germany’s DITIB employed 760 imams, each of whom was hired with the statute of public servant and a salary from the Turkish state for six-year terms. Known as the CCMTF in France, DIB offices abroad indirectly control half of all Turkish mosques in Europe. The two most influential international organizational networks of “political” Islam in Europe – also known as “dissident” Islam – are the Muslim Brotherhood (Ikhwan al-Muslimin, MB), of Egyptian inspiration, and the Germany-based Islamische Gemeinschaft Milli Gor ¨ us ¨ (IGMG), of Turkish background. The commonality among these political Islam organizations is their broad ideological outlook regarding the “inseparability of religion and politics” and their recruitment patterns, which attract individuals “far removed from traditional ulemas,” or religious authorities (Roy 2002: 29). The MB consists of a loose ideological network regrouping like-minded leadership figures, whereas IGMG provides formal organizational and financial support to Muslim associations across Europe. The MB was founded in 1928 by Hassan al-Banna (1906–49) in Egypt, the intellectual and political center of Islamism. The movement aimed to Islamize society from below by taking control of religious, academic, cultural, and social institutions. In the words of al-Banna: “Islam is faith and religion, country and nationality, religion and state, spirituality and action, book and spade” (Gritti and Allam 2001: 54). The Egyptian branch, in turn, is influential over Moroccan organizations in France (such as the UOIF). The non-governmental organizations based in the Arab-Muslim world – MB (Egypt and Syria), Refa party (Turkey), Jam’at-I Islami party (Pakistan) – maintain the “international nodes” of transnational forms of Islam (Eickelman 1997: 37). These are not simply political movements, Roy writes, but also a sort of religious brotherhood. These movements were able to spread in western countries through the 1970s and 1980s, Roy argues, thanks to globalized migration and communication technologies (2002). The Islamische Gemeinschaft Milli Gor ¨ us ¨ (National Vision, IGMG) was originally linked to the Refah party through the son of the party’s founder, Necmettin Erbakan. Established in Cologne in 1976/1982 as the Islamic Union of Europe (Islamische Union Europa), there are currently fourteen IGMG branches in Europe, including one in Brussels. The Cologne office is responsible for finances, while the Bonn office (established in 1994) oversees religious issues and mosque construction (called the Union of the New World Outlook in Europe, EMUG). Local Milli Gor ¨ us ¨ branches (called Islamic Federations in Germany) have

Managing transnational Islam

263

emerged as a major organizational force among Turks in Europe, principally as the arch-rival of the Turkish state’s directorate for religious affairs (DITIB). The IGMG defines itself in contrast to the DITIBs, which consider themselves as foreign organizations operating under diplomatic cover, and which the IGMG view as an obstacle to integration. The Union des Organisations Islamiques de France (UOIF) is the French branch of the FIOE. Headquartered in a defunct factory in the Parisian suburbs, the federation was founded in 1983 as a rejection of the Grand Mosque of Paris’s monopoly of representation, and as a result of other schisms amongst dissidents of “official” Franco-Algerian Islam. It federates approximately 250 of the many cultural, religious, and professional associations that have appeared since the 1981 reform that allowed foreigners to found associations. It claims control over 150 prayer spaces, but directly owns less than a third of these. The UOIF also runs a small theological seminary. Its current president and general secretary are both from Morocco, and both came to France to pursue advanced degrees in Bordeaux. Though the organization has no formal links to the Muslim Brotherhood, its president has used an MB slogan in interviews (“The Qur’an is our constitution”) and its general secretary meets regularly with a roving MB ambassador in Europe. UOIF representatives go on regular fundraising trips to the Gulf States and Saudi Arabia, partly with the help of the French offices of the Muslim World League and private donors. There are conflicting accounts as to whether it is one-third or two-thirds foreign-financed, but the organization’s directors speak openly of their wish to decrease their dependence on foreign aid. The federation maintains a “policy of non-intervention” with regard to its donors: the UOIF independently owns and administers the prayer spaces paid for with Saudi or Gulf-state money. The UOIF has sought to dispel any ambiguity that its sympathies lie with its adopted country, however, and in 1990 changed its name to the Union of Islamic Organizations of France (rather than “in” France). Two phases of strategies in Europe European governments have evolved from a laissez-faire policy of “outsourcing” State–Islam relations to Muslim diplomats (1974–89) towards a pro-active policy of “incorporation” (1989–2004), whose goal is to coopt the competing representatives of both “official” and “political” Islam described above. There were, at first, geopolitical and domestically rooted disincentives to engage Muslim minorities as if they were a permanent segment of national society. But this period was followed by new and powerful incentives to end a policy of laissez-faire. Just like the German government came to view the Turkish population in Germany as worthy

264

Immigration and the Transformation of Europe

of “einburgerung” ¨ (naturalization) by the late 1990s, Muslim religious representatives went from being treated as exogenous actors to serving as local government interlocutors (De Galembert 2000). This shift from outsourcing to incorporation led to a change in the demand for representative organizations in the practical and politically symbolic realm of State–Church relations. State–Islam relations emerged as the primary category of integration policy, replacing the emphasis on nationality or citizenship (Favell 1998; Bleich 2003). This has led to sustained efforts to institutionalize relations between religion offices and Muslim organizations along the lines of existing arrangements for other religions. It was not long after the end of mass migration in 1973–74 that national and local governments in Europe realized they would require an interlocutor in order to attend to the basic religious needs of their newly settled foreign populations: for example, prayer space, imams, and facilities for ritual lamb slaughter and travel visas for pilgrimage to Mecca (Boyer 1998). But host societies’ ambivalent attitudes towards the permanence of these new populations manifested themselves in the type of interlocutor that governments sought out. Official “return migration” incentives were in place into the early 1980s (Cesari 1994; Koopmans 1999), and a template of temporary migration defined the governments’ demand during the first phase. Guest workers and their offspring were not destined for citizenship; and Islam, as the religion of foreigners, was “an exogenous reality,” as De Galembert writes (2001). Governments thus largely entrusted embassies and representatives of the “official Islams” of the Muslim world, whether sending states (e.g., Algeria, Morocco, and Turkey) or centers of religious authority (e.g., Egypt, Saudi Arabia). Faced with a community of modest means, and given the legal and political difficulties of providing public funding, European governments encouraged the use of foreign funds for religious practice. From an electoral perspective, local officials undoubtedly viewed this as a safer route in the short term, at a time when extreme right parties were finding their bearings in response to the increasingly visible presence of immigrants in big cities by the late 1970s and early 1980s. This encouragement of a “home country” identity in the domain of religion dovetailed with a mutual fiction of an eventual “return home” for migrants and even their locally-born children. European governments tolerated the Islamic proselytism of foreign envoys from DITIB, MWL, and the GMP, for example, for a clear set of pragmatic reasons. The large, classical mosques that were planned and built across Europe during this period were justified as a fix for the practical needs of local Muslims. It made sense to rely on homeland governments for material requirements of religious observance since

Managing transnational Islam

265

those states had experience in its practical administration. Additionally, those homeland governments attended to the housekeeping of combating extremism in their own national interest. The implantation of official Islam in European national landscapes thus offered a security guarantee. As one French interior ministry official recalled, “Algeria, Morocco, Turkey, and Senegal were able to offer France a common front that was perhaps not pro-Western but at least anti-terrorist” (confidential interview, June 2002). Outsourcing (1974–89) The minimal accommodation of Islam that took place in the first period should be seen in this context. With the minor exception of prayer spaces created in some workplaces and public-housing units, governments outsourced relations to Muslim representatives to the embassies and consulates of sending states and the regional religious powerhouse, Saudi Arabia. When the local government in Bavaria created Turkish-language religious instruction in public schools in the early 1980s, for example, Turkish consular officials from DITIB were responsible for the curriculum and instruction; in the absence of religious education for Muslims in public schools in North-Rhine Westphalia, the Saudis created the King Fahd Akademie. French authorities allotted funds for Arabic language radio programs through the Fonds d’Action Sociale, and the foreign ministry created theological scholarships for foreign imams through a program known as Enseignement des Langues et Cultures d’Origine (ELCO). A rare European Community directive concerning migrant populations in 1976 allowed for mother-tongue classes to be sponsored by “sending” countries, taught by foreigners for third-country nationals in the European Community. What might ostensibly look like multicultural programs, however, were in fact the opposite: the relation with state Islam and homeland culture was intended to facilitate the eventual re-insertion at home. This policy of laissez-faire also served as a diplomatic nod to regional powers in the Muslim world, where a reshuffling of the power balance had taken place in the aftermath of Egypt’s 1973 defeat against Israel, and, in 1979, the Iranian revolution and the religiously-inspired coup attempt in Saudi Arabia (Rudolph 1997; Eickelman 1997). European countries sought to contain the regional aspirations of newly theocratic and Shi’ite Iran by supporting its Arab Sunni rivals in Saudi Arabia and the Maghreb. French President Giscard even sent troops to help alleviate the House of Saud’s difficulties with armed militants in Mecca. Tokens of good faith were offered by European governments eager to

266

Immigration and the Transformation of Europe

be on good terms with regional powers in the Arab world, who were the source not only of immigration but also of oil. In the aftermath of OPEC’s (Organization of the Petroleum Exporting Countries) oil embargo of the United States and The Netherlands in 1973–74, the Euro-Arab Dialogue (EAD) was institutionalized between twenty-one Arab states and the ten countries of the European Community. The EAD met several times a year to discuss trade issues alongside the theme of “cultural cooperation” (Benchenane 1983). Grandiose diplomatic gestures and monuments to the Islamic presence in Europe soon followed, such as the enormous Rome mosque at Monte Antenne (authorized by the local city council in 1974 in cooperation with Prime Minister Andreotti, who had personally asked for the Pope’s blessing of a minaret in the heart of western Christendom); the Belgian recognition of Islam as a national religion in 1974; and the MWL’s creation of major Islamic centers in Brussels and Vienna in 1975. The funds for mosque construction, decorations, and personnel salaries were provided by Saudi, Moroccan, Algerian, and Turkish embassies, and donors from the Gulf States – who were either solicited by non-state Islamic groups or the collaboration of state Islam with wealthy backers. Incorporation (1989–2004) Nineteen-eighty-nine marked a watershed year that initiated a second phase of State–Islam relations, after which governments sought to reassert state sovereignty over transnational Muslim networks. There were several confrontational events involving Islam in the international arena that year. First, the Ayatollah pronounced an unfavorable fatwa against Salman Rushdie for his allegedly blasphemous novel, The Satanic Verses. Then three headscarf-wearing girls were expelled from a junior high school outside Paris. Finally, that same year, Soviet troops withdrew from Afghanistan (Kepel 1991; Parekh 2000). The post-communist void in central Asia would soon reveal to Europe the extent of Saudi (and later, Turkish) institutional and financial deployment and proselytizing outside of the Arab world. These events pointed attention towards the European territory itself. These events reverberated within Muslim communities across the continent, and opened local governments’ eyes to the reality of transnational memberships among the minority populations. Soon thereafter, the Allied war to drive Iraq out of Kuwait provided further ripples across Muslim populations – where there were a few expressions of sympathy for President Hussein and a lack of understanding for the Saudi alliance with the United States-led coalition – and several incidents of Algerian jihadi terrorism in France in the mid-1990s culminated with a

Managing transnational Islam

267

deadly shoot-out between French special forces and a young Frenchman of Algerian origin. Cumulative integration failures among young Muslim Europeans contributed to the crystallizing sentiment that, as Heclo describes the moment preceding policy shifts, “spread a general conviction that something must be done” (Heclo 1974, 306). By the mid-late 1980s, the children and grandchildren of labor migrants had grown up, and the largely civic-based integration strategy had failed to achieve results in the second and third generations. The half-hearted strategies of inclusion that had stressed anti-racism or citizenship and electoral participation had, to a large extent, fallen flat: schools in the large urban centers that are home to populations of immigrant origin suffered from budgetary crises; military service was no longer obligatory; and voting rights had not led to much parliamentary representation. The shortcomings of the promise of integration and socioeconomic mobility were clear for all to see. Most significantly for the purposes of this chapter, however, by the late 1980s the convenient bargain of outsourcing was ultimately judged to be counterproductive in terms of the integration of Muslims. NGOs associated with the political Islam movements of transnational Muslim civil society in the Diaspora – such as the UOIF and IGMG discussed above – were increasingly assertive and behaved similarly to the peak associations of official Islam. They conglomerated sympathetic prayer rooms and cultural associations under common-law umbrella organizations that were ineligible for the status of State–Church associations, and therefore beyond the oversight and control of the government. In this same period, the second and third generations who had been expected to assimilate (or to return “home”) instead discovered religious identity in their new societies. Surveys have shown that many of these young people identified more with their inherited religion than with their nationality, place of residence or even gender – what might be called reIslamization or the “Ummah phenomenon” (Ummah refers to the larger Muslim community) (Bouzar 2004; Geisser and Finan 2002). Jean-Pierre Chev`enement, who as interior minister initiated the final round of consultations leading to the French Council for the Muslim Religion, pointed to this development in an interview: “It was only upon discovering that Islam was a form of identity affirmation for people who do not have much else that I [realized] it was important to engage in dialogue with these young people, who are having an identity crisis – we must not leave Islam outside” (personal interview, November 2003). This observation is emblematic of a broader change among public authorities, who adjusted their view of second and third generations from “youth of immigrant

268

Immigration and the Transformation of Europe

origin,” and came to see them, for limited public policy purposes, as “Muslims.” Authorities’ attitudes towards “official” Islam changed significantly during this period. It is not that European governments suddenly discovered the Islam of “sending” states to be anti-democratic or even fundamentalist; indeed, as “official” Islams they aim by definition for a peaceful relationship between religion and state. But their religious emissaries perpetuated a competing foreign tie among populations of immigrant origin – an Islam “in” rather than “of,” in politicians’ shorthand. Combined with a growing perception of transnational threats linked to global Islam, the maturation of migrant Muslim communities that retained strong organizational ties to homeland governments and nurtured connections with almost entirely unregulated dissident organizations of “political” Islam (Eickelman 1997: 31). In this second phase of State–Islam relations, Interior Ministries initiated consultations with a broader swathe of Muslim representatives, expanding their contacts with Muslims well beyond the “official” Islams of the homeland. This required delicate negotiations in which officials felt a need to tread lightly, and included not just diplomatic representatives (though they remained crucial) but also civil society organizations – including international NGOs affiliated with political or dissident Islam. This period has seen the reassertion of nation-state sovereignty over the informal influence of international religious NGOs and foreign embassies. This phase of Muslim incorporation has been about “de-transnationalization,” or undoing the power arrangement of the 1970s and 1980s that had privileged Saudi Arabia and other Muslim “sending” states in the practice of Islam in Europe – in addition to reigning in the unregulated associations of transnational “political” Islam active on national territory. Interior ministries provided the first impetus to organize Islam as a “national” religion, and the government-led consultations established a variety of national councils. In 1989, France began its fifteen-year journey to the Conseil Français du Culte Musulman; by 1992, Spain had a Comision Islamica; by 1998, Belgium had an Ex´ecutif Musulman; by 2000, there were special councils to guide State–Islam relations in seven German L¨ander; by 2002, official encouragement led to a newly consolidated Muslim Council of Britain; and by 2003, an Italian Consultation was initiated to create an interlocutor among the newly arrived Muslim communities there. These national processes are not identical, naturally: some place more weight on the role of “official” Islam and foreign government representatives (Belgium is seen as one case of this), while others rely more heavily on hand-picked local civil society organizations (Italy is a good example).

Managing transnational Islam

269

There are some striking differences: the French administration has been content to concentrate solely on religious representation: the CFCM is designed only to represent the 6–8 percent of observant Muslims who regularly attend prayer services (Laurence 2005). Italy has proposed something less official than a council: an informal consultation, or Consulta, which aspires not only to religious representation, but also social and political representation of the Muslim minority. Italian officials are aiming the creation of their Consulta at what Interior Minister Pisanu calls the 95 percent of moderate Muslims, whether observant or not: civil society representatives, secular Muslims, women’s groups, etc. (Laurence 2006). In Germany, by contrast, the councils in various L¨ander (since religion is a local competence) have resembled single-issue coalitions around specific tasks, such as organizing religious education in public schools. German accommodation of Islam has also heavily relied on court cases and a special role has been held by the office for the protection of the constitution, which has had the practical effect of excluding “political” Islam participants from consultations. But the commonalities of these national consultations with Islam in the second period are nonetheless very striking. European nation-states have undertaken a gradual institutional process of “de-transnationalizing” the practice of Islam. These governments have gone about integrating Islam into State–Church relations by negotiating a delicate settlement between “official” and “political” Islam, and they have used several specific instruments of nationalization familiar from previous instances of institutional incorporation. Space constraints prevent a full discussion of these instruments, but they include three crucial steps: a charter or founding document in which participating Muslim organizations confirm their respect for the rule of law; the establishment of technical working groups that include representatives of official and political Islams alongside state representatives; and, crucially, the nomination or election of a representative council that can serve as an interlocutor for State–Church affairs.

Conclusion The emergence of organized Islam as a permanent fixture in the European religious landscape was underway long before September 11, before a surge in anti-Semitic acts, or the departure of young British and French Muslims on suicide missions to London, Israel and Iraq. Contrary to the announcement of the state’s imminent death in the face of globalization and transnationalism, we can see that government activity is alive and well. European nation-states have established routines of contacts with

270

Immigration and the Transformation of Europe

Muslim leaders, leading to a new level of mutual acquaintance and a slow but steady process of nationalization of religious authority. State–Church relations are of vital importance because these institutional links with religious communities provide key elements of political integration. If unattended to, as it was in the first period of “outsourcing,” transnational religious networks have the potential to threaten the state and its maintenance of social order. By taking the initiative to incorporate and nationalize Islam in their respective institutional orders, European states have attempted to influence what kind of Islam young people discover – whether they search out religion as a reaction against European societies, or whether they are just satisfying curiosity about their heritage, or carrying on family traditions. Church–State relations are instrumental in achieving the state’s core duties of stability and security. Governments took into consideration the unintended consequences of their previous laissez-faire strategies in State-Islam relations and took stock of unanticipated developments among the immigrant populations. Of course, these populations did not “go home” and the networks of embassies and NGOs whose religious activities and proselytism European governments had uncritically tolerated for fifteen years turned out to be more tenacious than expected. But the strategy in the first period of keeping religion private, of keeping Islam out of the public sphere, and of using international diplomacy to manage the religion of immigrants was judged to be a failure. The national governments have assumed an active posture in State–religion affairs because Islam has emerged as a major factor of individual and group identity among the descendants of labor migrants.  AlSayyad, Nezar and Manual Castells 2002 Muslim Europe or Euro-Islam: Politics, Culture, and Citizenship in the Age of Globalization. Lanham, MD: Lexington Books. Amiraux, Val´erie 2001 Acteurs de l’islam entre Allemagne et Turquie. Paris: L’Harmattan. Baker, Don 1997 “World Religions and National States: Competing Claims in East Asia.” Chapter 6 in Transnational Religion and Fading States, Susanne Hoeber Rudolph, and James P. Piscatori (eds) Boulder, CO: Westview Press. Benchenane, Mustapha 1983 Pour un Dialogue Euro-Arabe. Paris: BergerLevrault. Bentley, Arthur Fisher 1949 The Process of Government: A Study of Social Pressures. Evanston, IL: Principia Press of Illinois. Bleich, Erik 2003 Race Politics in Britain and France: Ideas and Policymaking since the 1960’s. New York: Cambridge University Press. Bouzar, Dounia 2004 “Etude de 12 associations a` r´ef´erence musulmane: l’islam entre mythe et religion: le nouveau discours religieux dans las associations

Managing transnational Islam

271

socio-culturelles musulmanes.” Les Cahiers de la S´ecurit´e Int´erieure No 54. Paris: Institut National des Hautes Etudes de S´ecurit´e (IHESI). Boyer, Alain 1998 L’Islam en France. Paris: Presses Universitaires de France. Caldwell, Christopher 2004 “Islamic Europe? When Bernard Lewis Speaks.” The Weekly Standard, October 4. www.weeklystandard.com/Content/Public/ Articles/000/000/004/685ozxcq.asp. ˆ Musulman en France: Associations, Militants et Mosqu´ees. Cesari, Jocelyne 1994 Etre Paris: Karthala. Dassetto, Felice, Brigitte Mar´echal, Jørgen S. Nielsen and Stefano Allievi 2001 Convergences Musulmanes: Aspects Contemporains de L’islam dans l’Europe ´ Elargie. Paris: Harmattan. ´ De Galembert, Claire 2000 “Etat, nation et religion dans l’allemagne r´eunifi´ee.” Vingti`eme Si`ecle No. 66 (April–June): 37–51. ´ De Galembert, Claire 2001 “La R´egulation Etatique du Religieux a` L’´epreuve de la Globalisation” in La Globalisation du Religieux, Jean Pierre Bastian, Françoise Champion, and Kathy Rousselet (eds). Paris: L’Harmattan. Eickelman, Dale F. 1997 “Trans-State Islam and Security,” in Transnational Religion and Fading States, Susanne Hoeber Rudolph and James P. Piscatori (eds). Boulder, CO: Westview Press. Evans, Peter B., Dietrich Rueschemeyer and Theda Skocpol (eds) 1985 Bringing the State Back In. Cambridge: Cambridge University Press. Favell, Adrian 1998 Philosophies of Integration: Immigration and the Idea of Citizenship in France and Britain. 1st edn. New York: St. Martin’s Press in association with Centre for Research in Ethnic Relations, University of Warwick. Ferguson, Niall 2004 “Eurabia?” New York Times, Sunday, April 4, 13. Fetzer, Joel S. and J. Christopher Soper 2005 Muslims and the State in Britain, France, and Germany. Cambridge: Cambridge University Press. ´ Geisser, Vincent and Khadija Finan 2002 “L’Islam a` l’Ecole.” Paris: Institut National des Hautes Etudes de S´ecurit´e. Gresh, Alain and Tariq Ramadan 2000 L’Islam en Questions. Paris: Sindbad/Actes Sud. Grillo, Ralph 2004 “Islam and Transnationalism.” Journal of Ethnic and Migration Studies 30(5) (September): 861–78. Gritti, Roberto and Magdi Allam 2001 Islam, Italia: Chi Sono e Cosa Pensano i Musulmani Che Vivono Tra Noi. Milano: Guerini. Hammar, Tomas 1990 Democracy and the Nation State: Aliens, Denizens, and Citizens in a World of International Migration. Aldershot: Avebury. Heclo, Hugh 1974 Modern Social Politics in Britain and Sweden: From Relief to Income Maintenance. New Haven, CT: Yale University Press. Hunter, Shireen 1998 The Future of Islam and the West: Clash of Civilizations or Peaceful Coexistence? Westport, CT: Praeger. Hunter, Shireen 2002 Islam, Europe’s Second Religion: The New Social, Cultural, and Political Landscape. Westport, CT: Praeger. Kepel, Gilles 1991 Les Banlieues de l’Islam: Naissance d’une Religion en France. Paris: Editions du Seuil. Koopmans, Ruud 1999 “Germany and Its Immigrants: An Ambivalent Relationship.” Journal of Ethnic and Migration Studies 25(4) (October): 627– 47.

272

Immigration and the Transformation of Europe

Krasner, Stephen D. 1999 Sovereignty: Organized Hypocrisy. Princeton, NJ: Princeton University Press. Laurence, Jonathan 2001 “(Re)constructing Community in Berlin: Turks, Jews, and German Responsibility.” German Politics and Society 19(2) (Summer): 22–61. Laurence, Jonathan (guest editor) 2005 “The French Council for the Muslim Religion.” French Politics, Culture and Society 23(1). Laurence, Jonathan 2006 “Knocking on Europe’s Door: Islam in Italy.” US Europe Analysis Brief. Washington, DC: Brookings Institute. Lemmen, Thomas 2000 Islamische Organisationen in Deutschland. Bonn: Friedrich-Ebert-Stiftung. Long, Littwoon and Aristide Zolberg 1999 “Why Islam is like Spanish.” Politics and Society 27(1): 5–38. Mar´echal, Brigitte 2001 “Mosqu´ees, organisations et leadership,” in Convergences ´ Musulmanes: Aspects Contemporains de l’Islam dans l’Europe Elargie Felice Dassetto, Brigitte Mar´echal, Jørgen S. Nielsen, and Stefano Allievi (eds). Paris: Harmattan. Markell, Patchen 2003 Bound by Recognition. Princeton, NJ: Princeton University Press. McConnell, Michael W. 2000 “Believers as Equal Citizens.” Chapter 3 in Obligations of Citizenship and Demands of Faith: Religious Accommodation in Pluralist Democracies, Nancy L. Rosenblum (ed). Princeton, NJ: Princeton University Press. Mendes-Flohr, Paul R. and Jehuda Reinharz 1995 The Jew in the Modern World: A Documentary History. 2nd edn. New York: Oxford University Press. Minkenberg, Michael 2003 “The Policy Impact of Church–State Relations: Family Policy and Abortion in Britain, France, and Germany.” West European Politics 26(1): 195–217. Modood, Tariq 1998 “Multiculturalism, Secularism and the State.” Critical Review of International Social and Political Philosophy (CRISPP) 1(3) (Autumn): 79–97. Noiriel, G´erard 1988 Le Creuset Français: Histoire de l’Immigration, XIXe-XXe si`ecles. Paris: Seuil. Parekh, Bhikhu C. 2000 Rethinking Multiculturalism: Cultural Diversity and Political Theory. Cambridge, MA: Harvard University Press. R´emond, Ren´e 1999 Religion and Society in Modern Europe. Oxford: Blackwell Publishers. Rosenblum, Nancy L. (ed.) 2000 Obligations of Citizenship and Demands of Faith: Religious Accommodation in Pluralist Democracies. Princeton: Princeton University Press. Rousseau, Jean-Jacques, Roger D. Masters and Judith R. Masters 1978 On the Social Contract, with Geneva Manuscript and Political Economy. New York: St. Martin’s Press. Roy, Olivier 2002 L’Islam Mondialis´e. Paris: Editions du Seuil. Rudolph, Susanne Hoeber 1997 “Introduction: Religion, States, and Transnational Civil Society,” in Transnational Religion and Fading States, Susanne Hoeber Rudolph and James P. Piscatori (eds). Boulder, CO: Westview Press.

Managing transnational Islam

273

Sartori Giovanni 2000 Pluralismo, Multiculturalismo e Estranei: Saggio Sulla Societ`a Multietnica. 1st edn. Milano: Rizzoli. Savage, Timothy M. 2004 “The Changing Face of Islam.” The Washington Quarterly (Summer): 25–50. Shore, Zachary 2004 “Breeding New bin Ladens: America’s New Western Front.” Watch on the West (A Newsletter of the Foreign Policy Research Institute’s Center for the Study of America and the West) 5(11) (December) www.fpri.org/ww/0511.200412.shore.newbinladens.html. Soysal, Yasemin Nuhoglu 1994 Limits of Citizenship: Migrants and Postnational Membership in Europe. Chicago, IL: University of Chicago Press. Warner, Carolyn M. and Manfred W. Wenner 2002 “Organizing Islam for Politics in Western Europe.” PRPES Working Paper No 17. Cambridge, MA: Project on Religion, Political Economy, and Society; Weatherhead Center for International Affairs; Harvard University, Summer. www.wcfia.harvard.edu//papers/554 WarnerWennerIslampaper.doc. Weber, Eugen Joseph 1976 Peasants into Frenchmen: The Modernization of Rural France, 1870–1914. Stanford, CA: Stanford University Press. Wright, Lawrence 2004 “The Terror Web: Were the Madrid Bombings Part of a New, Far-reaching Jihad being Plotted on the Internet?” The New Yorker, August 2. Ye-or, Bat 2005 Eurabia: The Euro-Arab Axis. Madison, NJ: Fairleigh Dickinson University Press. Zaptcioglu, Dilek 2004 “Turkey’s Religion Council: Setting Guidelines for Islam and Politics.” www.quantara.de, May 10.

11

Migration mobility in European diasporic space Jacqueline Andall

Historical or cultural ties are important determinants of migration flows, as emphasized in chapter 3 of this volume. But recent migration trends have increasingly meant that labor migrants from countries with no previous strong historical or cultural ties to destination countries can be found throughout Europe.1 This chapter focuses on these new configurations of migration flows to Europe and explores how contemporary African labor migrants select their country of destination and settlement in Europe. The presence of a particular migratory group in a specific European country does not necessarily reflect an active choice, as another location (within or outside of Europe) may have been the preferred destination. What makes one European country more or less attractive and how do different state immigration regimes, local labor markets or the existence of a dispersed European component of a global diaspora impact on destination decisions and opportunities? Understanding migrants’ destination and settlement choices involves viewing the immigration regimes and labor market opportunities of single European countries as part of a complex European and global whole. In this chapter, I concentrate specifically on Ghanaian migration. During the 1990s, Ghanaians emerged as one of the largest groups of sub-Saharan Africans in European Union (EU) countries, with their movement classified primarily as a labor diaspora (Cohen 1997; van Hear 1998).2 I suggest that in a period of restricted access to labor migration in Europe, labor market opportunities and the quest for legality are significant factors shaping Ghanaian destination, settlement, and mobility decisions. However, the presence of Ghanaian communities dispersed across many 1

2

Italy is a very clear example of such a trend, especially if we compare the presence of nationals from Italy’s ex-colonies to other foreign nationals in Italy. Home Office data for 2003, shows that while there are only 6,318 Eritreans and 5,148 Somalis in Italy, there are 41,539 Sri Lankans, 73,847 Filippinos, and 23,060 Ghanaians (Caritas di Roma 2004). According to Eurostat migration data for 1995, there were 78,995 Ghanaians in European Union countries. This figure can be compared to the Senegalese (76,766); Nigerians (71,997); Zaireans (53,787); Ethiopians (51,446); and Somalis (47,507), cited in Mollel (2000).

274

Migration mobility in European diasporic space

275

European countries and knowledge sharing within this European diasporic space does play a role in directing country choice. Here, I establish how Ghanaian labor migrants “choose” Italy as a destination country, highlighting the significance of immigration regimes in EU countries, diaspora networks in Europe, informal and formal labor markets in Italy and the role of Italian immigration amnesties.3 Migration, mobility, and settlement Recent scholarship on post-cold war migrations has increasingly applied the concept of mobility to the type of movements occurring in central and eastern Europe (CEE) and has generally tended to differentiate between mobility and migration (Williams and Bal´azˇ 2002; Morokvasic 2003). Morokvasic (2003) not only contends that short-term shuttle mobility is the dominant characteristic of post-Berlin Wall migrations, but she additionally maintains that this pattern of mobility is “an alternative to emigration” (111), and indicative of migrants who “settle within mobility” (102). These migrants do not attempt to settle in the destination country, rather their mobility allows them to “improve or maintain the quality of life at home” (Morokvasic 2004:11). The new mobility patterns associated with “commuting” or “circulation” generally refer to movement between two countries (De Filippo 1994; Spano` and Zaccaria 2003). An additional typology of mobility which warrants further investigation relates to the sequential movement of migrants which involves varying degrees of settlement. While the settlement process constitutes an important dimension of migration research and has been a dominant paradigm for the analysis of African, Asian, and Caribbean migration to Europe, the research findings presented here indicate that we should broaden conceptual understandings of settlement which generally imply permanence. A transitory aspect characterises contemporary settlement patterns when aspirations regarding specific European countries remain unmet.4 The nature of immigration policies in Europe partially explains the emergence of different forms of mobility. The relaxed visa requirements for certain CEE countries introduced in the post-Berlin Wall era meant 3

4

The data presented draws from a qualitative research project conducted in Italy and Ghana between 2003–05. In Italy, semi-structured interviews were conducted with thirtyone Ghanaian nationals and reflect male patterns of mobility to Europe. Interviews were also conducted with local council officials, voluntary sector workers and Italian and Ghanaian trade unionists. These transitory forms of settlement should be considered as different from temporary or seasonal migration where the migration is for a defined period from the outset.

276

Immigration and the Transformation of Europe

that CEE nationals were able to circulate within, rather than settle in, EU countries “exploring opportunities, choosing more attractive destinations and abandoning those that [had] become difficult to access or to remain in” (Morokvasic 2004:12). By contrast, immigration policies in Europe have impeded rather than facilitated such mobility opportunities for African migrants (Carling 2002). As a consequence, contemporary African labor migrants can psychologically and physically exist in a process of “migration mobility” rather than in a process of permanent settlement. Instead of being “settled in mobility,” as with Polish commuters to Germany (Morokvasic 2004), they often continue to be mobile in order to identify an appropriate country of settlement. This does not mean that they are continuously mobile as they may spend several years in one country before movement onto another. So at what point can they be appropriately described as “settled” if they have active or latent intentions for further mobility and what factors induce them to consider sequential movement? The psychological migration mobility of such migrants is to be differentiated from Anwar’s (1979) “myth of return” as it is not determined by a desire to return to the home country. Instead, it describes migrants’ readiness to move to a different region/country within and beyond Europe should better opportunities arise.5 Physical migration mobility, on the other hand, is shaped by migrants’ ability to negotiate a way through the institutional and practical impediments to further movement. The demise of established colonial routes directing migration from Africa to Europe means that Senegalese, Cape Verdeans, or Ghanaian migrants no longer automatically migrate to France, Portugal, or the United Kingdom, respectively. Instead, migration mobility, following departure from the sending country, is facilitated by the emergence of African diasporas within Europe, which increasingly encompass a wider range of countries. Involuntary immobility (Carling 2002), previously applied to migration intention from the developing world to the West, becomes applicable to migration intention within the geographical space of Europe. The new forms of mobility to and within Europe are also gendered. While gendered labor market opportunities in Europe have contributed to the feminisation of some migratory streams, transnational caring responsibilities lead to gendered mobility actions (Andall 1999; Parrenas ˜ 2001; 5

The home country is not excluded from the framework as during the period of migration abroad, opportunities may arise which make it more profitable to return home. In contrast to the literature on transnationalism where the original “home” or sending country occupies a dominant place within the transnational framework (Basch et al. 1994; Portes et al. 1999), I maintain that following initial migration, the “home” country in the transnational field can develop into an integral rather than a dominant component of the transnational community.

Migration mobility in European diasporic space

277

Morokvasic 2003). Women and men’s assessment of what forms of mobility or migration may be risky or dangerous can also differ, as well as the issues they prioritise in relation to decision making regarding internal or international mobility. This partially explains, for example, why in Italy, a predominantly female group such as Cape Verdeans have demonstrated limited internal mobility within Italy (Andall 2005), while Senegalese nationals, the majority of whom are male, have been very mobile (Riccio 1999). In the following section, I address the growth of African Diasporas in Europe. I then examine the emergence of Ghanaians as a migratory group, focusing on why and how they choose to migrate to Italy. In the final section, I present the complex mobility patterns of selected interviewees. The proliferating channels of African migration to Europe By the early 1970s, the immigration regime change in several European countries meant that African migrants could no longer follow familiar routes to the former colonial metropolis. Data in relation to various African groups clearly demonstrate that nationals from specific sending countries are now migrating to a much wider range of countries (see Tables 11.1–11.6 below).6 The European component of a global diaspora can vary in terms of whether the diaspora is categorized as a labor or political diaspora or whether the diaspora is present in a limited or wide range of European countries. The gender composition of components of the diaspora can also vary,7 so, too, can the nature of its incorporation into local labor markets. These characteristics contribute to the variety of mobility actions occurring within different diaspora groupings. These data represent the European dimension of global diasporas, but we should note that for several groups, North America and Canada are major components of these diasporas. In 2001, for example, there were 97,000 Ghanaians in the United States and 16,985 in Canada.8 This figure was 73,000 and 14,345, respectively, for Ethiopians and 20,340 Somalis were also settled in Canada. Although my focus in this chapter is on African migration, these typologies of dispersal are applicable 6

7 8

Data record citizenship and thus do not capture the full extent of the size of the diaspora as it excludes those who have taken European citizenship and those who are undocumented. I have included countries which have registered presences of more than 1,000 nationals. For example, in the 1970s, Cape Verdeans in The Netherlands were primarily male, while Cape Verdeans in Italy were almost exclusively female (Andall 1999). Data is drawn from the statistical compilations of the Migration Policy Institute and data for Ghana is reported by country of birth.

278

Immigration and the Transformation of Europe

Table 11.1 Ghanaian nationals by citizenship in selected EU countries (1998) Belgium Germany France Italy Netherlands UK

1,260 22,170 2,809 15,293 4,375 23,000

Source: European Commission (2000) European Social Statistics: Migration.

Table 11.2 Somali nationals by citizenship in selected EU countries (1998) Denmark Germany France Italy Netherlands Sweden UK

11,890 9,035 1,077 10,787 13,645 13,122 53,000

Source: European Commission (2000) European Social Statistics: Migration.

Table 11.3 Nigerian nationals by citizenship in selected EU countries (1998) Germany Spain Italy Netherlands UK

15,919 1,092 9,486 1,675 44,000

Source: European Commission (2000) European Social Statistics: Migration.

to other migratory groups. For example, in 2001, Sri Lankan nationals were widely dispersed in Germany (46,632); Canada (91,670); the Netherlands (6,457); Sweden (5,964); and Italy (34,464).9 9

Data for Italy is drawn from the Caritas dossier which collates Home Office immigration data on an annual basis. Canada, Sweden and The Netherlands are reported by country of birth, and Germany and Italy by country of nationality.

Migration mobility in European diasporic space

279

Table 11.4 Ethiopian nationals by citizenship in selected EU countries (1998) Germany Greece Italy Netherlands Sweden UK

18,744 1,095 7,693 1,870 3,407 5,000

Source: European Commission (2000) European Social Statistics: Migration.

Table 11.5 Senegalese nationals by citizenship in selected EU countries (1998) Germany Spain France Italy

2,681 5,328 43,692 25,737

Source: European Commission (2000) European Social Statistics: Migration.

Table 11.6 Democratic Republic of Congo nationals by citizenship in selected EU countries (1998) Belgium Germany France Italy Netherlands UK

12,130 17,612 22,740 2,292 2,690 5,000

Source: European Commission (2000) European Social Statistics: Migration.

The case of Ghanaian migration and mobility Ghana’s history as a receiving immigration country in the West African region led observers to predict that Ghanaians would remain a “nonmigratory” group into the 1980s (Yeboah 1986). In fact, the migratory potential of Ghanaians would be rapidly transformed in the 1980s and 1990s. Poor trade terms in the 1970s, high inflation, and drought in the 1980s all contributed to a difficult economic situation (Yeboah 1986;

280

Immigration and the Transformation of Europe

Makinwa-Adebusoye 1995). Following a coup d’´etat in 1979, the country’s new leader, Jerry Rawlings, entered into agreements with both the World Bank and the International Monetary Fund in order to access foreign aid, leading to a reduction in the civil service, the devaluation of the currency and a reduction in salaries (Country Watch 2004). This economic background led to a migration turnaround in Ghana. Both skilled and unskilled Ghanaians began to emigrate in growing numbers and it is estimated that approximately 2 million Ghanaians left either for Nigeria or for the Ivory Coast between 1974–81 (Peil 1995).10 When the Nigerian economy began to struggle in the early 1980s, it expelled over 1 million Ghanaians in 1983 and implemented further expulsions in 1985 (Anwar 1979).11 These expulsions imposed limits on an important migration outlet and the “delayed diasporization” of Ghanaians in the 1980s and 1990s has been attributed to the partial removal of Nigeria as a migration destination country (van Hear 1998: 211). A number of studies confirm that Ghanaian communities in a range of countries began to emerge from the 1980s onwards. For example, the number of Ghanaians migrating to Canada substantially increased from a low baseline of approximately 150 entries per year during the 1970s to 11,070 by 1991 (Owusu 2000). Some 80 percent of this total had arrived between 1981 and 1991. In Germany, the majority of Ghanaians arrived between the late 1970s and early 1990s (Niewsand 2004). In The Netherlands, the Ghanaian population practically doubled in the mid-1990s (Mollel 2000). Reliable data regarding the presence of labor migrants in Italy is unavailable for the 1980s; however, it is probable that during this decade increasing numbers of Ghanaians began to arrive, since, in 1990, 11,443 Ghanaians were recorded as being legally resident in the country (Caritas di Roma 1992). As Peil (1995: 357) maintained in the mid-1990s: “There seems to be no major emigration stream for moves outside Africa; individuals take whatever opportunities they find, according to their education, training and contacts.” These migrations, and the consequent emergence of new destination countries for Ghanaian migrants, have meant that despite transformations in the Ghanaian economy and political democratisation processes, a migratory culture has become established (van Hear 1998: 210–11). Ghanaian labor migrants 10

11

In the early 1980s some 50 percent of Ghanaian doctors emigrated (Adepoju 1995). There are continuing health implications for Ghana as a result of this brain drain in the health sector. For example, it is estimated that some 50 percent of Ghana’s professional nurses have migrated over the last ten years to the UK, the US, and Canada (Nyonator et al. 2004). Ghana had expelled “aliens” in 1969.

Migration mobility in European diasporic space

281

are currently present in a wide range of African countries12 as well as in Europe, North America, and Canada. In Europe, there is a longer history of Ghanaian settlement in the United Kingdom as a result of colonial ties. Some 33,000 Ghanaians were identified in the British 1991 Census, primarily settled in the London area (Peach 1996; Ackah 2000). By 2001, additional important concentrations of Ghanaians in Europe were to be found in Germany (23,179), Italy (17,971), and The Netherlands (10,998).13 Ghanaian migration to Italy It is by no means obvious why Ghanaian migrants would choose Italy as a destination country above other European countries. There was no colonial relationship between the two countries, neither is there any common language or cultural affinity. As I show below, contemporary labor migrants from Ghana make their migration decisions against a framework of shifting opportunities. Their country aspiration when they are considering migration or when they are accumulating the necessary finances to embark on a migratory project may be transformed at the moment of migration. In this section, I describe the general characteristics of Ghanaian migrants in Italy, as the growth of a diaspora presence has contributed to Italy representing a known European destination country in the migratory repertoire of potential migrants. Labour immigration into Italy began slowly in the 1960s and 1970s and, by 2003, there were 2,193,999 legally resident non-nationals in Italy (Andall 2000; Caritas di Roma 2004). There are three dominant and inter-related factors which characterise the attractiveness of Italy in the context of migration to Europe. Firstly, there are opportunities for regular work within Italy’s formal economy.14 Secondly, there is an extensive informal economy in Italy that can readily incorporate undocumented migrants. It has been estimated that “the percentage of irregularly 12

13 14

Notably in Togo, Ivory Coast, Nigeria, Libya, and South Africa (Peil 1995). In 2003, there were an estimated 25,000 Ghanaians in Libya for example (Daily Graphic, June 19, 2003). The Netherlands is reported by country of birth, while Germany and Italy by country of nationality. There is an accepted need for skilled and unskilled labor in Italy and annual quotas are established regarding entry. In 2001, for example, the Ministry of Labor authorized 83,000 labor migrants, including 39,400 for seasonal work, 5,000 for nurses and workers in the IT sector and 15,000 for those in search of work but with a guarantor in Italy (Caritas di Roma 2002). Three broad models of regional labor market opportunities have been identified – the metropolitan model of urban conurbations (mainly service oriented – domestic work, restaurant work, etc.), the Southern model (services and agricultural work), and the industrial model of the North East (Ambrosini 2003; Reyneri 2004).

282

Immigration and the Transformation of Europe

employed immigrant workers never sinks below 31 percent, more than double the level for Italian workers” (Reyneri 2004:83). Thirdly, the use of immigration amnesties to manage immigration to the country facilitates migrants’ “settlement.” Over the past twenty years, no less than six immigration amnesties have been introduced to legalize migrants irregularly present in Italy (Carfagna and Pittau 2003). These Italian migration characteristics are not necessarily known to migrants before they migrate; instead they adopt a ‘go and see’ attitude, encountering specific conditions in Italy which lead them either to attempt settlement or to envisage further mobility. The labor migration of Ghanaians to Italy began around 1983–84, although there was some student migration prior to this (De Filippo 1993). In 2001, of the 17,971 Ghanaians present in Italy, some 16,246 were present in Italy’s wealthier Northern regions. The largest number of Ghanaians were domiciled in the Veneto (5,557), Lombardy (5,241), and Emilia Romagna (3,681) regions. The presence of Ghanaians in Italy’s southern regions, where the economy is weaker, is small – 510 in the Campania region (primarily Naples [385] and Caserta [124]); 560 in Sicily (primarily Palermo [539] and 56 in Puglia [48 in Foggia]) (unpublished data, Ministero degli interni). The limited numbers of Ghanaians in the south do not reflect actual presences, however, and disguise a generalized pattern of Ghanaian entry and mobility within Italy. The early Ghanaian migrants to Italy worked as undocumented labor in the agricultural sector in the south, only moving north to the industrial sectors once they had regularized their stay following a general immigration amnesty (Caritas di Roma 1992).15 This pattern of internal mobility was noted in the early 1990s (Pugliese 1990) and was confirmed as a common praxis during interviewee narratives. Interviews conducted with undocumented Ghanaian migrants in the south of Italy indicated the endurance of this pattern of migration and mobility as they all articulated a clear migration intention to move to the north as soon as they had regularized their situation. The largest community of Ghanaians reside in the Veneto region and numerically they represent the sixth largest ethnic minority group in the region. In the Veneto region itself, the main concentrations of Ghanaians are in Verona (1,924) and Vicenza (2,521). Today, the Veneto region has the third largest number of foreign nationals residing in the country after Lombardy and Lazio, indicating important transformations in the regional distribution of labor migrants in Italy.16 In the region as a whole, 15 16

A similar pattern has been noted for Senegalese labor migrants (Riccio 1999). In the early 1990s, the Veneto was not a region with a significant number of migrants when compared to other Italian regions. In 1991, the regional dimension of foreign

Migration mobility in European diasporic space

283

resident migrants make up 3.1 percent of the population and 4.4 percent of the province of Vicenza (Veneto Lavoro 2001). The Ghanaian presence in Italy is rising, up from 17,971 in 2001 to 19,160 in 2002, and to 23,060 in 2003. The Ghanaian presence has doubled over the last decade, while a different picture emerges for some other ethnic groups. Thus, the number of Ghanaian nationals grew from 11,443 in 1990 to 19,160 in 2002, while the number of Somalis fell from 11,842 to 4,894 during the same time frame (Caritas di Roma 1992; 2003).17 These data indicate that Ghanaians continue to view Italy as a viable destination choice, although it does not explain how and why they choose to migrate to Italy above other European countries. In the following section, I explore how Ghanaians select Italy as a destination country and discuss whether this reflected their aspirations in terms of destination choice. Aspiration, choice, and Italy While not the principal focus of this chapter, it is important to note that Ghanaian labor migrants are also an attractive proposition for Italian employers and Italian industry. Labor market opportunities in Italy constitute a critical dimension of the migration process. It is no accident that Ghanaians have been internally mobile within Italy and settled in the Veneto region. The Veneto region has developed a model of industrial development centerd on small and medium-sized enterprises geographically concentrated in industrial districts (Anastasia 1992; Anastasia and Coro 1992; Angelillo and Occari 1988). In the 1990s, the dynamic and internationally competitive nature of these districts began to be affected by a lack of available Italian labor, leading to the employment of migrant labor in the sector (Guolo 2001). A case in point is the leather industrial district in Arzignano (Vicenza) where the research with Ghanaians was primarily conducted. The Italian leather industry is an important industry, comprising some 70 percent of European leather production (EBAV 2001). Arzignano produces 40 percent of tanned leather in Italy and constitutes one of the two dominant leather-producing areas in the country. Arzignano has the second highest number of migrants employed

17

nationals’ presence was in numerical order, the Lazio region (201,961), Lombardia (149,985), Tuscany (69,816), Sicily (64,514), Emilia Romagna (61,380), Campania (53,639), Veneto (48,462) (Caritas di Roma 1992). Other labor diasporas to Italy show different levels of growth. Thus, the presence of Indian nationals grew from 11,282 in 1990 to 34,080 in 2002, while the number of Senegalese nationals grew from 25,107 in 1990 to 36,310 in 2002 (Caritas di Roma 1992; 2002).

284

Immigration and the Transformation of Europe

Table 11.7 Mobility patterns of selected interviewees1 Aspiration

First destination

Second destination

Third destination

Unknown Asylum seeker Europe Japan Europe Europe N/A UK USA Europe

Germany (d) Germany (d) Nigeria (x2) Thailand (irr) Libya (irr) Nigeria (exp) Algeria (stu) Libya (irr) Holland (irr) Malaysia (stu)

Austria (d) (asp. Canada) Italy (1996) Italy (1997) United Kingdom (d) Italy (2002) Italy (1986) Italy (1989) Spain (irr) Italy (2000) Denmark (irr)

Italy (1993) Holland (m) Italy (1998)

Italy (2002) Italy (1999)

Note: 1 d = deported; exp = expulsion; irr = irregular; stu = student; m = marriage.

(3,309) in the region and in Arzignano itself, there are 875 Ghanaians employed in the leather industry, the second largest group after migrants from the former Yugoslavia (1,197) (Veneto Lavoro 2001).18 Immigration in Arzignano has increased notably over a ten-year period. In 1991, foreign nationals represented only 2.7 percent of the town’s population. By 2002, foreign nationals represented 12.5 percent of the town’s 24,000 residents (Unpublished data, Comune di Arzignano). This growth confirms the need for migrant labor in Italy’s formal economy and also partially explains Italy’s attractiveness as a destination country for new migrants. Table 11.7 shows the mobility patterns of selected interviewees. These patterns indicate that in a context of restricted labor migration to Europe, a straightforward A to B model of international migration does not capture the reality of contemporary movements from Africa to Europe. Migration is more usefully conceived of as a staged process, although circumstantial events may induce modifications to short- or long-term plans and, for instance, lead to settlement in a country which was originally envisaged as a transitional location. There is, of course, an implicit bias in my sample, as I am taking Italy as the country of settlement. However, Italy should be viewed as a “current” settlement country rather than an absolute settlement country, as several interviewees articulated the desire to engage in further movement to other European countries.19 This is not to argue that none of the Ghanaians in Italy are settled. Rather, 18

19

Although I do not have a breakdown by nationality, there is evidence of mobility by labor migrants from the center and south of the country to Arzignano itself, with some 664 migrants moving from Lazio and 692 from the South (Veneto Lavoro 2001) Since the interviews in 2003, one interviewee has migrated from Italy to the UK.

Migration mobility in European diasporic space

285

Table 11.8 Asylum applications from Ghanaian nationals (1980–99)1 Year

UK

Germany

Netherlands

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999

29 13 407 689 337 141 196 125 172 330 1,330 2,405 1,600 1,785 2,035 1,915 675 350 225 1,995

2,768 3,378 4,113 1,611 2,670 3,994 5,769 783 1,304 3,178 3,786 4,541 6,994 1,973 300 781 676 698 308 227

– – – – 23 – – 2,515 920 812 715 465 140 73 70 n.d n.d n.d n.d n.d

Italy – ∗ ∗

31 234 67 27 18 6 10 13 12 ∗

– 11 25 – ∗ ∗

–

Source: UNCHR (2001) Asylum Applications in Industrialized Countries: 1980–99. Note: 1 ∗ = figures below 5; – = value is zero, rounded to zero, not available or not applicable.

the different rates of growth and the reduction in numbers of some ethnic groups suggests that Italy is an attractive country of settlement for some groups and less attractive for others. The immigration regime existing in a particular country affects the strategy utilized by migrants to enter specific countries (Castles 2004).20 For example, while the political regime in Ghana in the 1980s generated a number of asylum seekers, asylum claims by Ghanaians have generally been viewed as excessive in relation to actual political conditions (Peil 1995). Table 11.8 shows asylum applications from Ghanaian nationals to Germany, the United Kingdom, and Italy over a period of two decades. 20

This switching of strategies has been identified in other migrations. One of Morokvasic’s (2003) Polish interviewees applied for political asylum in Canada in 1987 as this was seen as the best way to obtain a secure status. Subsequent to his return to Poland after his application had been rejected, he migrated to Germany where he was able to work as a dental technician.

286

Immigration and the Transformation of Europe

As these figures indicate, applications to Germany have been comparatively high, peaking at almost 7,000 in 1992. In Italy, applications are low, in part because until 1990, refugee status was generally restricted to applicants from eastern Europe.21 Ghanaian asylum applications were not necessarily successful, however, and of the Ghanaians who sought asylum in Britain in 1991, for example, 99 percent were denied entry (Peil 1995).22 On the other hand, the majority of Ghanaians admitted to Canada throughout the 1980s were admitted as political refugees (Owusu 2000). Notwithstanding, the range of countries to which Ghanaians were attempting to migrate is evident when one considers the countries from which they were deported – no less than fifty-eight countries in 1993 (van Hear 1998).23 The political situation in Ghana in the 1980s had led some Ghanaian nationals to seek political asylum abroad, while the economic situation led others to embark on labor migration. Asylum seekers and labor migrants cannot always be neatly differentiated, however. In order to meet their objectives – personal safety or access to labor markets – refugees sometimes have to engage in labor migration while labor migrants conversely have to seek asylum. These choices are determined by the immigration or asylum rules of individual countries. Thus, it is more useful to consider the strategies Ghanaians have employed to migrate to Europe as symptomatic of changing immigration regimes and labor market opportunities in Europe which transform and shift migrants’ opportunities for settlement. By the beginning of the twenty-first century, migration to Europe was becoming increasingly expensive and physically risky for Ghanaians. Interviewees who were unable to meet the high costs of securing visas for specific European countries had travelled initially to Libya to accumulate enough money for the second leg of their journey to Europe. This usually involved embarking on a precarious boat journey from Libya to Italy. I present here five brief case studies which highlight the mobility patterns of individual labor migrants and which indicate how Italy was selected as a destination country. The interviewees include Jerome, a forty-five-year-old who migrated to Italy in 1988; Francis, a thirty-eightyear-old who migrated to Italy in 1995; Colin, a thirty-year-old who migrated to Italy in 1998, Robert, a thirty-four-year-old who migrated 21

22 23

As P´erouse de Montclos (2003) has shown in relation to the Somali refugee diaspora, their destination choice was influenced by individual countries’ asylum seeking position. Italy, for example, rejected at least 90 percent of requests, while the rate was 50 percent for The Netherlands. In 1993, 2,194 Ghanaians were deported back to Ghana (Peil 1995). 885 were deported from Germany, 312 from the United Kingdom, 184 from The Netherlands, and 142 from Italy (van Hear 1998).

Migration mobility in European diasporic space

287

in 1993, and Kenny, a twenty-six year-old who migrated in 2000. These narratives are useful for comparative purposes as interviewees’ different age, class, educational backgrounds, and time of arrival in Italy shed light on the shifting migration and mobility opportunities available to new migrants. Case 1 Colin was a middle class young man, who had been educated in a West African country where his father had been working as a teacher. He had started university and was taking courses in literature, philosophy, and politics. In explaining his personal motives for migration, he stated: “There was no future in Africa for anybody because even after studying you will finish and you have to go and teach . . . and the money is getting smaller. The dollar is getting bigger and our money is getting smaller . . . Everybody was going [to the West], almost anybody who had the chance was taking it.” His preferred destination choice was Japan because of its sophisticated technology and the only reason he did not get there was because he became nervous about getting onto a flight to Japan in Bangkok: “They didn’t actually stop me. I couldn’t do it. I was scared and I didn’t want to go to prison because I didn’t want any police problems. I called my mum and sister and they said that I should come back home.” Colin had been travelling on a false passport and had paid $4,000 (US dollars) for the privilege. He returned to Ghana for a year and once he had enough funds ($US 3,000) he received approval to go to the United Kingdom which he was pleased about. On arrival there, he claimed asylum and was eventually settled in a house in London. During his interview, he indicated some regret at not having absconded straight away: “If I wanted to try my way out, I should have done something quickly after that. After their initial interviews some people never speak to another official again, but lose themselves or they get married. My family in Ghana was a little bit political, which was why my father was living in [a West African country], but the British want actual proof, that you were beaten up, for example, and I couldn’t provide that proof.” In fact, after four or five months his application was rejected. He did not leave the United Kingdom, however, but found alternative accommodation and continued to work in a car accident repair shop. He was eventually tracked down by the police and deported back to Ghana. In Ghana, he drew on the money accumulated in London to pay the required $3,000 (US dollars) for his subsequent migration. This time his objective was Germany, because: “In Germany there are many places to hide and you can make the same money everybody is making all around you. But then just about the time I was going to leave [1998], Germany became too hard.” Rotte’s (2000) research on Germany confirms this. Asylum-seeking had become an important political preoccupation in Germany with applications

288

Immigration and the Transformation of Europe

rising rapidly (193,000 in 1990, 256,000 in 1991, and 440,000 in 1992). Asylum reform was introduced in 1993 and the safe country of origin rule did not permit admission to Germany if the country concerned was considered to be respecting human rights. Ghana was considered a safe country, and by the late 1990s, this knowledge about the transformed immigration regime in Germany had clearly trickled back to potential migrants in Ghana. Colin concluded that it would be too problematic to get to Germany and he received information that in Italy “they were doing papers for anybody who had been staying there for a few years, so if we moved to Italy, we might just be lucky and get some papers.” Using a scam, Colin travelled to Italy, staying initially with an uncle in Palermo. He obtained a job dish-washing in a fish restaurant and in order to fulfil the criteria for the 1998 immigration amnesty, he paid 600,000 Italian lire (300 Euro) so that an Italian lady would declare that he was employed by her to do domestic work. At the time of the interview, Colin had been working for several years in the formal economy in the leather industry. However, he deeply regretted that he had been unable to “make it” in London. Case 2 Jerome, by contrast, had limited formal education, having only completed primary school. He originally migrated to Nigeria where he worked for four years in road construction. He returned to Ghana for one year and then travelled again to Nigeria in 1987, with the aim of buying a ticket and visa to Europe: “When I was in Nigeria, I saw that the money was not enough for me . . . so I thought it was better for me to go to Europe so that I can take care of my children. Because I only have a little education, so if I didn’t travel I wouldn’t get enough money to look after my children.” In Nigeria, he encountered a businessman who travelled regularly to Italy buying Italian shoes to sell in Nigeria. This man became his contact to Italy and helped him to arrange his ticket and visa. Jerome had only been interested in migrating to Europe and was not especially concerned about which country. He was advised that Germany might be difficult because of the language and he knew that it had become difficult to travel to London, in part because of restrictions in the United Kingdom, but also because the air tickets to London from Ghana had become extremely expensive. Eventually he travelled to Rome in 1988 on a tourist visa. He headed straight for Naples in southern Italy where, according to his contact in Nigeria, he would encounter other Ghanaians. He was advised that when he reached Naples, he could ask any Anglophone black person for help. On arrival in Naples, he met a Ghanaian who was unable to help him, but directed him to the area where Ghanaians were living. He was helped by his compatriots to find accommodation and employment,

Migration mobility in European diasporic space

289

and having legalized his stay following the 1990 immigration amnesty, he moved to the north of Italy and began working in the leather industry. Despite having lived in Italy for more than fifteen years, he had recently entered the American immigration lottery, in the hope of migrating to the United States. Case 3 Francis, a thirty-eight-year-old, had been educated to middle school level and had trained as a mechanical technician. He was a professional footballer, playing in a second division team in Ghana. He had not planned to leave, but had been politically active with some friends over democratization issues and had been targeted by the authorities. He left Ghana in 1989, spent a month-and-a-half in Nigeria and arrived in Frankfurt, Germany, in 1990. He claimed political asylum, but his claim was rejected and he was deported back to Ghana in 1994. He tried to apply for a new visa to Germany as he had the support of the football team he had been playing with in Germany, but this was turned down by the German embassy in Ghana. He felt the political system in Ghana had not changed so he decided to leave again and to travel to Italy where his brother lived. Using a friend’s passport which had a visa for Germany, he travelled to Frankfurt. His scam was exposed but eventually he utilized another scam to travel to Italy. He would subsequently regularize his position via the 1998 immigration amnesty. He had worked in Italy in the leather industry, but at the time of the interview he was living in The Netherlands as he had married a Dutch-Ghanaian and was establishing his own business there. Case 4 Twenty-six-year-old Kenny arrived in Italy in 2000. He had finished his secondary education and planned to study microbiology in the United States but had been refused a visa in Ghana. He taught science and mathematics in a secondary school for a while but still planned to leave Ghana. His first choice in Europe would have been the UK, but he already had refusal stamps in his passport so he knew this would be impossible. He also used a scam to enter Europe, but as it was organized with the help of a friend so he did not have to pay much money. He first went to Holland and stayed there for a few months with a Ghanaian friend. He realized it would be very difficult to become legal in Holland, and also became aware that frequent security checks were implemented in the workplace. Thus, he concluded that it would be difficult to work there in the long term. He did, however, find some temporary work, working in a slaughterhouse. He was not that keen to go to Italy, as he had heard it was a difficult place and life was not so easy. However, as he could envisage problems if he remained in The Netherlands as an undocumented migrant, and he still had some time left on his visa, he decided to move to Italy. Some Ghanaian friends he knew there had informed him that

290

Immigration and the Transformation of Europe

there would soon (December 2000) be an amnesty in Italy. In fact, this interviewee complained bitterly about the misleading information being circulated amongst Ghanaians as the next amnesty was not implemented until 2002. In the meantime, he had worked as an undocumented worker, but at the time of the interview he had just heard that he had been successful with his amnesty application. He envisaged further mobility either to the UK or to the US. Case 5 Robert left Ghana in 1990 and flew to Germany where his father’s friend lived and where he claimed asylum. He stayed there for two-and-a-half years, but his application was rejected and he had to leave the country. He did not want to stay irregularly in Germany because, in his view, things would be difficult there without papers. He decided to try for Canada where a family member lived. He obtained some false documents and bought a ticket for Canada via Belgrade. He was denied entry in Belgrade and repatriated back to Germany. His relative in Canada advised him to try and leave from Vienna where he could get a direct flight to Canada. At Vienna airport he was arrested for traveling on a false passport and was detained there for nine months while his story was verified. As he said: “I knew everything was false but I had no alternative so I had to wait there.” Following his release, he contacted a Nigerian man he had met in prison and with whom he stayed for a few months. However, he could not see a future for himself in Vienna as he encountered many migrants engaged in marginal or criminal activities. He had the option of trying to go to the UK, as he was offered the use of a British passport, but he decided to go to Italy where he had a relative. He travelled by train via Switzerland and thought about claiming asylum in Zurich, but he met some Ghanaians who told him it was difficult to obtain refugee status in Switzerland so he would be better off trying somewhere else. He was able to successfully cross the border to Italy, arriving in 1993. He would eventually regularize his situation following the 1995 immigration amnesty. Since this time, he has worked in Italy’s formal economy and has no plans for further immigration. These complex “mobility for settlement” patterns across European borders demonstrate that contemporary labor migrants may aspire to specific countries as a desirable destination choice, but that these aspirations frequently have to be transformed in the face of contingent and unforeseen difficulties. These might include a change in immigration or asylum law, or deportation from a specific country. Interviewees who travelled in the early 1990s were more likely to have attempted the asylum route, while those travelling at the beginning of the twenty-first century were selecting different strategies for entry, often involving scams or risky boat journeys across the Mediterranean sea. While some interviewees

Migration mobility in European diasporic space

291

articulated a specific destination choice, such as Japan or the United States, most interviewees who had aspired to migrate to Europe did not always feel they could afford the luxury of specifying a particular country. Rather, the European destination choice was determined by a combination of the ease of entry to a particular country, the route that a particular migration scam might take them on, or the country where they had some contact. Nonetheless, the presence of Ghanaian communities dispersed among several European countries did play a role in terms of general assistance and information sharing (albeit occasionally flawed). In the early stages of the “delayed diasporization” of Ghanaiain migrants, labor migrants such as Jerome travelled to Italy knowing absolutely no-one, but safe in the knowledge that he would encounter other Ghanaians who would help him. Moreover, knowledge of diaspora communities in specific countries shaped perceptions in the sending country about the most desirable destinations. As the European component of the Ghanaian diaspora began to grow throughout the 1980s and 1990s, more information circulated about opportunities and difficulties in individual European countries. The very fact of the diaspora’s growth meant that potential migrants were also more likely to have contacts in a wider range of countries. It is to be noted that the presence of relatives or friends was frequently not sufficient to facilitate entry and settlement. Although migrants were able to stay with friends and relatives temporarily in a range of European countries, the problems associated with an undocumented status in Europe were significant enough to encourage further movement to a country like Italy, where it was not only possible to find work, but where it was seen as easier to live as an irregular migrant.24 Robert, for example, was not able to reach Canada and settled in Italy. Kenny felt unable to stay irregularly in The Netherlands and he too settled in Italy. In the contemporary European immigration climate, interviewees had generally been prepared to “settle for” any European country, despite having a desirable destination in mind. As a consequence, their “settling in” was a somewhat more precarious process and there were specific factors within Italy which caused them to consider further mobility. The majority of interviewees were incorporated into the formal labor market of Italy’s leather industry. While there are many stages to the leather production process, many interviewees were employed in the more physical and unpleasant aspects of tanning – cleaning and stretching the 24

Research conducted in Milan showed that migrants entered Italy because they envisaged it as a country where living and working was easy even without a residence permit as both inspections and deportations were limited (Reyneri 1998).

292

Immigration and the Transformation of Europe

leather. Tanning was the dominant occupational niche in the town and there were limited opportunities for work in other sectors. Despite the transition to legal status, interviewees continued to encounter discrimination in the labor market, limited social integration and partial citizenship.25 This social climate and working conditions contributed to a reluctance to accept that Italy was the final destination country. Family considerations were an additional impediment to permanent settlement in Italy. Men cited the limited employment possibilities for women outside of the leather and domestic work sectors, or the fact that they did not value an Italian education for their children as reasons for delaying family reunification.26 One interviewee, who had lived in Italy for eight years, had just brought his wife and two children to Italy. However, as his long-term goal was to settle in the UK, he was investigating the feasibility of his wife and children living there, while he continued to work in Italy. He needed two additional years of residence before being eligible to apply for Italian citizenship. Once he had acquired Italian citizenship, his plan was to move to the UK. Nonetheless, Italy also presented its own specific advantages – namely that Ghanaians could regularize their situation in Italy and generally count on stable work within the formal economy in the industrial districts of the North-East. However, the recent economic crisis which has affected some productive areas of the North-East may, for the first time, lead to unemployment among migrants in the area (Anastasia, Bragato, and Rasera 2004). For example, in 2003, leather production in the industrial district fell by approximately 20 percent (CGIL Vicenza 2004). If working conditions become more precarious as a result of the industry’s economic downturn, this may well impact on the latent mobility aspirations of Ghanaian migrants. Conclusion The Ghanaian case is suggestive of broader changes in patterns of migration to Europe. The migration of Ghanaians to and between European countries demonstrates that the immigration and asylum regimes of individual European countries are inter-linked. Decisions adopted in one country have a knock-on effect on migrants’ destination and mobility decisions in relation to other countries – creating “contamination” of 25

26

Interviewees complained of racism particularly with regard to housing. Private accommodation offices in the town confirmed that many landlords explicitly stated that they would not rent to foreigners. A high number of interviewees commented on the fact that Italian was not an international language.

Migration mobility in European diasporic space

293

national cases that is hard to capture in broad statistical studies. European immigration regimes encouraged interviewees to seek entry to Europe through a combination of overstaying visas, the use of scams, or travel on dangerous boat trips. These are modes of entry which can lead to detention, deportation, and death. Ghanaians undertook such financially and physically costly methods because they were confident that they would find employment in European labor markets. Undocumented Ghanaians were able to find employment in The Netherlands, the UK, and Italy – a clear indication of the availability of unskilled work for labor migrants in several European countries.27 Although interviewees were deported or suffered temporary setbacks, this did not deter them from trying again.28 Interviewees were not in a position to actively choose a specific European country of destination. Shifting opportunities directed them to different destinations within the known repertoire of European countries where Ghanaians have settled over the past twenty years. Nonetheless, the general social and political immigration environment in different European countries did impact on their propensity for settlement or further onward movement. Interviewees assessed whether it could be more beneficial to be undocumented in the UK or documented in Italy for example. The UK, Germany, The Netherlands, and Italy have become dominant poles of the Ghanaian diaspora in Europe, and information in relation to these countries is exchanged within diasporic networks, contributing to sequential mobility.29 The presence of a concentration of Ghanaians in a specific country is not in itself sufficient to lead to settlement. Interviewees sometimes had siblings or even parents in more desirable destination countries and yet were unable to access these countries because of other factors, such as the difficulty in acquiring legal status. Although Ghanaians were able to breach “Fortress Europe,” regulating their subsequent conditions of settlement, particularly in relation to employment opportunities and legal status was a paramount concern. Italy assumed a particular function in this regard. The overwhelming majority of interviewees had entered Europe either irregularly or had become undocumented through overstaying visas. However, they were all 27

28 29

While European countries are prepared to compete for skilled migrants (Vandamme 2000; Castles 2004), at present there is a reluctance to do so in relation to migrants prepared to work in unskilled sectors. Ghanaian responses to deportation have also been noted by Brydon (1985) and Peil (1995). As my previous research has shown in relation to Cape Verdean female migration, the existence of a transnational community proved especially useful for filtering information about conditions and opportunities in other countries, contributing to movement from Rotterdam to Rome (Andall 1999).

294

Immigration and the Transformation of Europe

eventually able to regularizes their status in Italy. Immigration amnesties were useful and exploited by Ghanaians.30 Amnesties in Italy are increasingly tied to the demonstration of a work contract, so the demand for migrant labor in a variety of economic sectors is an important dimension of Italy’s attractiveness to new migrants. This is likely to continue a trend whereby Ghanaians spend time in the more diffuse informal economy of the south of Italy before obtaining their papers and heading for the industrial districts of the north. European Union citizens’ current low rates of mobility are attributable in part to a lack of information about living and working conditions in individual European countries. “Mobility enhancing programmes” such as Socrates and Erasmus have been implemented to remedy this (Vandamme 2000: 451–3). Ghanaian migrants with access to different components of the diaspora in Europe are able to draw on information and “current knowledge” circulating within diasporic networks, as well as being able to have direct personal experience of other European countries through holiday visits to friends and relatives. In addition, if third-country nationals obtain mobility and employment rights within the EU, the distinction between destination aspiration and destination settlement may begin to fade. This will have implications for those countries that acknowledge their need for unskilled and skilled labor but have been negligent in their response to the wider social and political citizenship needs of labor migrants and their children. Ghanaian migration to Europe appears to be different to the circulatory migration referred to by Morokvasic (2004) and Castles (2004), which refers primarily to movement between the home country and a migration destination country. Rather, Ghanaian migrants were prepared to be geographically mobile across several European countries in a condition of “migration mobility.” More comparative research needs to be conducted in relation to other new diasporas in Europe to explore whether the patterns of mobility displayed by Ghanaian migrants are replicated in other migratory groups.  Ackah, W. 2000 “Diasporas of Faith: Exploring Ethnic and Religious Identity in the Ghanaian Seventh Day Adventist Community in Britain.” Paper presented to the Colloquium on New African Diasporas, SOAS, May 5. Adepoju, Aderanti 1995 “Emigration Dynamics in Sub-Saharan Africa.” International Migration 33(3–4): 315–90. 30

Research in relation to other migrant groups in Italy has highlighted the differentiated uptake of amnesty opportunities and confirms the importance of cooperation within groups to achieve a successful amnesty outcome (Reyneri 1998).

Migration mobility in European diasporic space

295

Ambrosini, M. 2003 “Per un inquadramento teorico del tema: il modello italiano di immigrazione e le funzioni delle reti etniche,” in Percorsi migratori tra reti etniche, istituzioni e mercato del lavoro, Michele La Rosa and Laura Zanfrini (eds). Milano: F. Angeli, 9–23. Anastasia, B. 1992 “Il settore del mobile in Veneto: Evoluzione storica e distribuzione geografica.” Oltre Il Ponte 40: 15–47. Anastasia, B. and G. Coro 1992 “I distretti industriali in Veneto: Una proposta di individuazione.” Oltre Il Ponte 38: 24–55. Anastasia, B., S. Bragato, and M. Rasera 2004 “Dopo la ‘grande regolarizzazione’ del 2002. Percorsi lavorativi degli immigrati e impatto sul mercato del lavoro,” in I sommersi e i sanati: le regolarizzazioni degli immigrati in Italia, Marzio Barbagli, Asher Columbo, and Giuseppe Sciortino (eds). Bologna: Il mulino, 103–38. Andall, Jacqueline 1999 “Cape Verdean Women on the Move: ‘Immigration Shopping’ in Italy and Europe.” Modern Italy 4(2) (November): 241–57. Andall, Jacqueline 2000 Gender, Migration and Domestic Service: The Politics of Black Women in Italy. Aldershot: Ashgate. Andall, Jacqueline 2005 “Cape Verdeans in Comparison to other Migrant Groups in Italy.” Paper presented to conference on Cape Verdean Migration and Diaspora, Lisboa, April 6–8. Angelillo, A. and F. Occari 1988 “Il distretto industriale dell’occhiale: struttura ed evoluzione.” Oltre Il Ponte 22: 87–121. Anwar, Muhammad 1979 The Myth of Return: Pakistanis in Britain. London: Heinemann. Basch, Linda G., Nina Glick Schiller and Cristina Szanton Blanc 1994 Nations Unbound: Transnational Projects, Postcolonial Predicaments, and Deterritorialized Nation-states. Basel: Gordon and Breach. Brydon, Lynne 1985 “Ghanaian Responses to the Nigerian Expulsions of 1983.” African Affairs 84 (337) (October): 561–85. Carfagna, Massimo and Franco Pittau 2003 “Italia: 20 anni di regolarizzazioni,” in Immigrazione dossier statistico, Caritas de Roma (ed). Rome: Anterem, 129–38. Caritas di Roma 1992 Immigrazione: dossier statistico. Rome: Sinnos editrice. Caritas di Roma 2002 Immigrazione: dossier statistico 2001. Rome: Anterem. Caritas di Roma 2003 Immigrazione: dossier statistico 2002. Rome: Anterem. Caritas di Roma 2004 Immigrazione: dossier statistico 2003. Rome: Anterem. Carling, Jorgen 2002 “Migration in the Age of Involuntary Immobility.” Journal of Ethnic and Migration Studies 28(1) (January): 5–42. Castles, Stephen 2004 “Why Migration Policies Fail.” Ethnic and Racial Studies 27(2) (March): 205–27. CGIL Vicenza 2004 “Le difficolt`a produttive e occupazionali a Vicenza.” unpublished manuscript. February 19. Cohen, Robin 1997 Global Diasporas: An Intoduction. London: UCL Press. Country Watch 2004 “Ghana 2004.” Houston, TX: Country Watch Inc. De Filippo, E. 1993 “I Ghanesi” in L’Arcipelago immigrazione: caratteristiche e modelli migratori dei lavoratori stranieri in Italia, Giovanni Mottura and Giovanna Altieri (eds). Rome: Ediesse, 277–87.

296

Immigration and the Transformation of Europe

De Filippo, E. 1994 “Le lavoratrici ‘giorno e notte’,” in Le mani invisibili: la vita e il lavoro delle donne immigrate, Giovanna Vicarelli (ed). Rome: Ediesse, 65–72. Ente Bilaterale Artigianato Veneto (EBAV) 2001 L’Azienda artigiana della concia. Arzignano: Grafiche Cora. Guolo, R. 2001 “Religione, Economia e Societa locale nel nordest.” Economia e societa regionale 3: 5–17. Makinwa-Adebusoye, P. K. 1995 “Emigration dynamics in West Africa.” International Migration 33 (3–4): 435–67. Mollel, A. 2000 “New African Diasporas: Transnational Practices of Ghanaians.” Paper presented to the Colloquium on New African Diasporas, SOAS. May 5. Morokvasic, Mirjana 2003 “Transnational Mobility and Gender: A View from Post-Wall Europe,” in Crossing Borders and Shifting Boundaries, Volume 1, Gender on the Move, M. Morokvasic, Umut Erel and Kyoko Shinozaki (eds). Opladen: Leske & Budrich, 101–33. Morokvasic, Mirjana 2004 “‘Settled in Mobility’: Engendering Post-Wall Migration in Europe.” Feminist Review 77: 7–25. Niewsand, B. 2004 “Ghanaians in Germany – Transnational Fields and Social Status.” Accessed May 28, 2004. Available from www.eth.mpg.de/people/ nieswand/project.html. Nyonator, F., D. Dovlo and K. Sagoe 2004 “The Health of the Nation and the Brain Drain in the Health Sector.” Paper presented to conference on Migration and Development in Ghana, Accra, September. Owusu, T. Y. 2000 “The Role of Ghanaian Immigrant Association in Toronto, Canada.” International Migration Review 34(4) (Winter): 1155–81. Parrenas, ˜ Rhacel Salazar 2001 Servants of Globalization: Women, Migration and Domestic Work. Stanford, CA: Stanford University Press. Peach, Ceri 1996 Ethnicity in the 1991 Census. Vol 2. The Ethnic Minority Populations of Great Britain. London: HMSO. Peil, M. 1995 “Ghanaians Abroad.” African Affairs 94(376) (July): 345–67. P´erouse de Montclos, Marc-Antoine 2003 “A Refugee Diaspora: When the Somali Go West,” in New African Diasporas, Khalid Koser (ed). London: Routledge, 37–55. Portes, A., L. E. Guarnizo and P. Landolt 1999 “The Study of Transnationalism: Pitfalls and Promise of an Emergent Research Field.” Ethnic and Racial Studies 22(2) (March): 217–37. Pugliese, E. 1990 “Gli immigrati nel mercato del lavoro.” Polis IV (1): 71–93. Reyneri, Emilio 1998 “The Mass Legalization of Migrants in Italy: Permanent or Temporary Emergence from the Underground Economy?” South European Society & Politics 3(3) (Winter): 83–104. Reyneri, Emilio 2004 “Immigrants in a Segmented and often Undeclared Labour Market.” Journal of Modern Italian Studies 9(1) (Spring): 71–93. Riccio, B. 1999 “Senegalese Street-sellers, Racism and the Discourse on ‘Irregular Trade’ in Rimini.” Modern Italy 4(2): 225–39. Rotte, R. 2000 “Immigration Control in United Germany.” International Migration Review 34 (2) (Summer): 357–389.

Migration mobility in European diasporic space

297

Spano, ` A. and A. Zaccaria 2003 “Il mercato delle collaborazioni domestiche a Napoli: il caso delle ucraine e delle polacche,” in Percorsi migratori tra reti etniche, istituzioni e mercato del lavoro, Michele La Rosa and Laura Zanfrini (eds). Milano: F. Angeli, pp. 193–224. van Hear, Nicholas 1998 New Diasporas: The Mass Exodus, Dispersal and Regrouping of Migrant Communities. London: UCL Press Ltd. Vandamme, Francois 2000 “Labour Mobility within the European Union: Findings, Stakes and Prospects.” International Labour Review 139(4): 437–55. Veneto Lavoro 2001 “Lavoratori Extracomunitari in Veneto: Un guadro aggiornato.” Mimeo. Williams, Allan M. and Vladimir Bal´azˇ 2002 “Trans-Border Population Mobility at a European Crossroads: Slovakia in the Shadow of EU Accession.” Journal of Ethnic and Migration Studies 28(4) (October): 647–64. Yeboah, Y. F. 1986 “The Crisis of International Migration in an Integrating West Africa: A Case Study of Nigeria and Ghana.” Africa Development 11(4): 217–56.

12

The new migratory Europe: Towards a proactive immigration policy? Marco Martiniello

Introduction Since the 1980s, the migration issue has returned to the top of the political agenda in various regions of the world, as we see in the evident return to 1920s nativism now taking place in the United States. This politicization of migration and its consequences has led to an over-dramatization, and sometimes an over-mediatization, of the human migration issue. Both migratory flows and the presence of immigrant populations are largely perceived as a cause of insecurity and even as a real threat (Martiniello 2001). Two significant features mark public debate concerning international migratory movements. Firstly, the emphasis is on real or potential migratory flow and on political and police measures to be brought into force for supranational regulation, in practice to keep the flow at a minimum level. The refugee issue is also on the way to becoming the major preoccupation of politicians and, to a certain extent, of certain sections of public opinion who are particularly aware of the migratory situation. Secondly, a certain “social alarmism” often predominates in these discussions where an essentially negative view of migration is presented and even seen as legitimate. There are many Europeans who accept the “invasion of Europe” doctrine and are ready to fight it. Immigration is thus presented as a fearful plague that must be swiftly vanquished, before it is too late. In other words, current migratory flows are often supposed to have a harmful effect on international security insofar as they may compromise relations between immigrant supplying states and immigrant host states. This is only a short step from claiming, as is often done in the popular press and by politicians, that migratory flows hinder the emergence of a new world order. This alarmist approach to migration is then logically followed by attempts to control, limit, or prevent it from doing harm by means of increasingly international political and police cooperation. 298

Towards a proactive immigration policy?

299

Migration is also perceived as a cause of insecurity inside states. The presence of immigrant populations is often presented as a threat to “native” economic well-being. Immigrants and their offspring are often accused of taking jobs from nationals or of taking unfair or fraudulent advantage of rich countries’ social security systems. Immigrant populations are also presented as a threat to law and order, so that immigration is associated with the rise in trans-frontier organized crime (drugs, prostitution, arms-dealing, and human-trafficking, mafias, etc.). Immigrants, particularly of the second generation, are associated with the rise in urban criminality affecting many towns and suburban areas of Europe. Consequently, since the presence of migrants is presumed to encourage feelings of insecurity in the native population, it is sometimes used simplistically as the main reason for the rise of extreme-right parties. In this way, immigration finally appears as a threat to democracy and immigrants as “internal enemies” who put in jeopardy our social benefits, our relative economic well-being, and even our cultural and national identity. Islam and Muslims thus become our bogey men. According to this view, Europe is suffering from galloping Islamization, threatening European cultures and values, including democracy and human rights. Stressing only the negative economic, democratic or security aspects of immigration is unsatisfactory from the academic point of view. Being too simplistic, it leads to manifest ambiguity and paradox. The growing economic exploitation, especially of illegal immigrants in certain sectors (for instance, in Spanish agriculture), coincides with declarations about the supposed costs for receiving countries. In cruder terms, the more we get out of immigrants the more they are accused of getting out of “us” (Adam et al. 2002). It is more realistic to recognize that migrations may have positive as well as negative effects both for the country of origin and the host country. There are no absolutes in the matter. It all depends on the general context of these migrations. We must also remember that migration is nothing new in human history. On the contrary, history is simply the long story of successive migratory flows across the planet. Humans have always moved over the face of the earth and nothing suggests they are likely to stop, whatever sort of restrictive migratory policies are brought into force. Modern means of communication and transport developments have, in fact, made it easier for people to travel to the four corners of the globe. Is it not a paradox, in the present state of the world, to make migration easier and cheaper for the many while trying at the same time to draw up more and more restrictive immigration policies?

300

Immigration and the Transformation of Europe

Migration is an inescapable feature of the start of this century, and is very likely to continue throughout it (Castles and Miller 2003). How could it be otherwise considering the number of existing and even increasing migratory pressures on a world-wide scale (demographic and economic inequalities, environmental and political causes, desire for better living conditions), and the technical possibilities for human mobility? In spite of restrictive immigration policies, frontiers between states remain more or less easy to cross. This is even more the case for the frontiers of democratic states which, in principle, are not willing to use force in order to deport human beings or to prevent their arrival or departure. This chapter shows in what way Europe can be considered as an immigration continent. It briefly follows migratory flows to Europe since 1945 and considers the outlook for the future. It also presents European immigration policies adopted since the postwar period, finally stressing the need to go further towards a common immigration approach in the European Union.

Immigration in Europe since 1945 Like other regions of the world, Europe has always been, from prehistoric times and all through the countless population movements up to the present day, a continent of immigration (Enzenberger 1994). During the prehistoric period, migrations took place all over the world, including Europe. The classical period was a time of great population movements known as the “great migrations.” Towns and cities like Athens and Rome were largely formed by drift from the land when peasants left for a more prosperous life in the urban centers. The Middle Ages saw the Germanic and Ottoman invasions as well as the Crusades. Learned men of the time traveled everywhere in spite of the difficulties to be faced. When Pope Alexander VI divided the New World between the Spanish and the Portuguese, he set off transoceanic migratory movements on a gigantic scale. Such movements were to mark the history of European populations’ right up to World War I. Millions of Europeans left their continent to take part in the conquest of the Americas. Between the wars, population shifts within Europe also gained considerable momentum for the following reasons: the depression, the closing of the United States to emigrants, and the appearance of dictatorships in certain European countries like Spain, Italy, and Germany. Belgium, at this time, was already recruiting foreign workers and accepting refugees from eastern and southern Europe (Caestecker 2000).

Towards a proactive immigration policy?

301

From 1945 to 1973–74 Just after World War II, demographic and economic indicators were announcing a new period of intercontinental emigration from Europe. The heady rise in fertility rates looked likely to lead to over-population in certain parts of the continent. In the economy, poverty was rife and restarting production was an enormous challenge. In such conditions, it was no wonder large numbers of Europeans went off to seek their fortune elsewhere, like in Canada or Australia. It was, however, mainly within Europe that migratory flows reached record levels in the period up to 1973–74 (Castles and Kosack 1985). During the industrial and economic reconstruction period of 1947–60, northern European economies suffered from a persistent manpower shortage in mining and heavy industry, whereas unemployment reigned in southern European countries such as Italy and Spain. These countries became immigrant labor pools for countries such as Belgium or Germany to draw on, and they recruited intensively. During the “Golden Sixties” (1961 to 1973–74), with industry constantly in search of fresh manpower, the demand for unskilled, cheap labor often outran supply. Immigration towards northern Europe continued, mainly on account of short-term economic requirements. At the time, the migration equation was generally perceived as relatively simple. When a manpower shortage appears, the immigration tap is turned on. When the vacancies have been filled, the tap is turned off again (Martens 1976). At a finer-grained level, the 1945 to 1973–74 period featured five migratory scenarios. In the first, northern European countries invited manual workers from southern Europe, then the Balkans, North Africa, and Turkey. This is the origin of the large numbers of Turks present in Germany and of the Italian and Moroccan communities in Belgium. In the second, colonial powers such as France and Great Britain encouraged immigration from their colonies and later from their ex-colonies. Algerian workers thus arrived in France (Sayad 1991); Caribbean or Indo-Pakistani workers settled in the British metropolis. In the third, the decolonization process led to the return of ex-colonials to the mothercountry. When Algeria became an independent state, numbers of excolonists came back to France. Some, called “Pieds Noirs”, met with serious problems of re-insertion. Similarly, the new dictatorships in exBritish East Africa turned out large numbers of resident migrants from the Indian sub-continent who, instead of returning to the mother-country, made for Great Britain. In the fourth, Europe was already facing the arrival of political asylum-seekers mainly from the communist bloc and later from Latin America where authoritarian regimes were in power.

302

Immigration and the Transformation of Europe

Finally, there was considerable movement of highly qualified elites thanks to industrial trans-nationalizations and the development of European and non-European international organizations. To different degrees and according to each European Union (EU) country’s specific timetable, most countries saw at least one of these five scenarios between 1945 and 1973–74 (Martiniello 2001). In Belgium, for example, the organization of southern Mediterranean immigration began very soon after the war, in 1946. With the independence of the Congo, Belgian colonists returned home. Belgium also accepted political refugees from eastern Europe and Latin America. As Brussels gradually became the European capital, a business and administrative elite moved there. In contrast, unlike France, Great Britain, and Holland, Belgium never encouraged immigrants from her colonies to come and work there. Two reasons are often quoted to explain the absence of colonial worker immigration in Belgium. Firstly, industrial exploitation of the Congo necessitated an abundant manpower supply which sometimes fell short locally. Secondly, on account of Belgian colonial racist attitudes, the government did not want to risk black workers settling in Belgium. At the European level, this period’s most significant migration scenarios were those of unskilled worker and colonial and post-colonial immigration. This led to considerable change in the ethnic and national composition of many European urban and industrial zones. The fact that Leicester is fast becoming the first non-white majority British town, that in Marseilles the northern area has a high proportion of French citizens of North African origin, that there is a significant Turkish presence in the Ruhr, that in Brussels, Charleroi, and Mons, an Italian accent is often heard, is largely due to migratory shifts that took place after the World War II and before 1973–74. Since 1973–74 The early 1970s marked the end of the “Golden Sixties” and the beginning of a global economic restructuring period. The first oil crisis, which struck in 1973–74, was of both economic and political importance, highlighting a fundamental transformation of the global economy and labor markets. There were several contributing elements: firstly, investment strategies underwent profound modifications; capital and jobs were increasingly being transferred to parts of the world where workers enjoyed less social security. Secondly, the micro-electronic revolution made for lower unskilled worker demand in certain industrial sectors. Thirdly, industrial decline took shape and gathered pace. Formerly prosperous mining and heavy industry areas faced a situation of almost unstoppable

Towards a proactive immigration policy?

303

job losses. Economic desertification (which actually began during the “Golden Sixties”) became rampant in Wallonia, Yorkshire, and the north of France. Fourthly, the service sector exploded. One of the main features of the workforce was flexibility, the essential element for operational efficiency. Fifthly, as from this period, we see the development of the “black” economy, with large numbers of illegal workers. Lastly, under-employment as a whole progressed, as well as job instability. All these developments took root during the following decades. They brought about a widening of the North–South divide and accentuated the poverty gap in the rich countries. They were also paralleled by global geopolitical transformations which took concrete form after 1989 with the fall of the Berlin Wall. These upheavals caused changes not only in the perception of European immigration but also in world-wide migratory scenarios. Until then, immigration had been seen largely as an essentially economic resource to be mobilized according to precise labor requirements. Generally speaking, the presence of immigrants was supposed to be temporary. In the case of unfavorable economic conditions, they were expected to go home. But in 1973–74, the expected departures did not take place, in spite of the recession and rising unemployment. Most European governments seem to have lost their bearings at this point. Several governments decided to end all further labor immigration. Yet despite its official termination, labor immigration into Europe continued. Certain previous migratory scenarios proceeded at a slower rate. Other older scenarios took a firmer hold. In addition, there were new immigration mechanisms appearing and developing both in Europe and on a global scale. Two distinct sub-phases can be seen in immigration within and into Europe after 1974 – the first up to 1989, and the second from the fall of the Berlin Wall up to the present day (Martiniello 2001). In the first sub-phase, European governments somehow never seemed able or inclined to see that their own decision to halt immigration was strictly respected. Foreign workers continued to obtain work permits in Belgium, France, Germany, and Holland after 1974. In Belgium, 30,000 new work permits were delivered to foreigners arriving directly from another country between 1974 and 1984 (Groenendijk and Hampsink 1994). Admittedly, these migrants did not have exactly the same profile as before. The proportion of highly-qualified workers tended to be higher. Also, family members continued to arrive. Not only did most migrant workers fail to leave the country after the 1970s oil crisis, but they increasingly brought in their family and gradually settled into less temporary situations in northern European countries. Refugees of more

304

Immigration and the Transformation of Europe

and more varied origins continued to arrive in relatively moderate numbers. In general, refugees from the communist bloc also continued to be welcomed – for reasons of international politics, among others. To some extent they were considered as desirable symbols of the failure of the communist egalitarian ideal. But the most striking development of this period concerned the Mediterranean countries. Historically speaking, Italy, Spain, and Greece had always been countries of emigration; now they became countries of immigration. Italy, for example, was historically the outstanding country of emigration. From 1876 to 1942 alone, over 18 million Italians left the country (Assante 1978). After the war and until the 1960s, the average annual number of emigrants stood at 300,000. It later fell to about 125,000. With industrialization in the north, however, the trend was gradually reversed. Fewer and fewer Italians emigrated and more and more foreigners entered Italy. Today the country has over 2.6 million legal immigrants, not counting “illegals” whose numbers are difficult to estimate with any accuracy (Caritas/Migrantes 2004). The second sub-phase came with the thawing of the cold war in the late 1980s, which brought a new sense of the potential for immigration. The nearer we got to the watershed of 1989, the more there was the feeling that political changes in the East might cause an unmanageable influx of migrants. There were alarmist forecasts. According to some sources, several million Russians were ready to leave as soon as they could. The fear of invasion gradually took root in Western Europe. In the summer of 1991, a boat carrying hundreds of Albanian emigrants, like bees in a hive, berthed in Italy. It was the media event of the summer, and aggravated the vague fear of a migratory flood. And yet we can but note that although immigrants continued to arrive in Europe after 1989, the feared invasion never took place. It is true that political crises in eastern Europe, Africa, or Asia have led to a global asylum crisis. There are indeed a growing number of refugees in the world (Joly, Kelly, and Nettleton 1997). In spite of the correct impression of a major rise in numbers of asylum seekers in the 1990s, Europe does not take in all the world’s refugees. Far from it. In fact, the vast majority of them take refuge in another poor country. Nonetheless, in the 1990s, European populations and their governments turned a colder and colder face to these immigrants. They were increasingly seen as “false refugees” or “disguised economic immigrants.” East–West migration after 1989 also included other new streams. Hundreds of thousands of ethnic Germans, who had lived for generations in the East, went back to re-unified Germany, the eternal mothercountry. Meanwhile, the German government tried to encourage Jewish

Towards a proactive immigration policy?

305

immigration into Germany. Lastly, migrant workers from the east came to do seasonal work in the west. The increasing politicization and mediazation of immigration was particularly striking in the 1990s and early 2000s, especially around the theme of illegal immigration and people-trafficking. Whether we consider the tragedy of the fifty or so young Chinese who perished in a crosschannel refrigerator truck going to England, the hundreds of Africans who disappeared while trying to cross the straits of Gibraltar in unseaworthy craft, or the hundreds of girls from eastern Europe forced into prostitution in western cities, these highly publicized dramas tend to produce a simplified impression of present migratory flows, which become connected with human tragedy, illegality and delinquency. This contributes strongly to the definition of immigration as essentially a security issue. However, different forms of non-legal immigration were partly the consequence of immigration and the hardening of asylum policy in Europe over the previous twenty years. The drastic restriction of legal immigration routes meant candidates were willing to try their luck by other means, either alone or with recourse to the services of illegal immigration professionals. It is also important to underscore the vague nature of the “illegal immigration” category. It covers, in fact, very different migratory trajectories. Some of these people enter the European territory legally, for example, with a tourist visa or as students, and then stay in the country after expiration of their visa. Others make a clandestine frontier-crossing and immediately become “illegal.” And some reside legally in Europe, but work illegally (Adam et al. 2002). This is not to deny the existence of illegal immigration and humantrafficking organizations. But neither can illegal immigration or immigration as a whole be reduced to the sordid activities of international organized crime. The many and complex causes of migratory flows in our globalized world are so compelling that, with or without illegal organizations, would-be immigrants will try their luck until they succeed or even die in the attempt. Complexification and multiplication of migratory scenarios It is sometimes claimed that economic globalization is the root cause of new migratory scenarios, making them fundamentally different from traditional immigration scenarios. In the globalized economy, it is said, it is becoming more and more difficult to identify and distinguish between emigrant countries and host countries (Martiniello 2001). Most countries may have somehow become both at once, as we see in those African states where large numbers of their population emigrate while immigrants

306

Immigration and the Transformation of Europe

from neighboring countries pour in. In Europe, Mediterranean countries, which have always known large-scale emigration have now become immigration countries. Moreover, it may be becoming more and more difficult to retrace the complex routes and trajectories followed by today’s migrants and thus give shape to multidirectional migratory flows. To simplify, it could be said that current migrations are increasingly less likely to feature a departure point A and an arrival point B with, in a certain number of cases, a final return of migrants to point A, or, for the majority, permanent residence at point B. In the age of globalization, migratory scenarios may actually involve points A, B, C, D, and E, between which some of the new migrants travel, making it impossible to identify their departure point (especially for those without identification papers) or their final point of arrival. The following anecdote illustrates this new migratory reality (as does Jacqueline Andall’s chapter in this volume). A young Senegalese man I met in autumn 1998 in the New York area of Battery Park, where he was selling T-shirts to tourists going to see the Statue of Liberty, explained in perfect French that he had worked for about a year in Italy, selling African statuettes in the streets. He also told me he was preparing to go for the third or fourth time to Belgium where he would stay a few months in order to buy second-hand cars and export them to the Ivory Coast. He thought he would stay a while in Belgium and then go back for a rest in Senegal before setting out again either for Europe or for the United States. Of course, there is no proof that this “global migrant” was telling the truth, but the migratory scenarios he describes, that of a young man ready to go where there are the best opportunities and stay as long as necessary, may at least seem more plausible and widespread today than was the case thirty years ago. It is, in any case, undeniable that migratory movements today correspond to more and more numerous, complex, global scenarios. Traditional migratory customs are disappearing before our eyes and new ones are appearing which will continue to develop. Thus, the patterns of postwar European migration ebb and flow across at least six major types of migration: (1) Unskilled labor migrations were crucial to industrial development after the war. Today, they remain important in certain economic sectors such as agriculture, building construction or catering. (2) Migration of highly qualified workers has steadily increased as technology has become more complex. This migration type is likely to spread in the future in Europe. (3) Armed conflicts and ecological disasters all over the world have caused, and will continue to cause, more or less forced population shifts towards Europe.

Towards a proactive immigration policy?

307

(4) Inasmuch as human beings aspire to family life, family regrouping is an important scenario. (5) Return migration has always existed, though of varying importance according to time and country. (6) Migratory movement is developing on a global scale, as shown above. This type of migration is not the most widespread, but it already affects our continent (Castles and Miller 2003). Serious observers no longer doubt that migration will be an inescapable feature of the twenty-first century. Migratory pressures continue to exist in the “global village” and are growing at the global level due to demographic and economic inequalities, environmental and political concerns, and the desire for better living conditions. Material possibilities for travel have multiplied and made for easier access. On the other hand, there is no prospect of anything that could be called a migrant invasion of Europe. The United Nations estimated total migrant numbers in the world in 1990 at 120 million (Castles 2002). Despite an increase in the 1990s, migrants still represent no more than 4 percent of world population. Moreover, only a small minority reaches Europe. The idea of a European invasion, therefore, in no way corresponds to current migratory reality. That said, European migratory prospects are hard to foresee with any accuracy. But it seems clear that the immigration policies brought into force may have an impact. Admitting that mobility has always existed and will always exist should not lead to an attitude of political fatalism or passivity. A clear, fair and proactive immigration policy is not only possible, but absolutely essential if migratory flows are to be to the benefit of all, here and elsewhere. Up to now, however, European policies have not always fully fulfilled these criteria. This takes another look across postwar European history to trace these policy choices. European immigration policies from 1945 to the present day Before starting to examine immigration policies some definitions are in order. Following the work of the Swedish political scientist, Thomas Hammar, a distinction is generally made between a state’s immigration policy and its immigrant policy (Hammar 1985). Immigration policy concerns migratory flow regulation, border-controls and admission of foreign nationals. It consists of all legal measures and procedures as well as administrative practices governing the selection, admission and entry of foreign nationals to state territory. Refusal of entry, removal, and expulsion measures are also included in immigration policy. Immigrant policy combines all legal measures and administrative practices governing the immigrant’s life in the new country. It covers work

308

Immigration and the Transformation of Europe

practices, social security, accommodation, education, social, and political participation as well as the immigrants’ cultural life. Immigrant policies plainly concern integration, incorporation, insertion or assimilation of newcomers to society. Immigration policy and immigrant policy are really two distinct, although linked, branches of what can be called migratory policy. They are distinct because the one aims to manage human movement while the other is concerned with aspects of migrants’ new living conditions. They are linked because, in certain countries (particularly the traditionally immigrant countries like Canada, the United States, or Australia), immigration policy is the initial phase of immigrant policy. The final objective of these countries is, in fact, to transform migrants into full citizens. These two branches of migratory policy are also frequently associated in political discourse. A much-used argument for a very strict immigration policy is that it represents an essential condition for ensuring the success of immigrant policies put in place. In Great Britain, for instance, politicians often attribute the success of race relations measures, which many would like to copy in continental Europe, to the strict frontier controls brought into force in 1962 (Favell 1998). Along the same lines, those who wish to follow a closed frontier policy today are ready to declare that higher immigration into Great Britain would harm the country’s ethnic and racial cohesion. British researchers, however, express open criticism of this simplistic linking of immigration to immigrant policy. This section is almost exclusively devoted to immigration policies in Europe. There are two other fundamental preliminary remarks to be made before delving into this sensitive issue. First, until recently, the EU states as well as outside states have had sovereign control over their immigration and immigrant policies. Each state is responsible for deciding, freely, but within international legal regulations, who may or may not enter the country, stay or reside and under what conditions. The state also has exclusive power of decision as to who may acquire citizenship. These issues are, in fact, at the heart of national sovereignty, which explains the difficulty and slow-going in drawing up a supranational European immigrant and immigration policy partly outside each state’s control. This point will be dealt with later in this chapter. Secondly, Western European and traditionally immigrant countries have widely different histories both in their approach to immigration and the type of immigration policy applied. The United States, Canada, and Australia consider themselves as nations of immigrants. Nationbuilding in these “New World” countries relies on the immigration myth. Of course, in the interest of historical objectivity, we can hardly accept the idea of virgin territory conquered by immigrants from all over the

Towards a proactive immigration policy?

309

world to form new nations. In the United States, genocidal treatment of Indians and slavery played an equally important part in the formation of the American nation. Immigration remains, however, a critical element in social imagery of the country’s history. The Ellis Island immigration museum in New York is one of the most popular in North America. There everyone can commemorate the immigration story, the startingpoint of United States history, as well as their pride in being American descendents of immigrants. In these circumstances, the idea of bringing in migrants to settle, exploit available resources, and contribute to development is historically rooted. They have admittedly often adopted restrictive immigration policies, but the idea of a proactive immigration policy is not seen as preposterous in immigrant nations. In Western Europe, on the other hand, the predominant notion is of age-old nations eventually achieving their own state (Smith 1986, 1995). What is the German nation? What is the French nation? There can be fierce debate on the subject, but rarely does anyone question the existence of these nations previous to immigration. In other words, European societies consider themselves as already populated and constituted entities, which needed the help of foreign labor at particular times in their economic development. They are much less likely to see immigration as a constituent part of the nation than are traditionally immigrant states. Immigration is considered as a temporary aid to national economic development and an essentially short-term factor, which does not necessarily lead to the granting of citizenship. Consequently, the idea of a reactive immigration policy adapts well to this perception of migratory events. According to the state of the economy, immigrants are invited with the idea that they can be got rid of when no longer needed. Yet European history shows that migration has also made a large contribution to the formation of European nations. France is a very good example. Though the French sense of a pre-existing nation is highly developed, it is also a long-standing immigrant state like the United States (Noiriel 1988). The difference between the two countries is that one of them has incorporated the migratory phenomenon into the romance of its national history, but not the other. To varying degrees, this would be true of all European countries. They have all experienced migratory flows, but this migratory history is more often ignored than integrated into national history. This is the case in Belgium. By reason of its geographical position, the country has always been criss-crossed by diverse populations. Yet the subject of immigration remains strikingly absent from Belgian history books. Even if immigration practically always leads part of the migrants in permanent residence, Europeans still broadly conceive of it as provisional and temporary. The French-Algerian sociologist, Abdelmalek

310

Immigration and the Transformation of Europe

Sayad, already pointed out, twenty years ago, that the distinction between labor migrations and population shifts correspond only approximately to reality (Sayad 1991, 1999). Human beings cannot be moved around like merchandise. Germany, for example, may well claim that it is not an immigrant country, but 2 to 3 million Turks are now a permanent part of its population. In the long term, many of them will become German citizens. Whatever the reality of permanent immigration, these differing perceptions of the importance of migratory facts for nation- and state-building are reflected in political attitudes to immigration and affect immigration policies put into practice on either side of the Atlantic. The trend in the western world towards strict frontier and immigration control – some would say the tendency to pretend we control the frontiers – has gathered force during the 1980s and 1990s. On the one hand, the United States has actually set up a wall that is several thousand kilometers long along the Mexican border, creating one of the most highly militarized frontiers in the democratic world. In Europe, very strict rules apply at checkpoints for entry into the EU. But in countries where immigration is traditional, there is still the possibility of legal immigration through various procedures (annual quotas, points system, lottery, family regrouping, etc.). Some regions such as Quebec even foresee increasing the number of visas to be issued to would-be immigrants in the next few years. In contrast, there is still no proactive immigration policy in most European countries. Legal immigration possibilities are at a minimum. The zero immigration doctrine is certainly running out of steam, and yet it has held out since 1973–74, when the European frontier was essentially closed to all new labor immigration. From the end of World War II to the first oil-crisis Between the end of World War II and the first oil-crisis, several types of immigration policy were set up in Europe. The reasons for these policies were chiefly economic. In a certain number of cases, they were also demographic and political. As Albert Martens writes, immigration was mainly seen as supplementary manpower to be mobilized according to the industrial system’s special requirements (Martens 1976). When labor demand outran supply in certain sectors, governments imported poorly-skilled migrant workers who were satisfied with low pay and accepted very hard workingconditions. In certain cases, recruitment took place in the country of origin, organized by the importing country’s government at the request of employers and with unwilling trade union consent. The unions feared

Towards a proactive immigration policy?

311

possible downward pressure on pay-packets due to the arrival of migrant workers. But rather than stand by passively during these manpower importations, they often decided to take part and try to limit the potentially harmful effects on local workers. The basic instruments of this immigration policy, which was no more than an appendage to employment policy, were bilateral agreements between countries. In June 1946, for example, Belgium signed an agreement with Italy providing for tens of thousands of Italian workers to be sent to the Belgian collieries. In exchange, Belgium would export coal to Italy on favorable terms (Martiniello and Rea 2001). As part of the so-called “contingent” system, the agreements provided for the workers to be brought to Belgium in special rail convoys. Belgium would also sign bilateral immigration contracts with other countries such as Spain, Morocco, Turkey, Yugoslavia, and Greece. Germany also had recourse to bilateral agreements to set up its system of Gastarbeiters (“guest workers”) through which foreigners were invited to come and work for a limited period (often five years). Switzerland invented a rotating immigration system in order to prevent foreigners settling in its territory. After their stint in Switzerland, immigrant workers had to go home and leave room for other foreign workers who would be subject to the same regime. In spite of these different immigrant worker recruitment and management methods, certain common features are present in the bilateral agreements signed by labor importing and exporting governments. Firstly, immigrant workers were given work and residence permits for a limited stay. Secondly, labor market access was initially restricted to certain industrial sectors, generally mining or iron and steel industries. Thirdly, the agreements concerned at first the recruitment of just the male worker – he could not bring his family. But from the early fifties, things began to change in certain countries. Fourthly, priority was given to strong, healthy young men who could be immediately productive. Whatever the immigration system chosen, the legal distinction between citizen status and foreign status was the central criterion for the migrant’s right to work, residence and social security. The immigrant was a foreign worker whose presence was seen as temporary. His only social existence was as a worker. In these circumstances, unemployment was an intolerable anomaly in the migrant’s trajectory. Either he worked or he had no place in the host country. In the 1960s, labor demand reached such a high level that certain governments adopted a laissez-faire policy. Foreigners, some of whom had already migrated under the bilateral agreements, arrived in Belgium or Germany simply on a tourist visa (Martens 1976). They found jobs in industry, without any trouble. Later they could

312

Immigration and the Transformation of Europe

regularize their situation as to residence and working rights. The granting of tourist visas thus plainly appeared as a part of reactive immigration policy. Certain colonial powers set up other immigrant labor recruitment methods. Prior to decolonization, colonial labor migration to the home country was in fact assimilated to internal migration. France and Great Britain had, in most cases, no need for collective labor recruitment in their colonies. On account of poverty and the demographic explosion, many colonial subjects were attracted by the opportunities they thought they would find in the metropolis. Leaving was easy thanks to their citizen or subject status, which gave them entry into France or Great Britain. In theory, colonial immigrants enjoyed the same rights as other citizens. In practice, however, they faced hostility from the local population and various types of discrimination in their social and working life. They essentially played the same economic role as immigrant workers in countries like Belgium or Germany. But in terms of status, they had the immediate advantage of a secure residence permit. Decolonization put an end to freedom of movement between colonies and home country. Both France and Great Britain would endeavor to control immigration from former colonies, which is the object of the British 1961 Immigration Act (LaytonHenry 1992). Immigration policies are also often put in place for internal demographic and political reasons. In the 1960s, Walloon demographic decline was seen as a danger for the region’s political equilibrium with Flanders. The French demographer, Alfred Sauvy, recommended the immigration of young families to compensate for the demographic shortfall. Without the Belgian population being really aware of it, this concern was partly responsible for the Belgian authorities’ comprehensive attitude to family regrouping, which had, in fact, already taken place since the 1950s. In short, it is obvious that European immigration policies from 1945 to 1973–74 were largely reactive. According to particular economic requirements and sometimes the demographic situation, governments called on labor immigration that they generally considered as temporary. In most cases, permanent residence, and the formation of new ethnic minorities resulting from these migratory processes were neither anticipated, desired, nor organized. Generally speaking, European migratory policy during this period was no more than an element in employment policy and a reaction by governments to market forces and the sometimes contradictory, sometimes converging, interests of migrants’ countries of origin and host countries. It was also very nationally oriented. Each state drew up its immigration policies on its own, according to its sovereign rights, without consulting European Community partner states

Towards a proactive immigration policy?

313

or taking account of special links with each of the European or other countries. Unfinished Europeanization (from 1973–74 to the year 2000) With the development of the European integration process since the 1957 Rome Treaty, immigration policy was to become an issue for discussion, and eventually decision at the supranational level. Insofar as European integration is based on four fundamental freedoms – free movement of capital, goods, and services as well as of persons – migratory policy was bound to appear sooner or later on the European political scene. Chapter 16 in this volume discusses these developments in greater depth, but a rapid summary of them is a crucial part of my overview. The baseline of the story is that the realization of the four freedoms has been extremely hesitant and complex. On the one hand, many European states have been very reluctant to give up their sovereignty on migration issues, especially as they have become more politicized. On the other hand, within fifty years, we have progressed from essentially national immigration policies to European-scale intergovernmental cooperation, with a growing tendency to common discussion and, to a lesser extent, a common immigration policy. This process has been hampered by running disagreements over the appropriate institutional format for cooperation, notably with several states objecting to the federal-style centralization of migration issues in truly common policies. Also crucial is the fact that this slow, laborious process of migratory policy Europeanization has taken place in a spirit of respect for the zero immigration doctrine. An important result is that the Europeanization of immigration policies – whether tending towards relative harmonization, convergence or more closely concerted national policies – has led to a considerable heightening of the restrictive nature of immigration policy and border controls at Europe’s external frontiers. In becoming more European, immigration policies have become even more security-based. Most of the developments in the Europeanization of immigration policy took place after 1973, occasionally taking giant steps. Two important periods stand out: the first from 1973–74 to 1989, the second from 1989 to the present day (with an important sub-phase of more accelerated activity since 1996). During the first period, migration was discussed at the European level mainly as a by-product of discussions about the construction of a common European market. During the second period, migration became a European issue in its own right. Two earlier dates provide the background to the former period: 1957 and 1968. The Treaty of Rome of 1957 establishing the European

314

Immigration and the Transformation of Europe

Economic Community (EEC) provided a legal basis for a European intracommunity migration policy. The principle of free labor force movement, one of the founding principles of the Common Market, implied abolition of all discrimination based on Member-State workers’ nationality with regard to employment, pay or other working conditions. The EEC’s Council of Ministers was to be responsible for adopting Directives and Regulations in this respect. In 1968 the EEC took its first actions in this sphere, adopting legal norms concerning freedom of movement for European workers. EEC Regulation 1612/68 established a legal difference between European and non-European workers. From this date, a line was drawn between the issue of intra-European mobility and that of extra-Community immigration. In other words, the difference of status between those who become European citizens and those from outside countries is fixed by a European legal norm. In the same year, Directive 68/130 widened the scope of freedom of movement for workers to certain members of their family, of whatever nationality. The aim of this Directive was above all to facilitate European worker mobility with the Common Market. And yet, in some ways, it laid the foundations for the future development of an extra-European immigration policy (Martiniello and Govaere 1989). The first steps toward elaborating such a policy in the 1973–74 to 1989 period began with an EEC summit in Paris in 1974, where the question of “non-Community foreign nationals” status was seriously examined for the first time. This was the logical consequence of the recently-affirmed idea of a “Citizens’ Europe.” How was this to be managed without considering similar treatment of non-European foreign nationals? Abandoning intra-European border controls while leaving each state free to adopt its own immigration policy did not seem feasible in the long term. In 1976 the Council therefore decided in favor of intergovernmental cooperation on the matter – if only in principle. The growing importance of the immigration issue led the Commission to present, in March 1985, a document on “Orientations for Common Migration Policy.” This text pleaded in favor of more elaborate European personal mobility legislation. The Commission also defended the idea of intergovernmental consultation on Member States’ non-European immigration policies. Already at this time, the Commission was playing a leading role in sensitizing European authorities to the importance of the migratory issue and in cautiously trying to plead for a European immigration policy. A few months later, the Commission presented the idea of European coordination of entry, residence and work-permit regulations for non-European citizens. It also considered the creation of a common visa policy.

Towards a proactive immigration policy?

315

In June of the same year, the Commission took a further step in this direction by deciding to set up a procedure for previous notification and concerted action on migratory policies with regard to third countries (Martiniello and Govaere 1989). Its aim was to encourage Member States to inform their Community partners and the Commission of any changes planned in existing non-European immigration legislation. This decision made it possible for the Commission and the Member States to coordinate their action without any binding constraints. Member States were reluctant to accept this decision, however, which many saw as an attack on national sovereignty. The Council thus adopted a resolution again stating that non-European immigration policy remained a national responsibility. Even this was not enough for France, Germany, Denmark, The Netherlands, and Great Britain, who asked the European Court of Justice to annul the Commission’s decision since (they argued) the Commission was not competent to deal with the matter. They thus indirectly contested the very idea of a common, coordinated non-European immigration policy. After extensive debates in which the European Parliament participated, and often spoke in defense of the Commission, the Court rescinded the Commission’s decision, while upholding the idea that a European migration policy is feasible. The parties met halfway. The Commission re-attacked in 1988 with a second decision setting up a previous notification and coordination procedure for non-European immigration. This amended version of the initial text caused less trouble. The Member States seemed divided between the wish to set up an internal market and saving their freedom of action on migration issues. This ambiguity is clear in the Single European Act of 1987, which revised the EEC treaty and launched the “Single Market 1992” program. On the one hand, governments promised to cooperate over non-Community foreign nationals, rights of entry, movement, and residence. On the other, they reasserted their right to take national measures for controlling nonEuropean immigration. This ambiguity endured. In 1989, during finalization of the Single Market program, the Palma Document (prepared for the June 1989 European Council summit in Madrid) drew up a list of problems to be solved for achieving personal freedom of movement in the EEC. It included border control on internal and external frontiers as well as visas and expulsions. Priority was given to the drawing-up of a list of countries whose nationals would need a visa for entering any Community country. Another priority was common external border control and surveillance measures. The document also attacked the question of improved cooperation and information exchange between customs and national police, as well as

316

Immigration and the Transformation of Europe

prevention of illegal immigration. But if immigration policy harmonization was now presented as desirable, in-depth discussion of the content and development of a future common immigration policy did not take place. Meanwhile, in the middle of the 1980s, senior officials in the Benelux countries, France and Germany were working, in secret and outside European institutions, on an agreement for gradually eliminating border controls on these five countries’ common frontiers. The final objective was to completely abolish internal frontiers in this group of states and place all border controls at their external frontiers. The Schengen Agreement signed in June 1985 raised a number of questions not only among migrant defense associations but also among parliamentarians from these states, who were left out of the negotiations. There was a regrettable lack of information. The agreement admittedly provided for visa and entry policy harmonization, but it heavily emphasized police and judicial cooperation – clearly associating immigration with criminality and terrorism. The Schengen Agreement application clause was signed in 1990. It stresses interior or exterior frontier crossing, visas, mobility conditions, residence permits, and state responsibility for dealing with asylum applications. This agreement also provides for the setting-up of the Schengen Information System (SIS). This huge data bank, mainly concerned with wanted persons, is more a tool for improved police cooperation than a means of ensuring personal freedom of movement (Collinson 1993). Some see Schengen as a practical method of work. Rather than proceeding through the usual intergovernmental channels, or accepting the leading role of the Commission, is it not better for a small group to make some progress towards an immigration policy, even if it is securityoriented, and later try to rally other European states to the same idea? And in fact Schengen did gradually attract most EU members, and even some from outside the EU. Among the EU Member States, only Great Britain and Ireland have not yet signed up to the Schengen Agreement. Moreover, although they cannot formally join, Iceland and Norway adopted the entire content of Schengen in March 2001. The Schengen method may be efficient. However, the fact that the basic agreement was negotiated in secret, hidden from the public gaze and from its representatives, poses a major problem for European democracy. Thus, until 1989, it cannot be said that a common non-European migration policy existed. Each state tried to regulate migratory flows and check out foreigners on its own. Cooperation between states was difficult and took place within the negotiations for setting up the Internal Market by 1992 (or in parallel to them, in the Schengen format). The maximum attempted was to coordinate national border-control policies at the

Towards a proactive immigration policy?

317

European level. In the absence of a more coherent collective discussion of migration, these national immigration policies practically all became more restrictive and security-oriented while remaining as reactive as ever (Martiniello 2001). As mentioned above, the fall of the Berlin Wall and the following collapse of nearly all communist regimes aroused the fear of a massive, uncontrollable migratory influx from the east. Illegal migrant and refugee flows from the south were also a cause of anxiety. It was increasingly recognized that an urgent political response to these migratory “threats” must be found. In a troubled geopolitical environment, external bordercontrol became a priority for the EU, if only in European political rhetoric. Migration became a major issue in the European and even non-European political context. Intergovernmental and multilateral cooperation gained increasing momentum from the end of the 1980s. In the early 1980s, five multilateral forums were dealing with migratory issues. A few years later, there were fifteen, including the Schengen group, the Rhodes group, the Trevi group, the “ad hoc immigration group”, the “mutual assistance” group, the European anti-drug committee, the Berlin group, the “horizontal group,” the European Security and Cooperation Conference, and others (Collinson 1993). This intense intergovernmental and multilateral activity is a main feature of the new migration policy regime set up during this period. The states increasingly realized that it was in their interest to cooperate in finding solutions to migratory dilemmas. Migration is in itself transnational. It may therefore be no use trying to regulate it simply between countries of origin and host countries. The Maastricht Treaty setting up the European Union came into force in 1993 and confirmed this trend towards an intergovernmental and multilateral approach to immigration policy. But, although visa policy was included in the supranational “first pillar” of core “Community competence,” all other features of immigration policy were relegated to Title IV of the Treaty, concerning Justice and Home Affairs cooperation. This is the famous “third pillar” of the Maastricht Treaty, the creation of which sought to limit centralized, supranational policy harmonization, and which pointedly did not provide for a common migratory policy as a whole. A particularly significant area in the post-1989 expansion of activity was political asylum. Cooperation on asylum dates from 1986 with the adoption of the Single Act. At this time, one of the aims pursued by the “ad hoc immigration group” was to put an early end to abuse of the asylum process in Europe. With the enormous increase in asylum cases after 1989, however, pressure for development of a European asylum regime accelerated. Considerable progress has since been made, though

318

Immigration and the Transformation of Europe

Member States have continued to disagree over how much their cooperation should operate through hortatory resolutions and declarations or through more centralized and legally binding common agreements and actions. Even given extensive coordination on asylum, certain states remain anxious to retain a degree of national sovereignty. There has been no shortage of resolutions and declarations on asylum since 1986. The 1989 Palma Document presented several asylum policy proposals. It called for a common Geneva Convention-based policy involving Member State acceptance of identical international commitments, simplification of screening procedures for “plainly unfounded” asylum applications, determining the state in charge of asylum application screening and the creation of a data bank to record the date and place of asylum applications. In 1991, the Commission presented to Member States an asylum policy communication that prioritized the harmonization of expulsion policy. The Maastricht Treaty of 1991 defined asylum policy as a matter of common interest to be dealt with through intergovernmental cooperation. A year later, the Edinburgh European Council again stated Europe’s intention of keeping to the asylum tradition but also of countering the potentially harmful effects of uncontrolled immigration. Among the principles inspiring European action was the recommendation to try to maintain displaced persons in the secure regions nearest to their home country. The idea of temporary protection appeared in these discussions, but did not lead concretely to a clearly defined status. In terms of legally binding agreements and measures, two major instruments are directly pertinent to the asylum issue. The Dublin Convention 1990 was mainly aimed at determining the country responsible for asylum application screening. To avoid the “orbiting refugee” phenomenon, it lays down that one Member State alone shall be in charge of screening each asylum application. Asylum seekers thus have only one chance of obtaining refugee status. The state responsible is that of the applicant’s first entry into EU territory. The Convention also considers the question of information exchange on asylum rights between different states. As to the Convention on European external frontier crossing of 1990, it deals with the policy to be adopted with regard to asylum-seeker conveyers. In 1995, it was still at a standstill because of disagreement between Spain and the United Kingdom about the Straits of Gibraltar. The debate continued for some time within the Schengen framework. With the opening of the 1996 Intergovernmental Conference that led to the Amsterdam Treaty in 1997, the European policy regime for international migration underwent four main developments that marked a new, more ambitious phase. Firstly, the political agenda widened considerably in comparison with previous periods. The Treaty assigned the EU

Towards a proactive immigration policy?

319

five major tasks: admission policy harmonization, development of a common illegal immigration approach, drawing-up of a labor immigration policy, solving of problems raised by non-EU foreign nationals’ presence within EU territory, and action to stem outside migratory pressure. Secondly, the emphasis was laid on joint international action. The basic notion was to set up a political regime stretching beyond the Schengen frontiers, and even beyond those of the EU and of the European Free Trade Association, stressing the importance of dialogue with the fiftyfive-member Organisation for Security and Co-operation in Europe, the forty-one member Council of Europe, and the thirty-member Organisation for Economic Co-operation and Development. The eastern enlargement of the EU was already on everyone’s mind, leading to an emphasis on coordination with the accession countries as well. Thirdly, the prevailing state of mind remained largely defensive. The EU and its Member States wished, above all, to protect EU frontiers by putting border controls on its external frontiers and creating buffer zones. Finally, this new regime was marked by a new level of confusion. It became even more difficult to discover exactly who does what in European immigration policy. During the Intergovernmental Conference preparing the Amsterdam summit, it was obvious that opposition from Great Britain, Denmark, and Ireland would block any sort of immediate immigration policy revolution. The other Member States claim to support the idea of a common immigration policy, leading to a compromise of sorts that has been decried by some political observers on both sides. Of all the difficult texts in EU history, the Amsterdam Treaty is one of the most complex and least readable, with fourteen protocols and fortysix appended declarations. Its migration-related provisions were particularly Byzantine. It provided for the incorporation of Schengen principles in the “first pillar,” except with respect to the three above-mentioned states. The Community thus adopted the Schengen restrictive, securityoriented approach. The Maastricht Treaty’s Title IV on freedom of movement, immigration and asylum was moved to the first pillar. In other words, most of the migration and asylum issues left the third intergovernmental pillar and became subject to the more supranational procedures of core “community competences.” The Amsterdam Treaty also foresaw the possibility, within five years of its ratification, of creating a common immigration policy, which would have to be unanimously adopted by the Council. But discussion of the content of this supposed future policy was avoided. Nonetheless, migration issues remained high on the agenda. At the 1999 European Council summit in Tampere, the Member States stressed the need to reach a common immigration and asylum policy within the

320

Immigration and the Transformation of Europe

five-year time frame in the Amsterdam Treaty – by 2004. Border-control and anti-illegal immigration measures were again emphasized. Two years after the Amsterdam Treaty came into force in 1999, nobody denied the slow progress made on common immigration and asylum policy. The Commission played its part with no less than eleven proposals. Only two of them have so far been adopted by the Council. In a speech delivered in Rome in July 2000, Commissioner Vittorino, in charge of the dossier, wrote off zero immigration. In November 2000, the Commission presented a communication on Community immigration policy. In one way, things began to change. More and more political actors recognize the limits of zero immigration doctrine as well as the security-oriented, restrictive approach to immigration. Otherwise, things remain the same. Firstly, Member States are still suffering from a sort of schizophrenia. On the one hand, they declare they want to set up a common immigration policy. On the other, national bureaucracies cling to their national sovereignty. Secondly, the restrictive, security-oriented approach is widely accepted by most Member States. A generous, proactive European immigration policy is still to come. We shall now look at some arguments for and against such a policy. Conclusion: For a proactive immigration policy in the European Union In the last few years, demands have been voiced all over Europe for official re-opening of frontiers to certain forms of labor immigration. Several European governments have already seriously studied this hypothesis, often under strong pressure from industry, in order to counter the highly skilled manpower shortage in certain sectors, or unskilled in others. Others see immigration as a solution to our demographic ageing and decline as well as a means of safeguarding our social security system, particularly our pensions. In this view, we would import large numbers of young immigrants, who would have a lot of children and contribute, through taxation, to the costs of our welfare state. Lastly, there are those who advance our commitment to human rights in support of a more generous immigration and asylum policy. Nevertheless, zero immigration is still considered the reference. Official re-opening of frontiers to new immigration is generally regarded with suspicion, if not hostility. Trade unions, for example, are reluctant to consider it. As to governments, some ministerial departments are afraid this new immigration will prove to be at the expense of labor market integration for the immigrant-origin youth. Most demographers are opposed to the idea of using immigration as a remedy for population ageing.

Towards a proactive immigration policy?

321

After September 11 had stopped any attempt to seriously move beyond the zero immigration doctrine at the EU level, the European Commission has tried hard to stimulate a debate on the necessity to have a global approach to immigration, asylum, and integration. In June 2003, the Commission presented a Communication on immigration, integration, and employment (COM [2003]: 336) advocating a holistic approach targeting all dimensions of immigration and integration (economic, social, and political rights, cultural and religious diversity, citizenship and participation). Referring to a 2000 Communication (COM [2000]: 757), the text determined the targeted population as essentially composed by migrant workers, their reunited family members, refugees, and persons under international protection. In November 2004, the Common Basic Principles (CBP) on Integration were adopted under the Dutch presidency. They are aimed at designing a common European integration policy. They suggest a framework to serve as a reference for the implementation and evaluation of current and future integration policies. Finally, the Commission released a Green Paper on an EU approach to managing economic migration in early 2005 (COM [2004]: 82 final). The paper raised numerous issues and discusses various scenarios relating to an EU approach to economic migration. It has led to a consultation of all the stakeholders throughout the Union. At the time of writing, the position of the Member States still look difficult to reconcile on this matter. Analysis of immigration policy and declarations on the subject reveal, as the American researcher James Hollifield has been stressing for some years, a flagrant contradiction between market logic and political logic (Hollifield 1997; Entzinger et al. 2004). On the one hand, liberal states are obliged to support frontier opening as a part of their foreign policy in a free-trade economy, which presupposes freedom of movement for highly-skilled labor. On the other, they are afraid a migratory influx will bring into question the social security system, national identity and social cohesion, which is why they sometimes support very restrictive immigration and border-control policies. This tension, or, as Hollifield writes, this paradox of liberalism, is at the heart of public debate on immigration in Europe and one of the major causes of the slow progress towards a more generous proactive common immigration policy in the EU. Furthermore, anti-immigration and anti-immigrant sentiment is widespread among the population and the idea of re-opening frontiers is far from meeting with unanimous approval. These contradictory pressures sometimes cause divisions within governments. Some ministerial departments seem willing to discuss an end to zero immigration, while others seem mentally stuck in a timid, restrictive and security-oriented attitude to migratory flows.

322

Immigration and the Transformation of Europe

The realistic argument in favor of a more generous, proactive common immigration policy in the EU rests, on the one hand, on the findings of migration theories and analysis of the main immigration trends in the last fifty years and, on the other, on an assessment of the restrictive policies in force since 1973–74 in Europe. Several important points must be stressed. First, as stressed in the first two chapters of this volume (by a demographer and economists), migration is a complex phenomenon that cannot be reduced to its economic, demographic or political aspects. Economists often explain migratory flows by a juxtaposition of individual rational decisions taken by migrants on a cost–benefit basis. According to this, they leave their homes when they think they will get an economic advantage from migration. This is largely a simplification, as can also be said of certain demographic analyses explaining migration by the existence of demographic inequalities between different regions of the world. There would be emigration from overpopulated to less populated regions. Yet we observe that it is often heavily overpopulated urban zones that attract most migrants. Migration can only be explained within a transdisciplinary approach, alone capable of expressing its complexity. For the present, the most efficient theories are migratory system theories. In their attempt to explain migratory movements, they take into account both the role of the state and that of individual decision. In this way, they combine the “macro” and “micro” approaches. On the “macro” level, they recognize the importance of institutional and structural factors such as the state of the market as well as the legislation and policies adopted by different states. On the “micro” level, they take notice of the role played by migrants’ beliefs and practices as well as by social networks. Many studies highlight the importance of networks to explain the formation of same-origin migrant communities, for example, those from China or Italy, in certain European or North American towns. They also stress international comparisons as well as historical links existing between the migrant’s home country and the host country, for instance, their cultural proximity, political influences or remaining traces of colonization (Castles and Miller 2003; Verdusco 2004). Second, and as a consequence, migration can only be understood in a global context. It means little to explain immigration in Belgium, for example, while isolating this country from its European partners. Immigration can only be understood and explained if we also understand and explain emigration. This means taking note of what is happening all over the world, not imagining that immigration begins when foreigners cross the frontier. Third, North–South inequalities, whether economic, social, demographic or environmental are the context in which migratory movements

Towards a proactive immigration policy?

323

are formed rather than their fundamental causes. Even if such inequalities were to disappear overnight, migration would not stop. It would simply take place in another context and take on another form. Fourth, international migrations are, in reality, the expression of growing interdependence between different regions of the world in the globalized economy. Different parts of the world are all, at the same time and to various extents, both emigrant countries and immigrant countries. Fifth, migration is certainly a feature of the economic and labor market restructuring process in the wealthy, developed parts of the world. This is as true for legal migration as it is for the illegal kind. In this latter case, if hundreds of thousands of clandestine immigrants are doing agricultural work in southern Europe and California, it is because the demands of competition impose cost reductions, which, in turn, keep prices down for the consumer and increase profits for the producer. In other words, illegal immigration plays an important economic role. This is perhaps why it is flourishing despite all the policies designed to combat it and which hypocritically denounce it as a calamity. Sixth, we must not omit to mention the cultural aspect of emigration. In certain parts of the world, emigration is a compulsory phase of life for an individual or a family. It represents a sort of rite of passage or an obligatory service rendered to the community. In front of so complex a phenomenon, it will be easily understood that restrictive, security-inspired policies relying on the zero immigration doctrine are plainly inadequate. Policies applied in Europe since 1973–74 have definitely not achieved the objectives set. Moreover, these policies have produced perverse and apparently unexpected results. In the first place, as mentioned above, they have not put a stop to immigration into Europe. Declaring an end to labor immigration does not mean that it will actually stop. Furthermore, ambiguities can be found in working visa policy. On the one hand, a total ban on labor immigration is announced and on the other, visas are still delivered to certain categories of foreign workers. Secondly, while claiming to have closed the front door on immigration, that is, legal labor immigration, European governments have, in a way, encouraged entry through the side door, the one used by asylum seekers, and the back door, which lets in illegal and clandestine immigration. So immigrants who might in other circumstances have tried to enter Europe legally have chosen to seek asylum or enter by illegal means. Growing illegal immigration and people-trafficking can be largely explained by the extremely limited legal means of access to European territory. As to asylum seekers, many of these people fall into the legitimate Geneva Convention categories but may also be potential economic migrants as well. This is in no way contradictory. Thirdly, it

324

Immigration and the Transformation of Europe

is particularly inappropriate to privilege, as European governments have long been doing, a strictly national approach to a transnational and, as already noted, global phenomenon. This is all the more true for the fact that, through the European integration process, a group of European states was becoming more and more interdependent. In the international migration sphere, each state’s policy has an impact on its partners’ situation. For example, when Germany decided to abolish visas for Polish nationals, there was a potential risk of migratory flows for neighboring countries whose borders with Germany are now much easier to cross. In other words, by opening its frontiers, Germany was also opening those of its partners in the EU. In these circumstances, the absence of common tools for European immigration management raises serious problems. Chapters 11, 16, and 17 all stress this theme as well. Three options are in fact offered if we recognize the fact that European immigration policies have been, for several decades, poorly adapted and inefficient. The first consists of a general opening of frontiers. Morally speaking, it is possible to dream of a totally open world, as mentioned above. However, in present circumstances, with such an array of economic, social, political, and environmental inequalities, only the well-off would benefit from a general opening of frontiers. As to the poor, they would run the risk of becoming a new proletariat, even a lumpenproletariat, globalized like the globalized economy. The second possibility is to make immigration policies even more restrictive by increasingly militarizing immigration control, by further detaining asylum seekers and increasing the expulsion of foreigners. We then risk the systematic undermining of the human rights we claim to defend. This choice would mean a great loss for democracy. Between the naive total open borders hypothesis and the hypocritical reassertion of the zero immigration hypothesis, there is maybe room for a third option: a proactive EU approach to immigration which could take into account the interests of sending and receiving countries and regions but also the interests of migrants themselves. The construction of such an approach will be one of the most important challenges of the next decade.  Adam, Ilke, Nadia Ben Mohammed, Bonaventure Kagne, Marco Martiniello, and Andrea Rea 2002 Histoires sans-papiers 1. Brussels: Vista Editions. Assante, Franca 1978 Il movimento migratorio italiano dall’Unit´a nazionale ai giorni nostri. Geneva: Droz. Caestecker, Frank 2000 Alien Policy in Belgium, 1840–1940. New York: Berghahn Books.

Towards a proactive immigration policy?

325

Caritas/Migrantes 2004 Immigrazione. Dossier Statistico 2004. XIV Rapporto sull’ immigrazione, Rome: Caritas/Migrantes. Castles, Stephen 2002, “Migration and Community Formation under Conditions of Globalization.” International Migration Review 36: 4, 1143–68. Castles, Stephen and Godula Kosack 1985 Immigrant Workers and Class Structure in Western Europe. 2nd edn. Oxford: Oxford University Press. Castles, Stephen and Mark J. Miller 2003 The Age of Migration: International Population Movements in the Modern World. 3rd edn. New York: Guilford Press. Collinson, Sarah 1993 Beyond Borders: West European Migration Policy towards the 21st Century. London: Royal Institute of International Affairs. Commission of the European Communities 2000 “Communication from the Commission to the Council and the European Parliament on a Community Immigration Policy,” COM (2000) 757 final. Brussels. Commission of the European Communities 2003 “Communication from the Commission to the Council, the European Parliament, The European Economic and Social Committee and the Committee of the Regions on Immigration, Integration and Employment,” COM (2003) 336 final. Brussels. Commission of the European Communities 2004 “Green Paper on an EU Approach to Managing Economic Migration,” COM (2004) 811 final. Brussels. Entzinger, Han B., Marco Martiniello and Catherine Wihtol de Wenden 2004 Migration between States and Markets. Burlington, VT: Aldershot, Hants. Enzenberger, Hans M. 1994 Civil Wars: From LA to Bosnia. New York: New Press. Favell, Adrian (1998) Philosophies of Integration. Immigration and the Idea of Citizenship in France and Britain. London: Macmillan Press. Groenendijk, Kees and Hampsink, Ren´e (1994) Temporary Employment of Migrants in Europe. Nijmegen: Reeks Recht & Samesleving. Hammar, Tomas 1985 European Immigration Policy: A Comparative Study. New York: Cambridge University Press. Hollifield, James F. 1997 L’immigration et l’Etat-nation a` la recherche d’un mod`ele national. Paris: L’Harmatton. Joly, Daniele, Lynette Kelly and Clive Nettleton 1997 Refugees in Europe: The Hostile New Agenda. London: Minority Rights Group. Layton-Henry, Zig 1992 The Politics of Immigration: Immigration, “Race” and “Race” Relations in Post-War Britian. Oxford: Blackwell Publishers. Martens, Albert 1976 Les Immigr´es: Flux et reflux d’une main-d’oeuvre d’appoint: La Politique belge de l’immigration de 1945 a` 1970. Leuven: Universitaire Pers. Martiniello Marco and Govaere, Inge (1989) “Place de l’immigration et politiques migratoires dans l’Europe de demain.” Contradictions 56: 143–61. Martiniello M. and Andrea Rea 2001 Et si on racontait . . . Une histoire de l’immigration en Belgique. Brussels: Communaut´e Française de Belgique. Martiniello, Marco 2001 La Nouvelle Europe Migratoire: Pour Une Politique Proactive de l’ immigration. Brussels: Labor. Noiriel, G´erard 1988 Le Creuzet Français: Histoire de l’immigration, XIXe–XXe si`ecles. Paris: Seuil.

326

Immigration and the Transformation of Europe

Sayad, Abdelmalek 1991 L’immigration, ou, Les Paradoxes de l’alt´erit´e. Brussels: De Boeck Universit´e. Sayad, Abdelmalek (ed.) 1999 La Double Absence: des illusions de l’´emigr´e aux souffrances de l’immigr´e. Paris: Seuil. Smith, Anthony D. 1986 The Ethnic Origins of Nations. Oxford: Basil Blackwell. Smith, Anthony D. 1995 Nations and Nationalism in a Global Era. Cambridge: Polity Press. Verdusco G. 2004 “The Mexican Labor Force and Economic Interaction in North America,” in H. Entzinger, M. Martiniello, and C. Withol de Wenden (eds) Migration between States and Markets. Aldershot and Burlington Ashgate: Research in Migration and Ethnic Relations Series, 133–51.

13

European immigration in the people’s court Jack Citrin and John Sides

Introduction What is Europe to do about immigration? Mortality is up and fertility is down, resulting in an ageing population and a steadily shrinking workforce in most European countries (United Nations Population Division 2000). Until the invention of robots paying taxes, the policies required to preserve the kind face of European capitalism are either painful or unfeasible. Are Europeans prepared to have more children, work until the age of seventy-five, pay steeper taxes, or accept reduced retirement and health-care benefits? Because immigrants tend to be younger than the native-born populations of European countries and to have larger families, an active immigration policy seems an obvious response to the dictates of demography and economy (Joppke 2002:259). And there is no shortage of people clamoring to get in. Nevertheless, since the 1970s, most European governments have tried to “stem” rather than to “solicit” migrants (Joppke 2002; but see also Freeman 1995). For all the talk of post-nationalism and globalization among scholars, the movement of people across borders remains less politically acceptable than that of capital or goods. As this chapter will show, among European publics enduring cultural loyalties seemingly trump the economist’s rational calculus in which workers are infinitely substitutable. The current phase of European integration is supposed to produce common immigration and asylum rules. Yet most EU leaders have rejected proposals to establish a Union-wide immigration quota as a response to the numbers of workers and refugees eager to get in.1 Germany’s interior minister put it bluntly in October 2003: “We Germans have to be able to decide what our absorption capacity is. A system in which the Commission decides cannot function. Let us have no illusions; even a country that says it agrees with a given quota will nevertheless have to be able to say who 1

See Luedtke (2005) and the chapter in this volume for discussions of the difficulties in “Europeanizing” immigration policy.

327

328

Immigration and the Transformation of Europe

can and who cannot enter” (BBC, October 27, 2003). Allegedly “postnational” European states still seek to “monopolize the legitimate means of movement” across their borders, even if they pool their administrative resources to do so (Torpey 2000: 34–6). Deciding how to draw the lines between insiders and outsiders is national sovereignty’s final stand. If descent or cultural affinity is what creates the “imagined community” that is a nation (Anderson 1983), then the immigration of mainly non-white and Muslim populations poses a threat to the very identity of a people. This is the recurring and common theme of the anti-immigrant parties and movements that have surfaced in Austria, Belgium, Denmark, France, The Netherlands, and elsewhere. Even before September 11 sparked fears of terrorism and the radicalization of Muslim youth, the ideological pendulum in Europe was swinging from celebrating multiculturalism to demanding assimilation (Brubaker 2001). To cite just the well-known Dutch case, Pim Fortuyn declared that Holland is full, that there should be no further immigration, and that newcomers were to blame for increases in crime. Pronouncing Islam a “backward” culture for its treatment of women and gays, he insisted that immigrants living in The Netherlands assimilate to Dutch values and culture (Dreher, National Review Online, May 7, 2002). The LPF party founded by Fortuyn has proposed making fluency in Dutch a condition of naturalization, and other parties have endorsed LPF proposals requiring foreign spouses joining partners settled in The Netherlands to complete integration courses in their country of origin, partly as a barrier and partly to cut costs (Fleishman, Los Angeles Times, April 18, 2004). In Rotterdam, the city council has proposed a five-year moratorium on foreign residents (Ehrlich, Christian Science Monitor, December 19, 2003). In February 2004, the Dutch government made the controversial decision to deport 26,000 asylum seekers, mostly to Iraq, Afghanistan, and Somalia, over a three-year period (Migration News, April 2004). And the assassination of Theo Van Gogh by the Dutch-born child of Moroccan immigrants sent shock waves throughout Europe. In Germany, the Christian Democrat politician, Marcus Soder, warned of a potential “clash of civilizations” a` la Huntington (Leiken 2005). Top-down governance has its limits. Where officials ultimately must win a verdict in the people’s court, public opinion both constrains public policy and has the potential to reshape it. The rise of populist parties has battered the elite consensus that tolerated immigration and preached multiculturalism, if only because of the long-running need for population growth. Even where anti-immigrant parties win a small share of the vote, their electoral presence moves more

European immigration in the people’s court

329

mainstream parties toward a more restrictionist stance for fear of being outflanked. This chapter, therefore, describes European beliefs about how many immigrants should be admitted, the kinds of immigrants that should be admitted, and the economic and social impact of immigration on the receiving country. It then considers the role of economic and cultural motivations underlying public opposition to immigration and examines whether and how the size of a country’s foreign-born population and the state of its economy matter. The evidence for this paper comes from the European Social Survey (ESS) of 2002–03. The ESS is an ongoing research project funded by the European Commission’s Fifth Framework Program. It is designed to collect data on the attitudes and behavior of representative samples of European populations aged fifteen and above and to achieve a minimum effective response rate of 70 percent. The first survey included a module of fifty-eight questions relating to immigration and refugee matters and these are the centerpiece of our analysis. Data regarding immigration attitudes from the first round of surveys comprise responses from twenty countries. These include fourteen members of the European Union (EU) at that time (Austria, Belgium, Denmark, Finland, France, Germany, Great Britain, Greece, Ireland, Italy, Luxemburg, The Netherlands, Portugal, Spain, and Sweden), three eastern European countries that entered the EU in 2004 (the Czech Republic, Hungary, Poland), a useful feature that allows the creation of a baseline for studying trends in attitudes once outsiders and potential immigrants turn into insiders, and two non-members (Norway and Switzerland).2 Interests and identities as causes of immigration attitudes Two themes have dominated efforts to explain attitudes about immigration. One is the familiar hypothesis of economic self-interest (Citrin et al. 1997; Esses et al. 1998; Lahav 2004; Sniderman et al. 2004). From this perspective, antagonism toward immigrants, who presumably are identifiable by some visible marker, is based on the threat they pose to one’s material circumstances. The political debate about immigration tends to focus on this socially acceptable issue, with protagonists generally weighing the economic costs and benefits of immigration, including the threat 2

The ESS also included samples in Israel and Slovenia, which we do not utilize here.

330

Immigration and the Transformation of Europe

to jobs and wages versus the need for people to do the dirty and dangerous jobs many native-born workers eschew. Interest-based explanations of attitudes about immigration assume that perceptions of costs and benefits are “realistic” (Hardin 1995). People are motivated to protect and improve their material circumstances. Immigrants sometimes are a threat, and sometimes not. Scholars have shown that opposition to immigration and support for anti-immigrant political parties increase when unemployment increases (Fetzer 2000; Jackman and Volpert 1996). Permitted levels of immigration tend to decrease during economic hard times (Money 1999). Similarly, others have found cross-sectional evidence that opposition to immigration is higher among lower-income or less skilled workers (Scheve and Slaughter 2001). Advocates of this line of reasoning sometimes differ over whether the threats to which people respond are perceived as personal (“to me”) or collective (to the country or society as a whole). But the underlying logic of preference formation is the same. It is important to note also that from a theoretical perspective, prejudice does not enter into the perception of threat or the calculation of self-interest. It should not matter whether immigrants are white, yellow, brown, or black, or whether they are Christians, Moslems, or Jews. If there is a preference for immigrants of one ethnicity over another, this should reflect differences in the estimates, presumably accurate, of the impact of these groups on one’s material interests, including one’s tax burden or access to government services. The main alternative to the group interest explanation for attitudes about immigration is symbolic politics theory. This perspective emphasizes the potency of general values and identifications on opinion formation (Sears 1993; Chong 2000). Of particular relevance in the present research is the role of national and cultural identities. Ernest Gellner’s (1983) pithy remark that in the modern world a national identity is as natural as a nose and ears does not reveal the content of a specific nation’s identity or whether people prioritize this group membership above other affiliations, such as those based on class or religion. It is important to know how people subjectively define their nation’s identity because dominant beliefs about what makes someone “truly” British, French, Dutch, Hungarian, or American function as rules for admitting someone into the national community. By definition, immigrants are outsiders in contexts where national identity is the basis of self-categorization. Accordingly, concerns about immigration may reflect fears about the persistence of a nation’s identity. If a sense of cultural threat is the foundation of negative attitudes toward immigration, then anti-immigrant sentiments should be more prevalent among people with a strong sense of

European immigration in the people’s court

331

national identity and those who define their nation’s identity in ethnic or cultural terms. Moreover, negative attitudes toward immigrants should be triggered when the sense of threat is heightened, a condition that seems likely to be met when there is a large influx of immigrants who are visibly different in appearance, customs, and values. A testable implication is that because immigrants from EU countries are less likely to have such prominent differences from the native population, they are less likely to be considered a cultural threat and evoke less antagonism than the generally darker and Muslim migrants from the “South.” How many? Preferred levels of immigration How do Europeans feel about the appropriate level of immigration into their respective countries? The ESS asked a series of questions that referred to different kinds of immigrant populations.3 Two questions centered on race and ethnicity: “To what extent do you think [country] should allow people of the same race or ethnic group as most [country] people to come and live here?” and “How about people of a different race or ethnic group from most [country] people?” Respondents were given these options: allow many, allow some, allow a few, or allow none. The modal preference of respondents in the pooled set of samples was for the vague category of “some,” which we interpret as an acceptance of a modest level of immigration. In regards to immigrants of the same race or ethnicity, 16% answered “many,” 49% answered “some,” 29% answered “a few,” and only 6% answered “none.”4 Respondents were a little less welcoming when asked about immigrants of a different ethnic background: the majority said “many” or “some” (10 and 43%, respectively); while 36% said “a few,” and 11% said “none.” Answers to these two questions are highly correlated (r = .76). Of course, asking the questions in this precise order may have engendered a social desirability bias, such that respondents were less willing to express negative sentiments towards immigrants of a different ethnic background.5 3

4

5

The ESS immigration battery began with this preamble: “People come to live in [country] from other countries for different reasons. Some have ancestral ties. Others come to work here, or to join their families. Others come because they are under threat. Here are some questions about this issue.” Whenever we report results for the entire ESS sample, we weight responses so that each country’s sample is represented in proportion to that country’s actual population (using the ESS variable pweight). Results for individual countries are weighted to account for an unequal probability of selection into the sample within those countries (the ESS variable dweight). McLaren (2001) examines similar Eurobarometer questions and finds that “most individuals [appear] not to make a distinction between EC/EU and non-EC/EU immigrants”

332

Immigration and the Transformation of Europe

The ESS then differentiated immigrants in terms of place of origin and in terms of the economic status of this place of origin. Respondents were asked about “people from the richer countries in Europe” and “from the poorer countries in Europe,” and then about “people from the richer countries outside Europe” and “the poorer countries outside Europe.” In every case, the most common response (40–45 percent) was to allow some immigrants. It is noteworthy that there were only slight differences based on place of origin: Europeans were not significantly more likely to welcome immigrants from within Europe as opposed to outside Europe, or to welcome immigrants from richer countries as opposed to poorer countries – though, again, we suspect that a measure of political correctness is operating here.6 Do these beliefs vary across countries? To evaluate this question, we created an index that averaged responses to the six items about admitting immigrants. The index is coded from 0 to 1, where 1 indicates strong opposition to immigration.7 Table 13.1 presents the average score within each of these 18 countries. These results demonstrate that intensely exclusionary attitudes are not the dominant point of view, in that the means in most countries are below the midpoint of the scale (.5), but this inference depends, of course, on how one interprets the ambiguous word “some.” These results reveal that the aggregate level of opposition to increased immigration is strongest in Greece, Hungary, and Portugal, and weakest in Sweden and Switzerland. The perceived consequences of immigration Xenophobia and racial prejudice aside, beliefs about the effects of immigration, its potential costs and benefits, should be the determinants of preferences about how many and who to admit. How do people perceive the effects of immigrants, and do these perceptions vary across different domains – for example, the economy, crime, and culture? The ESS

6

7

(85). However, Lahav’s (2004) examination of other Eurobarometer indicators does find differences between attitudes towards African and Asian immigrants and attitudes towards European immigrants. Finally, Sniderman, Hagendoorn, and Prior (2004) randomize survey respondents into different experimental conditions that describe immigrants in different ways, and find that respondents react more harshly to immigrants who are described as less educated or as culturally dissimilar. As Pettigrew and Meertens (1995) discuss, there is a developing European norm against blatant prejudice, though this does not obviate the existence of what they term “subtle prejudice.” Moreover, subtle prejudice is in fact strongly linked to views of immigrants (Pettigrew 1998). We computed this index by scaling each individual item to range from 0 to 1 (such that 1 indicates allowing no immigrants into the country), and then averaging these items. The reliability of this index is very high overall (alpha = .94) and is comparably high in each individual country.

European immigration in the people’s court

333

Table 13.1 European beliefs about immigration and immigrants, by country

Country

Mean score on preferred level index

Country

Mean score on perceived consequences index

Greece Hungary Portugal Austria Finland UK Netherlands France Luxembourg Czech Republic Belgium Spain Poland Denmark Norway Germany Ireland Italy Switzerland Sweden

0.64 0.62 0.58 0.53 0.51 0.50 0.49 0.49 0.48 0.48 0.47 0.46 0.44 0.44 0.44 0.43 0.42 0.41 0.39 0.31

Greece Hungary Czech Republic Belgium UK Portugal Poland Germany Netherlands Ireland France Italy Spain Austria Norway Denmark Switzerland Finland Luxemburg Sweden

0.69 0.62 0.62 0.57 0.56 0.56 0.55 0.55 0.54 0.54 0.54 0.53 0.52 0.52 0.52 0.51 0.51 0.49 0.45 0.45

TOTAL

0.47

TOTAL

0.55

Cell entries are means from six-item indices that run from 0 – most accepting of immigration to 1 – most opposed to immigration. Source: 2002–2003 European Social Survey.

included a series of scales ranging from 0–10 where the endpoints indicate a bad outcome, such as “take jobs away,” and a good outcome, such as “create new jobs.” Collapsing responses into positive (0–4), neutral (5), and negative (6–10) positions of the scale, by and large these assessments suggest that negative views of the consequences of immigration are typically more common than positive views – though, as one might expect, there is a notable lump of respondents at the midpoint of each scale. Thirty-eight percent of respondents believe that immigrants make the country a worse place to live, and 34 percent believe that immigrants are bad for the economy.8 There is slightly more concern that immigrants 8

Other ESS items suggest a somewhat more positive view of immigrants. For example, only 38 percent agree that, “Average wages and salaries are generally brought down by people coming to work and live here,” while 60 percent agree that, “People who have come to live and work here help to fill jobs where there are shortages of workers.”

334

Immigration and the Transformation of Europe

take away jobs (41%) and demand more in services than they pay in taxes (47%). By a substantial margin, fear that immigration breeds crime is the dominant concern: 69 percent of all respondents believe that immigrants make crime “worse.” One indication of the salience of this anxiety is that, in response to another question, 81 percent of ESS respondents agreed that immigrants who commit a serious crime should be deported. Somewhat surprisingly, most respondents do not believe that immigration has negative cultural consequences. Indeed, a majority (51%) place themselves on the positive, “cultural enrichment” side of the scale. Given previous evidence regarding concerns about national and cultural identity, we speculate that the wording of the item about cultural enrichment may be eliciting positive feelings about the introduction of new foods, music, dance, and clothing styles rather than focusing attention on the consequences for the erosion of the dominant national culture. Which of these perceived effects is most strongly related to views about the level of immigration? Although these items are highly correlated, the large sample size permits the separation of their separate connections to preferences about the right level of immigration through multiple regression analysis. This is not to insist that the beliefs about consequences “cause” preferences about the volume of immigration. The opposite flow also is plausible, with beliefs about consequences functioning as rationalizations for generalized hostility to immigrants. Here we view these relationships in correlational terms, although the reality probably is that the beliefs about impact and policy preferences are mutually reinforcing. In this simple regression model, the perceived cultural consequences of immigration are most strongly related to restrictionist attitudes (b = .242), with economic consequences a very close second (b = .225). Crime has an impact about one-half the size (b = .110). Beliefs about the consequences for jobs and taxes/services have a smaller impact. Thus, even though the majority of respondents believe that immigrants tend to enrich their host country’s cultural palette – and thus the distribution of this item is not particularly anti-immigrant – it is strongly associated with opinions about the “right” level of immigration. Estimating this same simple regression model separately for each country and comparing the coefficients for each category of consequence reveals notable heterogeneity across the twenty countries in the magnitude of these coefficients. However, in most cases (fourteen of twenty countries) fears that immigration will undermine a nation’s culture and hurt its economy are the two most potent predictors. However, their relative influences vary, with cultural threat mattering relatively more in some countries (Austria, Belgium, Denmark, Finland, Poland, Spain, and

European immigration in the people’s court

335

Sweden) and economic threat weighing more heavily in others (United Kingdom, Czech Republic, Italy, Portugal, Hungary, France, and The Netherlands). The several assessments of the consequences of immigration were combined into a single index.9 As with the previous index, this index is scaled from 0 to 1, where 1 indicates the most negative assessment of the effects of immigration. Table 13.1 also presents the means of this index by country. In the majority of countries, the average assessment is on the negative side (i.e. greater than .50). The rank order of countries appears quite similar to the order of countries for the index of preferred levels. At the country level, the preferred level of immigration is, as one would expect, strongly correlated with assessments about its effects (r = .69). Contextual explanations for attitudes Do contextual factors explain the observed variation across countries? To answer this question, we test two hypotheses derived from previous literature: that opposition to immigration derives in part from, first, the level of immigration into a country and, second, from the economic situation of that country. More specifically, opposition should be strongest in countries with large immigrant populations and those that are economically disadvantaged. Table 13.2 reports the coefficients and measures of fit from a series of bivariate regressions between country-level measures of immigrant numbers and economic health and the country-level means on these indices of perceived consequences and preferred levels. The statistical and substantive significance of these coefficients and the amount of variance these country-level variables explain will suggest whether attitudes about immigration are grounded in “objective” conditions within countries. The present data find that attitudes towards immigration at the country level are not related to the size of the overall foreign-born population. The bivariate relationship between the country-level opposition to immigration reported in Table 13.2 and the proportion of the country’s population that is foreign-born are small, insignificant, and, interestingly, in the opposite direction from the hypothesis.10 (See Appendix Table 13.A for a description of this and other country-level measures.) The same is true when the contextual factor is measured as the percent change in the proportion foreign-born from 1992–2001. One can distinguish 9 10

The reliability of this index is .84 overall, and it is greater than .70 in all countries. If one removes the outliers of Luxembourg and Switzerland, both of which have large foreign populations mostly composed of EU nationals, the figures in Table 13.3 are still negative and insignificant.

336

Immigration and the Transformation of Europe

Table 13.2 Relationships between demographic and economic context and immigration opinions Dependent variable Mean score on preferred level index

Demographic measures Percent Foreign in Population (2001) % Change in Foreign (1992–2001) Percent non-EU Foreign in Population (2001) % Change in non-EU Foreign (1991/1996–2001) Percent Asylum-Seekers in Population (2002) Economic measures Unemployment 2001 Unemployment 2002 % Change in Unemployment 2001–02 GDP 2001 GDP 2002 % Change in GDP 2001–2002 Economic Hardship Index(GDP & Unemployment ’02 and change ’01–02)

Mean score on perceived consequences index

b

r2

b

r2

−.002 .0001 −.002 −.05

.03 <.01 .01 .09

−.003 .0002 −.001 −.005

.14 .01 <.01 <.01

−.037∗

.21

−.025∗

.18

.001 .001 −.0003 −.003# −.003# .02# .03

<.01 <.01 <.01 .12 .11 .10 .06

.004 .004 −.003∗ −.004∗∗ −.004∗∗ .02∗ .05∗∗

.10 .07 .23 .45 .44 .15 .37

Table entries are unstandardized regression coefficients and r-squared statistics from bivariate regressions between each country-level measure and the country-level mean on each of the two indices. N = 20. # p < .10; ∗ p < .05; ∗∗ p < .01. Sources: 2002–2003 European Social Survey, OECD.

the foreign-born who originate from within and outside the EU, following Quillian (1995) and Lahav (2004), but the percent of a country’s population comprised of non-EU foreigners has a similarly unimpressive relationship with both measures of attitudes towards immigration (b = −.003 and b = −.002, both n.s.).11 The change in this proportion from either 1991–2001 or 1996–2001 (the time frame varies depending on data availability) has a similarly insignificant statistical relationship to the measure of immigration attitudes. Another potential population that “stands out” within European countries is asylum seekers. This type of immigrant may engender a stronger 11

We also explored other measures of immigrant demographics, including the percentage of immigrants who hailed from various parts of the world (e.g., Africa, the Middle East, etc.). In no case were any of these measures significantly and positively related to attitudes across these twenty countries.

European immigration in the people’s court

337

reaction from native populations, as they are usually ethnically different, less educated, and in greater need. ESS respondents reacted more negatively to asylum seekers than to immigrants as a whole, as 64 percent agree that their country “has more than its fair share of people applying for refugee status,” and 47 percent agree that “Most applicants for refugee status aren’t in real fear of persecution in their own countries” (vs. 22 percent who disagree). However, the presence of more refugees does not breed antagonism toward immigrants in general. The volume of asylum seekers from 1993–2002 – when expressed as a percentage of each country’s population in 2002 – is actually associated with less hostility towards immigration (b = −.025 and b = −.037, p < .05), largely because the countries with the largest proportions of asylum seekers, Sweden and Switzerland, are those countries most welcoming of immigrants. This suggests that the causal arrow may run in the opposite direction: asylum seekers gravitate to the countries most tolerant of them.12 To examine the relationship between a country’s economic situation and its views of immigrants, we consider four measures of the economy: the unemployment rate, the gross domestic product (GDP) per capita (based on purchasing power parities), and the change in each of these indicators.13 Though immigrants are commonly thought of as a threat in the job market, there is no relationship between attitudes towards immigration and the unemployment rate in 2001 or 2002. The change in the unemployment rate between these two years is actually negatively associated with hostility to immigrants. There is a very modest relationship with GDP per capita in both 2001 and 2002. In general, opposition to immigration decreases in wealthier countries. However, the change in GDP, like the change in unemployment, has a counterintuitive relationship with attitudes towards immigrants: a positive change in GDP from 2001–02 is associated with a higher level of opposition to immigration. This relationship could arise because economic growth creates rising expectations, exacerbating the perceptions of relative deprivation that can fuel resentment towards immigrants.

12

13

A more contemporaneous measure – the proportion of asylum seekers in 2001 – generates this same finding. An alternative measure of demography – the change in the percentage foreign-born from 1992–2001 (OECD 2004: 308) – was also unrelated to attitudes. Finally, substituting the country-level means on the individual items in these indices generates results that are at best comparable in their modesty and at worse substantially weaker. We also examined measures of the size of welfare state and the type of welfare state (see Esping-Anderson 1990). These measures did not have any substantively or statistically significant relationships with attitudes towards immigration.

338

Immigration and the Transformation of Europe

Given these mostly insignificant and unexpected relationships, we attempted to build a more robust measure of a country’s economic status, drawing on information about both GDP and unemployment, and about both levels and change over time. We created an index of “economic hardship” by standardizing and then averaging GDP and unemployment in 2002, as well as the change in GDP and unemployment from 2001–02. This index has a high degree of reliability (alpha = .70), and higher values indicate a worse economy. As Table 13.2 shows, this measure is positively, but not significantly, related to the desired level of immigration and both positively and significantly related to the perceived consequences of immigration. Country-level differences in demographic make-up or economic attainment thus have limited power in explaining aggregate differences in attitudes toward immigration. The findings based on more recent data and more comprehensive measures of immigration attitudes diverge from Lahav’s (2004) assertion that “the degree to which Europeans reported immigrants and asylum-seekers to be a problem seemed to be explained largely by the numbers of resident foreigners” (117).14 Of course, changes in demography or economic health might explain changes over time within countries (Fetzer 2000; Jackman and Volpert 1996; Money 1999), but similar measures do not help us differentiate across these 20 nations. Perception and misperception of immigrant levels Perceptions as well as reality shape preferences. The prominence of differences between immigrants and the native population makes their presence more vivid and salient. This, along with the tendency of migrants to congregate in large cities, may lead to overestimates of their number which in turn influence opinions about whether more should be admitted. Freeman (1995: 883) argues that “there are serious barriers to acquisition of 14

The differences may arise because her measures are separate single-item questions asking whether immigration is a “big problem” and whether there are too many non-EU foreigners in the nation, because her data come from Eurobarometer surveys that date from 10–15 years ago and that contain a slightly different set of countries, or because the empirical reality has simply changed over the fourteen years between those surveys and the ESS. Lahav finds that the countries which express the strongest anti-immigrant sentiments are Germany, Greece, France, Belgium, The Netherlands, Ireland, and Denmark. This contrasts sharply with our Table 13.1, where of those countries only Greece is demonstrably more anti-immigrant. When we limit our analysis to the EU Member States in 2000 that were included in the ESS sample, the correlation between attitudes and Lahav’s preferred measure, the proportion of non-EU immigrants in the country, is larger (r = .27), though a bivariate regression reveals no significant relationship. It is also true that Quillian (1995) finds no direct relationship between the percent of non-EU immigrants in the country and prejudice towards immigrants in every model specification (see his Table 2: 600–1).

European immigration in the people’s court

339

information about immigration” and thus there is every reason to expect the public to be poorly informed. However, Lahav (2004) concludes the opposite after finding a direct relationship between the number of immigrants a country receives and the number of Eurobarometer respondents who cite that country as having the most immigrants: “We may infer that attitudes toward immigrants are neither random nor biased in some systematic direction, but based on informed views” (78). The ESS asked respondents for a comparative assessment: “Compared to other European countries of about the same size as [country], do you think that more or fewer people come to live here from other countries?” The response demonstrates that most Europeans view their own country as a relatively popular destination: 19 percent of respondents said “far more people come to live here” and 38 percent said “more people come to live here.” Only 14 percent of respondents said that “fewer” or “far fewer” immigrants come to live in their country than in another of comparable size. A second question asked for an absolute estimate of the size of the foreign-born population: “Out of every 100 people living in [country], how many do you think were born outside [country]?” Answers to this question ranged between zero and 100, with a mean of twenty. Roughly 28 percent of respondents gave an answer between zero and nine, 26 percent between ten and twenty, 19 percent between twenty-one and forty, and the remaining 27 percent greater than forty. How accurate are these estimates? Figure 13.1 reveals them to be largely inaccurate. This figure plots the country-level estimates of the foreign-born population against the actual foreign-born population, with a line that corresponds to an accurate estimate. Every single country, with the exception of Switzerland and Germany, is above this line, meaning that respondents routinely overestimate the percentage of immigrants in their country. At times, these misperceptions are quite significant: the estimates of respondents from France, The Netherlands, the United Kingdom, and Portugal, for example, are all about 20 percentage points too high. Thus, when confronted with the most direct test of knowledge about immigrant levels, respondents tended to fall short of accuracy. At the individual level, perceptions of immigrant numbers are strongly related to preferences about the level of immigration. For example, among those who believe that their country has far fewer, fewer, or about the same amount of immigrants relative to countries of comparable size, only 6 percent believe that no immigrants of a different race should be allowed in the country. Among those who believe there are “more” immigrants coming to their country, 10 percent support an absolute ban on immigrants of a different ethnic background. Among those who believe there are “far more” immigrants coming, 24 percent support this ban. A similar

340

Immigration and the Transformation of Europe Lux

Estimated percent foreign-born

40

30 Switz

Fr

Port

UK Neth Belg Aus Swe Greece

20

Ger

Italy Spain Hung

Ire Nor Den

10 Czech Pol Fin

0 0

10

20 30 Actual percent foreign-born in country

40

Figure 13.1. Estimated vs actual percent foreign-born in country. Source: 2002–2003 European Social Survey, OECD.

relationship emerges when considering people’s estimates of the absolute number of immigrants. Among those who estimate that 10 percent or less of the country’s population are foreign-born, only 7 percent support barring immigrants who are ethnically distinct. Among those who estimate that 40 percent or more of their country are foreign-born, 19 percent have this opinion. If one creates a measure of “misperception” by subtracting the actual percent foreign-born from these estimates, a similar relationship emerges (r = .14). The implication here is that if people were provided with more information about actual levels of immigration, they might be more willing to admit immigrants.

Beliefs about cultural unity We have argued that public opinion towards immigrants may depend both on the strength of the individual’s sense of national identity and the extent to which cultural homogeneity is a fundamental component of how nations conceive of themselves. Earlier research indicates that when immigrants are widely regarded to threaten a country’s self-concept or way of life, opposition to immigration and the insistence that immigrants assimilate to the dominant culture are likely to increase. The ESS asked a

European immigration in the people’s court

341

series of items that permit a more extended analysis of how beliefs about the virtues of cultural unity are linked to opinions about immigration. An overwhelming majority of European publics believe in the benefits of a common language: 91 percent of ESS respondents agree that, “It is better for a country if almost everyone is able to speak at least one common language.” There also is general agreement that, “It is better for a country if almost everyone shares the same customs and traditions”: 51 percent hold this view, compared to 26 percent who disagree, with the remainder uncertain or ambivalent. In an increasingly secular society, religious unity ostensibly matters less than a common language or cultural tradition, although the ESS did not include a specific question about Islam. When asked if, “It is better for a country if there are a variety of different religions,” 36 percent of the ESS respondents agree and 32 percent disagree. At the same time, there is majority opposition (54 percent to 30 percent) to the idea that, “Communities of people who have to come to live here should be allowed to educate their children in their own separate schools if they wish.” There is, then, a general belief in the benefits of cultural unity for social harmony and the desire that immigrants acculturate, particularly by learning their new country’s language. This conclusion is reinforced by responses to a question asking respondents to indicate the importance (by giving a score on a 1–10 scale) of particular attributes in deciding whether or not someone “born, brought up, and living outside the country should be able to come and live here.” The criterion emphasized most often was people “committed to our way of life”: 80 percent of respondents gave this attribute a score of between six and ten. The proportions giving a similar ranking were 72 percent for the attribute of “work skills that the country needs” and 70 percent for “speaking the country’s language.” By contrast, having close family already in the country was ranked high by barely 50 percent of respondents and being wealthy was emphasized by just 29 percent. There is a strong relationship between attitudes towards cultural unity and attitudes towards admitting more immigrants. For example, 24 percent of respondents who favor overall cultural unity – those who say, “It is better for a country if almost everyone shares the same customs and traditions” – support an outright ban on immigrants of a different ethnic background, and only 4 percent advocate allowing “many” immigrants of a different ethnicity into their country. Among those who strongly disagree with this measure of cultural unity, only 4 percent support such a ban and 39 percent actually favor allowing “many” such immigrants into the country. Finally, the cultural unity item is also strongly correlated with the index of preferred immigration levels (r = .36). This finding suggests

342

Immigration and the Transformation of Europe

the important role of cultural motivations in shaping preferences about immigration among ordinary citizens. A model of attitudes towards immigration To complete the empirical analysis of public opinion, we present multivariate models of perceptions of the consequences of immigration and opinions about the appropriate level of immigration, as measured by the indices described above. These models incorporate both individual-level and country-level attributes as predictors and the statistical technique is a hierarchical, or multilevel, estimation procedure (see Raudenbush and Bryk 2002; Luke 2004; Steenbergen and Jones 2002). At the individual level, the primary concern is the causal role of three main factors – interests, identities, and information – after controlling for other relevant factors. (Appendix Table 13.B summarizes the question wording and coding for each individual-level variable.) To begin with the measures of economic interests, the predictor is the standard sociotropic question asking about the respondent’s perception about the state of the national economy. To assess personal economic concerns, we construct a two-item measure addressing the respondent’s sense of security about being able to cope with financial stress.15 In addition, we include as predictors measures of the respondents’ objective economic circumstance: their income and whether they are unemployed. The expectation is that respondents who are pessimistic about the nation’s economy, financially insecure, poorer or unemployed will be more opposed to immigration – because of a generalized economic anxiety that they project onto outsiders or because they see immigrants as a direct threat to job opportunities and wage levels. The predictors of cultural and national identities in the multivariate model include agreement with the item, “It is better for a country if almost everyone shares the same customs and traditions.”16 Agreement may be interpreted as support for an “ethnic” definition of nationhood, which should be associated with opposition to immigration. Unfortunately, the ESS does not include any general measure of the overall strength of 15

16

These items are: “Which of the descriptions on this card comes closest to how you feel about your household’s income nowadays?: living comfortably, coping, finding it difficult, finding it very difficult” and “If for some reason you were in serious financial difficulties and had to borrow money to make ends meet, how difficult or easy would that be?” These items were combined into a scale (alpha = .59). We do not incorporate the other items in that battery of questions, e.g., into an overall index, because they are not strongly correlated and as such any index’s reliability would be questionable.

European immigration in the people’s court

343

national identity – for example, degree of pride in country. However, there is an alternative means to measure this concept. Citrin and Sides (2004) show that European respondents who said they were attached to Europe as well as their country were more tolerant of ethnic minorities and new members of the EU. Although the ESS did not ask respondents about the strength of their attachment to nation or Europe, it did ask about preferences for policy-making at either the European or national level for eight policy domains, including immigration and refugees. Citrin and Sides also find that how one prioritized attachment to the nation and Europe was strongly associated with a willingness to delegate policy-making jurisdiction to the EU (see also Luedtke 2005). Accordingly, a measure of “preference for supranational authority” can be constructed by summing answers to the eight ESS items in this domain and we employ this measure as a proxy for a direct assessment of the strength of the respondent’s national identity. The expectation here is that a greater willingness to cede political authority to European and international institutions will mitigate opposition to immigration because this indicates both a less intense attachment to the nation as a focus of political identity and a greater willingness to go with the official European attitude toward immigration, which is benign. The role of information is assessed by the comparative estimate of immigrant levels and a measure of absolute “misperception,” calculated as the difference between each respondent’s absolute estimate of immigrant levels and the actual proportion of foreign-born residents in the respondent’s country (see Figure 13.1).17 We hypothesize that perceptions of the level of immigration as higher than the European norm and as high in the absolute will create greater opposition to immigration. We also examine the effect of the interaction of these two measures, because it is likely that an assessment of immigration that is high in both absolute and relative terms will engender even stronger opposition than the additive effects of each belief. An additional hypothesis is that respondents who perceive immigrants to be different from their country’s majority will have more negative views of immigrants. The ESS asked respondents whether they thought most immigrants in their country are “of the same race or ethnic group as the majority of” people in that country, whether most were of a different race or ethnic group, or whether it was “about half and half ” and similar 17

Approximately 17 percent of the ESS sample could not provide a numerical estimate of the immigrant population. To avoid throwing out these cases, we imputed values on a country-by-country basis, drawing on respondent’s level of education and frequency of political discussion as predictors.

344

Immigration and the Transformation of Europe

questions about whether most came from the richer countries in Europe, the poorer countries in Europe or from outside Europe. Beliefs that most immigrants are ethnically different or from poor countries are hypothesized to foster more negative attitudes regarding the consequences of immigration and a preference for restricting future access. Among the variables included as controls are political awareness and political ideology. Because the dominant outlook among Europe’s political, business, and academic elites is favorable toward immigration and tolerance toward immigrants (Lahav 2004), those who are attentive to politics should be more likely to learn and adopt a similar outlook than the less politically interested and engaged (see Zaller 1992). Following other studies of European attitudes “cognitive mobilization,” a cognate of political awareness, is measured by the self-reported frequency of discussing politics, and is related to support for strengthening the EU (Gabel 1998) and for immigration (McLaren 2001). Political ideology also is consistently related to immigration attitudes in European politics, with opposition concentrated on the political right. So location on the left–right continuum, measured here by self-placement on a 0–10 scale, should be related to immigration attitudes. However, the impact of the respondent’s ideological position should be conditioned by political awareness. That is, if there is a public debate about immigration in which the elite consensus is challenged, as, for example, by Pim Fortuyn, then the politically aware on the right should move toward greater opposition while those on the left continue to follow the lead of the more liberal elites. In other words, the polarization of the left and right on questions of immigration occurs most strongly among those who are high in awareness (Zaller 1992). To test this proposition we included a multiplicative interaction term of ideology and awareness. The model elaborated here also includes factors deemed to increase tolerance of ethnic minorities. The contact hypotheses (Pettigrew and Meertens 1996) holds that inter-ethnic interactions will boost acceptance of immigrants, so we include a measure recording the respondent’s report on personal friendships with immigrants. Another potential psychological predictor is social trust. The social trust measure is a core component of Putnam’s (2000) notion of social capital and captures a willingness to cooperate with others in collective enterprises. Those who are more trusting (and therefore less “alienated”) should be more supportive of immigration. Similarly, we include a measure of overall life satisfaction, which is the response to this question: “All things considered, how satisfied are you with your life as a whole nowadays?” Those who are satisfied

European immigration in the people’s court

345

with their lives should be more supportive of immigration, whereas the unsatisfied, like those who are cynical rather than trusting of other people, will be less supportive, perhaps through a process of displaced aggression and scapegoating of those who are weak and different (Adorno et al. 1950). The remaining components of the model are demographic factors. These include self-identifications as a member of a minority ethnic group, immigrant status, age, and education. Social and political incorporation into the host country should diminish support for immigration and promote the development of views similar to those of the native majority. By contrast, recent immigrants, those who are more “marginal” in Fetzer’s (2000) terminology, should be more likely to see themselves as the target of negative beliefs about the consequences of immigration and therefore reject them to maintain a sense of self-esteem. Moreover, recent immigrants should have a stronger interest in keeping the borders open, as they are more likely to seek to bring family members to join them. Finally, the young and those with more formal education are consistently more likely to express tolerance for racial and ethnic minorities (see Hagendoorn and Nekuee 1999). In addition to the long list of individual-level factors summarized above, the model includes two different contextual measures: the percentage of the foreign-born population that originates outside of the European Union, and the index of economic hardship. As discussed previously, these are related to attitudes towards immigration, with opposition to immigration increasing when a country has a larger proportion of foreign residents from outside the EU, and decreasing when GDP per capita increases. While these effects were not as pronounced as in some extant literature, their influence is worth testing in the more elaborate model articulated here. Results The models are estimated using the restricted maximum likelihood routine in the HLM 6.0 package developed by Raudenbush and Bryk (2002). The first model estimated is a random-intercepts model, where the attitude of an individual within a country is a function of both individual-level and country-level effects. Initially, the individual-level slopes are treated as fixed in this specification. Each of the dependent variables and each of these individual-level independent variables are coded 0 to 1. This means that the individual-level coefficient estimates are somewhat comparable across variables, in the sense that they can be regarded as an estimate of the numerical change on the 0–1 anti-immigration index that results

346

Immigration and the Transformation of Europe

from moving from the lowest to the highest category of the independent variable.18 One advantage of hierarchical models is that they allow one to estimate the relative contribution of each level of data to the overall variance in the first level of data. In this case, it allows us to determine how much of the variance in individual attitudes towards immigration is explained at the individual level and how much is explained at the country level. Decomposing variance in this fashion entails estimating a simple multilevel model that includes only intercepts at the country and individual levels (see Luke 2004: 20–1). Such a model reveals that individual-level factors account for 89.3 percent of the variance in preferred levels of immigration, while country-level factors account for the remaining 10.7 percent. The same figures for perceived consequences of immigration are 89.2 percent and 10.8 percent. This demonstrates that individual-level factors matter much more in explaining attitudes. That country-level factors explain only a small fraction of variance dovetails with our previous finding that few country-level factors are significantly associated with attitudes (see Table 13.2). Table 13.3 reports the results of the estimated models. The effects of the individual-level variables confirm many of our expectations. Economic satisfaction tends to decrease opposition to immigration. In line with previous research (Citrin et al. 1997; Lahav 2004), sociotropic orientations outweigh personal financial concerns. For example, in the model of perceived consequences, the effect of sociotropic evaluations (b = −.091) is much greater than that of personal evaluations (b = −.018). In the model of preferred levels, this gap shrinks, but sociotropic concerns still have twice the substantive effect of personal evaluations. Income and unemployment are at best weakly associated with these two dependent variables – suggesting a smaller role for labor market position in boosting opposition to immigration than reported by researchers studying American public opinion (Scheve and Slaughter 2001). Turning to the role of symbolic attitudes, the two measures of cultural and national identities also have significant effects. The preference for cultural unity and the preference for retaining national authority, our 18

The country-level variables are centered around their grand mean, so that there is a meaningful zero-point for each variable (see Luke 2004: 48–51). See Appendix Table 13.A for more information about these measures. We also performed diagnostic tests on the hierarchical models to confirm that various assumptions were met, in particular that the level-1 residuals and the level-2 random effects were normally distributed around 0. The level-1 residuals appeared to meet this assumption. There were small deviations from normality in the level-2 random effects, which is not surprising given that there are only 20 level-2 units.

Table 13.3 Predicting opposition to immigration

Concept Individual-level Economic interests

Variable

Satisfaction with personal finances Satisfaction with economy Income Employment: unemployed Employment: student Employment: retired, etc. Cultural and Prefer cultural unity national identities Preference for retaining national authority Information about Comparative estimate of immigration Absolute misperception of immigration immigration Comparative × absolute estimate Perceptions of Perceive immigrants to be different immigrant race attributes Perceive EU immigrants to be low SES Perceive non-EU immigrants to be low SES Contact alienation Have immigrant friends Social trust Life satisfaction Political awareness Political discussion Conservatism and ideology Conservatism × Political discussion Immigrant status Self-identified minority Second generation Naturalized (>10 yrs in country) Naturalized (<10 yrs in country) Non-citizen (>10 yrs in country) Non-citizen (<10 yrs in country) Other controls Education Age Female Country-level Percent of population from outside EU Economic hardship Constant % reduction in error (individual-level) 2×Log-likelihood

Preferred levels index

Perceived consequences index

−.05∗∗ −.047∗∗∗ −.028∗∗∗ −.003 −.046∗∗∗ −.005 .169∗∗∗ .052∗∗∗

−.018∗∗ −.091∗∗∗ .006 .003 −.019∗∗∗ .001 .147∗∗∗ .043∗∗∗

.102∗∗∗ .019 .125∗∗∗ .031∗∗∗

.098∗∗∗ −.016 .106∗∗∗ .037∗∗∗

.023∗∗∗ −.020∗∗∗

.014∗ .004

−.077∗∗∗ −.135∗∗∗ −.009 −.073∗∗∗ .032∗∗ .071∗∗∗ .001 −.026∗∗∗ −.006 −.042∗∗ −.069∗∗∗ −.030∗∗∗ −.087∗∗∗ .046∗∗∗ .001 −.027 −.002 .494∗∗∗ .12 −13589.9

−.060∗∗∗ −.136∗∗∗ −.020∗∗∗ −.042∗∗∗ .025∗∗ .055∗∗∗ −.022∗∗∗ −.026∗∗∗ −.024∗∗∗ −.055∗∗∗ −.059∗∗∗ −.109∗∗∗ −.061∗∗∗ −.019∗∗∗ .0003 .002 .008 .562∗∗∗ .17 −41637.2

Table entries are coefficients from hierarchical linear models. All dependent variables coded 0 (most favorable to immigration) to 1 (least favorable to immigration). All individual-level independent variables are coded 0 to 1. Level-1 N = 34,427. Level-2 N = 20. Sources: 2002–2003 European Social Survey, OECD. ∗ p < .05; ∗∗ p < .01; ∗∗∗ p < .001.

348

Immigration and the Transformation of Europe

proxy for a sense of patriotism and national attachment, boost negative assessments of immigration. In particular, fears about the consequences of cultural pluralism make people view immigration in a negative light. Indeed, the effect of this variable is arguably the largest of any in the model. Beliefs about the consequences of immigration do grow more negative when people perceive that immigrants to their country are of a different ethnicity and a lower socioeconomic status, but the substantive effects of each of these variables are quite small. Table 13.3 also indicates that believing that one’s country has a high number of immigrants relative to other countries promotes a negative assessment of immigration. In addition to this significant direct effect on immigration attitudes, there is an interaction between the comparative estimate and the absolute magnitude of one’s overestimate of the number of immigrants in one’s country. The joint effect of these comparative and absolute estimates is substantial. For example, a person who believes that his country received many more immigrants than countries of similar size, and who overestimates the proportion of immigration by the sample mean (about 10 percentage points), will have a score .125 greater on the preferred levels index than a person who believes that his country receives many fewer immigrants and who estimates the proportion of immigrants correctly. This shift is equal to about one-half of a standard deviation in this dependent variable. Political variables matter as well, confirming the results of earlier research. Cognitive mobilization, as captured by the self-reported frequency of political discussion, is associated with more positive views of immigrants. Conservatism engenders negative judgments about immigration, as expected, and that influence is particularly pronounced among those who engage in more frequent political discussions. The effect of ideology among the politically inactive is less than half the size that it is among the most politically active. Social-psychological factors have a surprisingly strong association with beliefs about the effects of immigration. Social trust and having immigrant friends tend to produce attitudes that are less anti-immigrant. The influence of socialization also is apparent: the young, the better-educated, and students, all embedded in social networks that generally are favorable to immigrants, express more favorable views of immigration. Beliefs about immigrants and immigration predictably are more positive among self-identified minorities, though the statistical effect is quite small. By contrast, nativity, naturalization, and longer residence in the receiving country increase the tendency to characterize the effects of immigration more negatively. It is as though integration (or assimilation) involves adopting the attitude of the native-born, which is less favorable

European immigration in the people’s court

349

to immigrants. Nevertheless, it seems clear that a continuing influx of immigrants is likely to generate political divisions about the issue. Finally, neither contextual factor in the model has a statistically significant effect on the perceived consequences of immigration.19 Once these various individual-level variables are controlled for, thereby taking into account the compositional differences among countries, neither a country’s demography nor its overall economic well-being affects its citizens’ perceptions of immigrants. Of course, context is measured here at the country level and this may be too crude or remote a measure to capture the environmental influences that shape experiences with and beliefs about immigrants.20 Although both cultural attitudes and economic concerns matter, these data suggest that cultural factors are more potent. Immigration is opposed when immigrants are perceived as a threat to one’s “way of life.” This interpretation is buttressed by the significant effect of perceptions rather than the facts about the level of immigration. Overestimating the number of immigrants predicts opposition to more immigration, and it is plausible that these misperceptions are a response to interactions with or media stories about visibly different groups such as Muslims and non-whites. In other words, the distinctiveness of some immigrants raises their salience and engenders exaggerated views of their numbers. In this context, the results pertaining to friendship with immigrants and social trust have policy relevance. They suggest that “genuine” contact across group lines can reduce one’s sense of threat and thus, increase acceptance of more immigration (Allport 1954). Variation across countries By combining the twenty country samples into a single European public, we have created an artificial leviathan. With such a large sample, small yet statistically significant effects are easy to attain. More importantly, the methodological approach adopted may conceal important countrylevel differences in the underpinnings of attitudes about immigration. To 19

20

We also examined other measures of economic health, including GDP and the unemployment rate. Neither was significant. Following Quillian (1995), we also included an interaction between the two contextual measures. This measure was not significant either. Hood and Morris (1997) find evidence that respondents in the US who live in counties with larger Asian and Hispanic populations tend to feel more positively towards immigration. Scheve and Slaughter (2001) do not find the same, however, using the metropolitan statistical area (or MSA) as the contextual unit. Perrineau (1985) finds that voting for the National Front is actually more strongly related to immigrant presence in a larger contextual unit (the department) than in a smaller unit (communes). Specifying the correct contextual unit and isolating its effect is a significant task for future research.

350

Immigration and the Transformation of Europe

explore variation across countries in the causal underpinnings of resistance to immigration, we conducted several initial analyses. One approach was to estimate the model for the pooled sample with the inclusion of country-specific dummy variables. The result, not reported here for reasons of space, indicated the presence of residual differences across countries after the effects of both the individual-level causes and the aggregate differences in the number of foreign-born and economic conditions are taken into account. A second approach was to estimate the model presented in Table 13.3 (absent the country-level variables) for each of the twenty countries in the ESS sample. By comparing individual-level effects across countries, one tests for the robustness of the hypotheses regarding interests, identities, and information and also indicates whether contextual conditions “prime” certain of these individual-level factors, such as nationalism or economic anxiety.21 The focus of this research is on the consistency and relative strength of cultural and economic factors in explaining opposition to immigration in Europe. The country-specific coefficients for the measure of cultural unity show that this variable is statistically significant in every country. Indeed, this variable is the most robust component of the model. However, the magnitude of its effect is not related to the percentage of non-EU foreigners in the country. The effects of preferences of national authority are also quite robust, as they are statistically significant in the majority of countries. However, there is only a very slight positive relationship with the percentage of non-EU foreigners. Thus, neither measure of cultural and national identities is “primed” by a putatively more visible immigrant “threat.”22 The role of sociotropic economic evaluations is generally more robust than personal evaluations: there are a greater number of coefficients among the 20 countries that are statistically distinguishable from zero. However, no measure of economic circumstances or anxiety has as consistent an impact as the preference for cultural unity. So it appears that “symbolic” orientations tend to be more consistent predictors of immigration attitudes than calculations of economic interest. We also find that economic evaluations are not primed by the economic situation in the country. Whereas one might think that evaluations more strongly predict attitudes towards immigration when the economy is struggling, we find no such evidence. Table 13.4 provides a summary of the effects of 21 22

For reasons of space we do not present a table of these twenty regressions, but they are available on request from the authors. The individual-level coefficients for perceptions of immigrants’ racial and socioeconomic background are also unrelated to the percentage of non-EU foreigners.

✓

✓ ✓ ✓ ✓

✓ ✓

✓ ✓ ✓

✓ ✓ ✓ ✓ ✓ ✓ ✓

✓

✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓

✓ ✓ ✓ ✓ ✓

✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓

Prefer cultural unity

✓

✓ ✓

✓ ✓

✓ ✓

Personal financial situation

✓ ✓ ✓ ✓

✓

✓ ✓

✓

✓

✓ ✓

National economic satisfaction

✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓

Social trust

✓ ✓ ✓

✓ ✓ ✓ ✓

✓ ✓ ✓

✓ ✓ ✓ ✓ ✓

Prefer national authority

✓

✓ ✓

✓ ✓

✓ ✓ ✓

✓

Ideology1

✓ ✓ ✓

✓ ✓ ✓ ✓

✓ ✓ ✓ ✓ ✓

✓ ✓ ✓ ✓ ✓

Education

A checkmark denotes that a particular variable was significant at p < .05 or better (one-tailed). The model estimated was identical to the first column of Table 3, the Preferred Level Index. 1 Ideology was counted as significant if it had a significant direct effect or a significant effect when interacted with frequency of political discussion.

Belgium Switz Czech Germany Denmark Spain Finland UK Greece Hungary Ireland Italy Luxemburg Netherlands Norway Poland Portugal Sweden Austria France

Have immigrant friends

Comparative estimate of immigration

Table 13.4 Statistical significance of selected variables in predicting preferred level of immigration in twenty European countries

352

Immigration and the Transformation of Europe

theoretically relevant variables on opposition to immigration across countries. The table makes clear the consistent significance of cultural identities and social trust and the more variable influence of economic circumstance and attitudes. These results leave us with several empirical questions and puzzles to explore in the future. A critical issue is to clarify further the role of contextual influences. Measures of context at sub-national geographic units may be one important step. It may also be that attitudes toward immigrants have become increasingly divorced from context as the issue has become politicized and people’s perceptions of immigration and immigrants rely more on elite messages and less on the objective circumstances around them. In this regard, political context as measured by laws regarding immigrant rights, the relative strength of left and right wing parties, and the nature of a country’s identity (e.g., ethnic versus civic) may be more significant environmental influences than the demographic and economic conditions that have been the main focus of research to date. In sum, the microfoundations of public opinion about immigration are fairly stable and consistent throughout European countries. At the individual level, one can conclude with some confidence that European attitudes toward immigration sit firmly on these four legs: cultural and national identities, anxiety about the state of the economy, membership in social groups and networks that communicate a particular outlook about accepting immigrants, and psychological dispositions that govern highly generalized emotional reactions to other people. What explains the relative weight of these factors across space and time requires additional analysis. Discussion Europe is not calling for other nations to send their huddled masses. Instead, a succinct summary of the modal opinion evident in most countries might be “send us a few, but make sure they think and look like us, and that they won’t take our jobs or make trouble.” For Europeans, the ideal immigrant is someone who works in jobs natives cannot or will not perform, pays taxes, stays home at night, speaks the common language, never breaks down, and never grows old. In short, the preferred target of a “soliciting” immigration policy is an easily programmed machine. In the real world, of course, a thoroughly self-regarding immigration policy probably is impossible for governments to implement, constrained as they are by permeable borders and international pressures to accept at least some of the people uprooted by war and persecution. Moreover, migration is largely a voluntary behavior. People leave their own

European immigration in the people’s court

353

countries to seek a better life and this limits a state’s capacity to solicit just the right kind of immigrant. Indeed, countries such as France, Germany, and Great Britain have to compete with each other, Canada, and the United States for highly qualified professionals in health care and information technology. Yet it is well-established that a migration flow, once begun, induces its own flow. As chapter 3 showed in broad numbers, chain migration is quickly institutionalized, with people from particular points of origin going to particular points of destination. The immigrant stream from Algeria goes to France and similar channels link Morocco and Spain, Pakistan and Bangladesh and England, and Turkey and Germany. The homogeneity of immigrants and the tendency for residential enclaves to develop both makes it difficult for government to have a “designer” immigration policy and probably retards the process of integration. Immigration now is forcing Europe to confront the challenge of multiculturalism; that is, to take seriously the popular belief that immigration poses a cultural threat to the nation’s identity and way of life. The cultural basis of opposition to immigration is a recurring theme in this and other recent studies of European attitudes. Publics are sensitive to the economic consequences of immigration, even if they are not personally affected and especially during periods of economic recession, but more important are deeply held symbolic attitudes, including nationalism, political ideology, and racial prejudice. In the United States, immigration goes to the heart of how Americans conceive of their unique national identity. Historically, immigration has played a completely different role in Europe, where most nations define themselves as ethnically homogeneous communities, immigrants are workers not settlers, and even if they seek “integration” the populace’s conception of its national self may foreclose fully accepting people of different cultural origins into what David Hollinger (1995) has dubbed the national “Circle of We.” Controlling immigration is a separate issue from the treatment of immigrants already admitted. The concept of “integration” is the prevalent term in European thinking to refer to what happens to immigrants after entry. Favell (2003) outlines various aspects of “integration” policy, such as state-sponsored measures addressing basic legal and social protection for immigrants, anti-discrimination laws, naturalization, language courses, changes in the education curriculum, funds for economic development and housing, and changes in patterns of social relations, intermarriage, and self-identification. Only now are we beginning to learn about popular opinion regarding immigrant integration as both goal and process. The evidence reported here indicates that linguistic assimilation is expected. And despite the

354

Immigration and the Transformation of Europe

declining significance of religion in most European countries and seeming opposition to a religious test for citizenship, there seems little doubt that the cultural as well as the political dimension of Islam increasingly is viewed as a threat to European attitudes toward women, family, and homosexuals. If to be European is to be post-modern and post-national, immigration may well threaten European as well as national identity. Indeed, the problem of redefining nationhood to accommodate cultural differences seems greater in Europe than in the United States, in part because so many immigrants to Europe are Muslims. This raises the question of whether nationalism in Europe truly is disappearing as globalization and political integration proceed apace. If so, then on the one hand cultural assimilation beyond learning the majority’s language becomes a less divisive issue, but on the other, the search for a cultural glue thick enough to bind a much more diverse society together becomes a critical political problem.  Adorno, Theodor W., Else Frenkel-Brunswik, Daniel J. Levinson and R. Nevitt Sanford 1950 The Authoritarian Personality. 1st edn. New York: Harper. Allport, G. W. 1954 The Nature of Prejudice. Oxford: Addison-Wesley. Anderson, Benedict R. O’G. 1983 Imagined Communities: Reflections on the Origin and Spread of Nationalism. London: Verso. Brubaker, Rogers 2001 “The Return of Assimilation? Changing Perspectives on Immigration and Its Sequels in France, Germany, and the United States.” Ethnic and Racial Studies 24(4): 531–48. Chong, Dennis 2000 Rational Lives: Norms and Values in Politics and Society. Chicago, IL: University of Chicago Press. Citrin, Jack, Donald P. Greene, Christopher Muste and Cara Wong 1997 “Public Opinion toward Immigration Reform: The Role of Economic Motivations.” Journal of Politics 59(3) (August): 858–81. Citrin, Jack and John Sides 2004 “More than Nationals: How Identity Choice Matters in the New Europe.” Chapter 8 in Transnational Identities: Becoming European in the EU, Richard K. Herrmann, Thomas Risse-Kappen and Marilynn B. Brewer (eds). London: Rowman and Littlefield. Esping-Andersen, Gøsta 1990 The Three Worlds of Welfare Capitalism. Cambridge: Polity Press. Esses, Victoria M., Lynne M. Jackson and Tamara L. Armstrong 1998 “Intergroup Competition and Attitudes toward Immigrants and Immigration: An Instrumental Model of Group Conflict.” Journal of Social Issues 54(4) (Winter): 699–724. Favell, Adrian 2003 “Integration Nations: The Nation-State and Research on Immigrants in Western Europe.” Comparative Social Research 22: 13–42.

European immigration in the people’s court

355

Fetzer, Joel 2000 Public Attitudes toward Immigration in the United States, France, and Germany. Cambridge: Cambridge University Press. Freeman, Gary P. 1995 “Modes of Immigration Politics in Liberal Democratic States.” International Migration Review 29(4) (Winter): 881–902. Gabel, Matthew 1998 “Public Support for European Integration: An Empirical Test of Five Theories.” Journal of Politics 60(2) (May): 333–54. Gellner, Ernest 1983 Nations and Nationalism. Ithaca, NY: Cornell University Press. Hagendoorn, A. and Shervin Nekuee (eds) 1999 Education and Racism: A Cross National Inventory of Positive Effects of Education on Ethnic Tolerance. Brookfield, VT: Ashgate. Hardin, Russell 1995 One for All: The Logic of Group Conflict. Princeton, NJ: Princeton University Press. Hollinger, David A. 1995 Postethnic America: Beyond Multiculturalism. New York: BasicBooks. Hood, M. V. III and Irwin L. Morris 1997 “¿Amigo o Enemigo?: Context, Attitudes, and Anglo Public Opinion toward Immigration.” Social Science Quarterly 78(2) (June): 309–23. Jackman, Robert W. and Karin Volpert 1996 “Conditions Favouring Parties of the Extreme Right in Western Europe.” British Journal of Political Science 26(4) (October): 501–21. Joppke, Christian 2002 “European Immigration Policies at the Crossroads.” Chapter 14 in Developments in West European Politics 2, Paul Heywood, Erik Jones and Martin Rhodes (eds). Basingstoke: Palgrave Macmillan, 259–75. Lahav, Gallya 2004 Immigration and Politics in the New Europe: Reinventing Borders. Cambridge: Cambridge University Press. Leiken, Robert S. 2005 “Europe’s Majahideen: Where Mass Imigration Meets Global Terrorism.” Washington, DC: Center for Immigration Studies, April. www.cis.org/articles/2005/back405.html. Leudtke, A. 2005 “European Integration, Public Opinion, and Immigration Policy: Testing the Impact of National Identity.” European Union Politics vol. 6, No. 1. Luke, Douglas A. 2004 Multilevel Modeling. Thousand Oaks, CA: Sage Publications. McLaren, Lauren 2001 “Immigration and the New Politics of Inclusion and Exclusion in the European Union.” European Journal of Political Research 39: 81–108. Money, Jeannette 1999 Fences and Neighbors: The Political Geography of Immigration Control. Ithaca, NY: Cornell University Press. OECD 2004 Trends in International Migration SOPEMI 2003. Paris: OECD. Perrineau, Pascal 1985 “Le Front National: un e´ lectorat autoritaire.” Revue Politique et Parlementaire 87: 24–31. Pettigrew, Thomas F. 1998 “Reactions toward the New Minorities of Western Europe.” Annual Review of Sociology 24: 77–103. Pettigrew, Thomas F. and R. W. Meertens 1995 “Subtle and Blatant Prejudice in Western Europe.” European Journal of Social Psychology 25(1) (January– February): 57–75.

356

Immigration and the Transformation of Europe

Putnam, Robert D. 2000 Bowling Alone: The Collapse and Revival of American Community. New York: Simon & Schuster. Quillian, Lincoln 1995 “Prejudice as a Response to Perceived Group Threat: Population Composition and Anti-Immigrant and Racial Prejudice in Europe.” American Sociological Review 60(4) (August): 586–611. Raudenbush, Stephen W. and Anthony S. Bryk 2002 Hierarchical Linear Models: Applications and Data Analysis Methods. 2nd edn. Thousand Oaks, CA: Sage Publications. Scheve, Kenneth F. and Matthew Slaughter 2001 “Labor Market Competition and Individual Preferences over Migration Policy.” Review of Economics and Statistics 83(1) (February): 133–45. Sears, David O. 1993 “Symbolic Politics: A Socio-Psychological Theory,” in Explorations in Political Psychology, Shanto Iyengar and William James McGuire (eds). Durham, NC: Duke University Press, 113–49. Sniderman, Paul M., Louk Hagendoorn and Markus Prior 2004 “Predisposing Factors and Situational Triggers: Exclusionary Reactions to Immigrant Minorities.” American Political Science Review 98(1): 35–49. Steenbergen, Marco R. and Bradford S. Jones 2002 “Modeling Multilevel Data Structures.” American Journal of Political Science 46 (1) (January): 218–37. Torpey, John 2000 “States and the Regulation of Migration in the Twenty-first Century North Atlantic World.” Chapter 3 in The Wall Around the West: State Borders and Immigration Controls in North America and Europe, Peter Andreas and Timothy Snyder (eds). Lanham, MD: Rowman and Littlefield, 31–54. United Nations Population Division 2000 “Replacement Migration: Is it a Solution to Declining and Ageing Populations?” ESA/P/WP.160. New York: United Nations, March 21. www.un.org/esa/population/publications/ migration/migration.htm. Zaller, John 1992 The Nature and Origins of Mass Opinion. Cambridge: Cambridge University Press.

European immigration in the people’s court

357

Appendix Table 13.A Country-level data

Country

% foreign % % (nonasylum- GDP foreign EU) seeker 2002

Belgium Switz Czech Germany Denmark Spain Finland UK Greece Hungary Ireland Italy Lux Neth Norway Poland Portugal Sweden Austria France

10.68 25.26 3.59 19.03 7.19 5.32 2.63 8.28 10.27 2.87 10.37 2.40 32.95 10.10 7.33 3.55 6.29 12.01 12.49 10.03

4.69 13.65 3.29 17.35 4.94 3.71 1.73 5.19 8.17 2.62 2.14 2.04 6.30 7.85 4.49 3.11 4.48 7.49 9.72 6.61

2.15 3.45 0.55 1.49 1.69 0.19 0.35 1.11 0.26 0.45 1.31 0.17 1.82 2.22 1.84 0.08 0.04 1.98 1.83 0.51

Unemp GDP 2002 (%)

26.05 7.30 29.84 3.20 14.06 7.30 24.99 8.60 28.66 4.60 21.00 11.30 26.09 9.10 26.03 5.10 17.32 10.00 12.82 5.60 30.59 4.30 25.11 9.00 49.59 2.80 27.14 2.70 36.94 3.90 10.53 19.80 17.26 5.10 27.23 4.90 28.31 4.30 25.85 8.80

0.22 −0.88 2.14 −0.09 0.66 1.32 1.98 1.31 3.39 3.66 4.13 0.19 1.37 −0.08 0.79 1.40 −0.35 1.74 1.08 0.65

 Unemp Econ (%) Index 8.22 18.75 −9.59 9.30 6.52 6.19 0.00 1.96 −4.00 0.00 9.30 −4.44 25.00 7.41 7.69 6.57 19.61 0.00 16.28 3.41

−0.24 −1.11 0.98 −0.21 −0.35 0.42 0.44 0.01 1.13 0.91 0.14 0.26 −1.46 −0.59 −0.64 1.26 −0.56 0.10 −0.56 0.08

Percent foreign-born and percent foreign-born (Non-EU) The raw data are from the OECD: www.oecd.org/document/51/0,2340,en 2825 494553 34063091 1 1 1 1,00.html This file does not contain data on Italy, which were derived from another OECD publication (Table B.1.5 in OECD 2004: 341–50). Historical data used to compute changes in immigrant levels over time also come from OECD (2004). Non-EU immigrants are defined as those from outside of Western Europe, the US, Canada, New Zealand, Australia, and Iceland. The data derive from country census data collected in 2000 or 2001, except for France (1999) and Ireland (2002). Percent asylum-seekers See Table I.4 (OECD 2004: 34) and Table A.1.3 (OECD 2004: 306). The measure is the total number of asylum-seekers 1993–2002, divided by the country’s 2002 population. Unemployment rate See www.oecd.org/dataoecd/41/13/18595359.pdf. GDP The specific figure is GDP per capita (in 2002 US $1000), based on purchasing power parities. See www.oecd.org/document/28/0,2340,en 2825 495684 2750044 1 1 1 1,00.html.

358

Immigration and the Transformation of Europe

Table 13.B Individual-level questions and measures Variable or concept

ESS variable

Question wording and coding

Preferred level of imsetn immigration imdfetn eimrcnt eimpcnt imrcntr impcntr

“Now, to what extent do you think [country] should allow people of the same race or ethnic group as most [country] people to come and live here?” “How about people of a different race or ethnic group from most [country] people?” “Now, to what extent do you think [country] should allow people from the richer countries in Europe to come and live here?” “And how about people from the poorer countries in Europe?” “To what extent do you think [country] should allow people from the richer countries outside Europe to come and live here?” “How about people from the poorer countries outside Europe?” Response options were: many, some, a few, or none. Each indicator was coded 0–many to 1–none and average to make the index.

Perceived effects of immigration

imtcjob imbgeco imeuclt imwbcrm imbleco imwbcnt

“Would you say that people who come to live here generally take jobs away from workers in [country], or generally help to create new jobs?” “Most people who come to live here work and pay taxes. They also use health and welfare services. On balance, do you think people who come here take out more than they put in or put in more than they take out?” “Would you say it is generally bad or good for [country]’s economy that people come to live here from other countries?” “Would you say that [country]’s cultural life is generally undermined or enriched by people coming to live here from other countries?” “Are [country]’s crime problems made worse or better by people coming to live here from other countries?” Is [country] made a worse or a better place to live by people coming to live here from other countries? All recoded from original 0–10 scale to 0 (most positive view) to 1 (most negative view). Index averages six items.

Perceived race of immigrants

imgetn

“Thinking of people coming to live in [country] nowadays from other countries, would you say that most are of the same race or ethnic group as the majority of [country] people, most are of a different race or ethnic group, or is it about half and half?” Coded 0–most of same race, .5–half and half, 1–most of different race.

European immigration in the people’s court

359

Table 13.B (cont.) Variable or concept

ESS variable

Question wording and coding

Perceived SES of eimgrpc EU immigrants

“Now thinking about people coming to live in [country] nowadays from other countries within Europe, would you say that most come from the richer countries of Europe, most come from the poorer countries of Europe, or is it about half and half?” Coded 0–most from richer EU countries, .5–half and half, 1–most from poorer EU countries.

Perceived SES of imgrpc non-EU immigrants

“And what about people who come to live in [country] nowadays from countries outside Europe, would you say that most come from the richer countries outside Europe, most come from the poorer countries outside Europe, or is it about half and half?” Coded 0–most from richer non-EU countries, .5–half and half, 1–most from poorer non-EU countries.

Comparative estimate of immigration

cpimpop

“Compared to other European countries of about the same size as [country], do you think that more or fewer people come and live here from other countries?” Coded 0–far fewer to 1–far more.

Have immigrant friends

imgfrnd

“Do you have any friends who have come to live in [country] from another country?” Coded 0–none, .5–few, 1–several.

Prefer cultural unity

pplstrd

“It is better for a country if almost everyone shares the same customs and traditions.” Coded 0–strongly disagree to 1–strongly agree.

Personal financial uncertainty

hincfel brwmny

Which of the descriptions on this card comes closest to how you feel about your household’s income nowadays: living comfortably on present income, coping on present income, finding it difficult on present income, or finding it very difficult on present income?” “If for some reason you were in serious financial difficulties and had to borrow money to make ends meet, how difficult or easy would that be: very difficult, quite difficult, neither easy nor difficult, quite easy, or very easy?” Each indicator was recoded to 0–1 scale and then average to form variable. Coded from 0–most certain to 1–most uncertain.

Satisfaction with economy

Sfteco

“On the whole how satisfied are you with the present state of the economy in [country]?” Recoded 0–10 scale to 0–1, where 1 indicates most satisfied. (cont.)

360

Immigration and the Transformation of Europe

Table 13.B (cont.) Variable or concept

ESS variable

Income

Hinctnt

Employment status: unemployed student retired, homemaker, other

Question wording and coding

Coded on 0–1 scale, with missing data recoded to country mean. pdwrk And which of these descriptions describes your situation edctn (in the last seven days)? In paid work (or away uempla temporarily) (employee, self-employed, working for your uempli rtrd family business); in education, even if on vacation (not cmsrv paid for by employer); unemployed and actively looking hswrk for a job; unemployed, wanting a job but not actively dngoth looking for a job; permanently sick or disabled; retired; in community or military service; doing housework, looking after children or other persons; or other?” “Unemployed” is coded 1 if respondent is unemployed (whether looking for work or not) and 0 otherwise. “Student” is coded 1 if edctn=1 and 0 otherwise. “Retired, homemaker, etc.” is coded 1 if rtrd, cmsrv, hswrk, or dngoth=1, and 0 otherwise.

Social trust

ppltrust pplfair pplhlp

“Generally speaking, would you say that most people can be trusted, or that you can’t be too careful in dealing with people?” “Do you think that most people would try to take advantage of you if they got the chance, or would they try to be fair?” “Would you say that most of the time people try to be helpful or that they are mostly looking out for themselves?” Variable averages together three items and ranges from 0 (least trusting) to 1 (most trusting).

Life satisfaction

stflife

“All things considered, how satisfied are you with your life as a whole nowadays?” Recoded 0–10 scale to 0–1, where 1 indicates most satisfied.

Political discussion

discpol

“How often would you say you discuss politics and current affairs: every day, several times a week, once a week, several times a month, once a month, less often, or never?” Coded 0–never to 1–every day.

Conservatism

lrscale

“In politics people sometimes talk of ‘left’ and ‘right.’ Using this card, where would you place yourself on this scale, where 0 means the left and 10 means the right?” Coded 0–left to 1–right.

European immigration in the people’s court

361

Table 13.B (cont.) Variable or concept

ESS variable

Preference for supranational authority

dclenv dclcrm dclagr dcldef dcldef dclwlfr dclaid dclmig dlcintr

“Policies can be decided at different levels. Using this card, at which level do you think the following policies should mainly be decided: protecting the environment, fighting against organised crime, agriculture, defence, social welfare, aid to developing countries, immigration and refugees, interest rates.” Each item was coded 0–national or local/regional level and 1–European or international level. These items were then summed up to make an index from 0 to 8, where higher values indicate a willingness to delegate more policy reams to a supranational authority and this index was converted to the 0–1 scale.

Self-identified minority

blgetmg

“Do you belong to a minority ethnic group in [country]?” Coded 0–no and 1–yes.

Question wording and coding

Immigrant status brncntr facntr mocntr ctzcntr livencntr

Each variable is coded 0 or 1. “Second generation” means that R is born in country but father and mother are not. The remaining variables all involve foreign-born Rs and are based on whether R is naturalized and how long R lived in country.

Education

edulvl

“What is the highest level of education you have achieved? No qualifications; CSE grade 2–5/GCSE grades D-G or equivalent; CSE grade 1/O-level/GCSE grades A-C or equivalent; A-level, AS-level or equivalent; Degree–postgraduate qualification or equivalent.” (Note: combined measure based on country-specific codes.) Five-category item recoded 0–least educated to 1–most educated.

Age

yrborn

Female

gndr

Coded as 2002 minus value in yrbrn. Three cases with age < 14 were dropped. Values greater than 98 were coded as 98. Coded 1–female and 0–male.

14

The politics of immigration in France, Britain, and the United States: A transatlantic comparison Martin A. Schain Introduction: The problem

Both Europe and the United States are “countries” of immigration. Each year about 1.5 million immigrants legally enter the countries that comprise the European Union (the EU-15), with considerable variation among countries. A generation ago, the most important differences within Europe were between countries that had historically needed and received immigrants (France, Germany, Switzerland, and the United Kingdom), and those that had been the providers of immigrants (Italy, Spain, Ireland, and Portugal). Now, however, all of the senders are receivers, and the variation is among the levels of immigration. Indeed, Europe now receives between 4.7 immigrants per thousand population on the high end (1992) – 3.9 on the low end (2001), compared with about 3.8 per thousand in the United States (SOPEMI 2004: 305–10). The most important differences between Europe and the United States are not those of levels of immigration, but differences in the politics of immigration: immigration policy and the dynamics that drive this policy. The United States has a relatively open immigration policy, one that sets a (flexible) ceiling on the number of legal immigrants admitted each year, the basis on which they are to be admitted, and the principle criteria that govern admission. Although the ceilings set on immigration never exactly correspond to the actual number of immigrants admitted to the United States each year, there is a general relationship between the intent of the law (to admit and limit immigrants) and the results. In Europe, the relationship between the stated intent of the law and the results is quite different. Levels of legal immigration have remained high in Europe over a long period of time, even after the strong declarations of “zero” immigration policies at the governmental level. Although the main objectives of policy have not changed over a 25-year period, large numbers of legal immigrants – workers, as well as family members through family reunification – have been admitted to all European 362

A transatlantic comparison

363

countries, mostly through administrative decisions that have facilitated labor migration, and court decisions that have established access rights for family members of settled immigrants. Therefore, the first question raised in this paper is how we can understand the differences in policy – as well as the relationship between policy and policy outcomes – in Europe compared with the United States. At least superficially, this difference between a relatively open policy in the United States and restrictive European policy orientation seems logical. After all, the United States is a “country of immigration,” a “nation of nations,” which has long accepted the multicultural diversity and challenge to national identity implied by continuing immigration. European countries, on the other hand, which have a deeper cultural sense of national identity and are far more homogeneos than the United States, have a weaker capacity to deal with high levels of immigration. However, this kind of understanding simply ignores the fact that in both Europe and the United States there have been periods of open and exclusionary policies during the past 100 years. At least part of the explanation for the differences in policy flows from the dominant dynamics that are driving policy on either side of the Atlantic. With some variation among countries, the dominant concern that drives policy in Europe is the challenge to national identity that is posed by immigration from countries outside of Europe. These concerns about national identity are evident in public opinion surveys, as documented in chapter 13, and have been for many years. The immigration issue, defined in these terms, has been politicized by political parties, and has become integrated into party competition at the national and subnational levels. This electoral dynamic has been a key to the development of policy and its implementation throughout the European Union (EU). Defined as a challenge to national identity during the period since the 1960s, the resultant exclusionary policy in Western Europe has resisted the challenge of labor market and welfare state needs that have developed in more recent years. In the United States, questions of national identity that were at the core of the development of immigration policy at the turn of the last century have played a far less important role in the development of more recent policy. Instead, what has increasingly dominated the politics of immigration have been considerations of electoral gain or future party gain as a result of immigrants voting. Defined in relation to a multi-cultural society after 1965, immigration policy has remained relatively open through good and bad economic times. The core of this policy has also resisted nationalist challenges in the 1990s and the more recent challenge of postSeptember 11 security concerns. Overall, although policy on both sides

364

Immigration and the Transformation of Europe

of the Atlantic has been driven by electoral considerations, in Europe the focus has been on the presumed impact of immigration on voters other than immigrants, while in the United States, increasing attention has been given to immigrants themselves as voters and potential voters. In this chapter, I analyze these questions by focusing on a comparison between the United States and two European countries: France and Britain. I argue that two aspects of policy development are of particular importance for understanding differences in policy, as well as the complexities of developing an immigration policy for the EU. How questions of immigration are defined and constructed This implies the larger issue of issue formation or agenda-setting. Since relatively few issues make it on to the policy agenda, a process through which immigration is presented and defined is related to how it becomes a political priority. The way that these issues are defined by public authorities is a crucial aspect of policy-making that is also linked to which publics are mobilized and within which political arenas policy decisions are taken. The construction of the issue of immigration may be related to pressures of public opinion, to pressures of organized interests, to initiatives within administration, or to all three. The point is that issues do not simply emerge. They are constructed within specific institutional arenas in specific ways for specific purposes. In his recently published study of race policies in Britain and France, Erik Bleich (2003) explains their sharp differences of ideas in the form of “frames” that drove the dominant group of policy-makers in each case. While many differences in policy outcomes can be explained by conflict theory (groups and parties), and differences in the specific content of policy can be understood through a problem-solving approach that focuses on the role of policy communities or by the institutionalist perspectives, Bleich sees these as secondary, if what we want to understand is differences in the orientation of policy choices that are made. He argues that the same kinds of political actors in each country can produce very different kinds of policies if they are operating with very different sets of ideas about the need for policy, its goals and the ways it should be effective (Bleich 2003, 172–4). This approach forms the core of our understanding of the approach to immigration policy in France, Britain, and the United States. However, we find that the ideas that structure the policy orientation are informed by strategic thinking about politics which is related to the distribution of immigrant communities within the country.

A transatlantic comparison

365

How political space and arenas of decision-making influence the policy-making process and its outcomes The arenas within which issues are developed provide first, a key to how the issues are defined, and second, a key to the dynamics through which they develop further. The importance and impact of immigration can, and generally does, vary considerably across geographic space. It seems reasonable to presume that public opinion, interest pressure, and electoral pressure would vary with the variation in the concentration of immigrant populations, since the concentrations of immigrants across spatial areas are also related to political, social, and economic costs and benefits (see Freeman 1989 and Money 1999, ch 1). Thus what may appear to be “national” opinion may be strong where there are high concentrations of immigrants and indifferent where there are not. The electoral arena is, in fact, a network of local constituency arenas, each electing its own representative, and each influenced by local political forces. Although immigration policy and regulation is usually determined in the national rather than the local political arena, it is crucial to understand the conditions under which local demands are transmitted to the national level. The link between local dynamics and the national arena depends on the importance of specific localities for shifting national elections. Jeanette Money has argued convincingly that the politicization of immigration has been driven by electoral dynamics within spatially defined arenas – areas in which immigrants are concentrated. In Money’s analysis, the size and safety of the constituencies (possible swing constituencies) are important, as is the importance of the constituency for building national coalitions. Even where size is irrelevant (as in Britain), the number of constituencies in which immigration issues are important must be related to the winning party’s electoral margin in the House of Commons. Cross-nationally, the common thread is the need for politicians to build a national electoral majority. Driven by electoral competition, local politicians will shift their policy positions in response to changing community preferences, toward either greater openness or greater closure. But immigration control is determined in the national rather than the local political arena. Therefore, it is crucial to understand the conditions under which local demands are successfully transmitted to the national level. (Money 1999: 62.)

Similarly, each institutional arena within which decisions are made and policy developed, structures decisions in a different way. It has long been recognized that the network of actors and influence within an institutional

366

Immigration and the Transformation of Europe

framework strongly influence policy outcomes. Indeed, the struggle over the arena itself may be an important key for understanding the ultimate outcome (the concept of power was first developed by Schattschneider [1960]; Lowi [1969, chapter 1] then applied it more extensively). Each institutional or geographic arena provides a different opportunity structure for political actors, but within that structure, a variety of outcomes are possible. Defining and constructing the immigration issue France In many ways France is the most normal case – the most consistent with expectations. The French government moved to restrict immigration in 1974, in the midst of the oil crisis and at the beginning of what would be the end of the period of postwar expansion, which had been the stated reason for labor immigration in earlier decades. The sharp move towards immigration restriction during this period was directed specifically against immigration from outside of Europe, however, and for some time immigration policy had been developed by administrative authorities with the presumption that non-European immigration (for the most part from former French colonies) was quite different from immigration from Europe: Must we recall that migrations of European origin were understood as migrations for work and settlement, while those from the former colonies of North Africa were considered, at least until the recognition of family reunification, as migrations of “celibate” workers? (Viet 1998: 509)

Indeed, part of what went into the decision to suspend immigration in 1974 was the long administrative dialog that differentiated between European and North African immigration. This dialog, in turn, was at the root of the tension that underpinned the policy-making process. It was certainly not anticipated that one consequence of the suspension of immigration from the former colonies would be: That France [would integrate] communities organized according to their own rules, and that the rigidity of the republican model of integration [would be] distorted by forms of representation, of institutionalization, of negotiation and mediation that betrays as much the unspeakable recognition of community and cultural forms as the search for a modus vivendi or a new social contract based on the recognition of the possible limits of differences. (Viet 1998: 509)

There were solid labor market reasons for suspending immigration in 1974, but there were also other fears that reflected patterns of

A transatlantic comparison

367

public opinion. The first attempts by the French state to define a coherent policy about the new immigration began after the May crisis of 1968, and is summarized in a report written by Correntin Calvez for the Economic and Social Council in 1969. The report recognized the continuing economic need for immigrant labor, but for the first time clearly differentiated European from non-European workers. Europeans were assimilable, and should be encouraged to become French citizens, argued Calvez, while non-European immigrants constituted an “inassimilable island”: It seems desirable, therefore, more and more to give to the influx of non-European origin, and principally to the flow from the Maghreb, the character of temporary immigration for work, organized in the manner of a rapid process of introduction which would be linked as much as possible to the need for labor or the business sectors concerned and in cooperation with the country of origin. (Calvez 1969: 315)

Thus, from the beginning of the process of defining and implementing immigration policy, the idea of difference was asserted, a difference that was frequently posed in (ethno-cultural) racialized terms. The new forms of racial differentiation expressed by the Calvez report had little to do with eugenics or the biological inferiority of the new immigrants. Rather it was an expression of a perceived chasm between immigrant and French culture, what French philosopher Pierre-Andr´e Taguieff (1988) has called “differential racism.” During the 1970s, the government struggled to develop a policy informed by the main lines of the Calvez Report, but was unsuccessful. The Left opposition was generally successful in checking and limiting what they sometimes termed the “racist” labor market policies of the governments of the Right in the 1970s at the national level. At the same time, however, representatives of the Left were defining issues and developing integration policies at the local level that were based on similar racial assumptions. In contrast to “the tradition of solidarity” that Communist-governed municipalities had developed towards predominantly European immigrants, by the 1970s many of these same local governments began to treat non-European immigrants, as well as non-white French citizens from the overseas departments, as temporary residents who must be encouraged to return home (Schain 1985). By the 1980s the framing of the issue of immigration in racialized terms had taken on positive as well as negative aspects at the national level. The government of the Left was persuaded that non-European immigrants would not soon depart, and therefore the state needed to deal more decisively with problems of integration. At the same time, however,

368

Immigration and the Transformation of Europe

the government retained an approach that set these immigrants apart as a special case. As Gary Freeman has noted, In a sense, once the state had committed itself to this racially discriminatory policy, it had more incentive than before to increase the effectiveness and generosity of its social policies towards migrants. The fact that a large part of its immigrant population would be permanent persuaded officials that more needed to be done on their behalf. (Freeman 1989: 169)

After the suspension of immigration in 1974, an increasingly dense network of state institutions and programs were developed to deal with problems of incorporation and integration. The first initiatives in the 1970s were taken in the areas of education and housing. While the objectives of these programs were often contradictory, they were all efforts that tended to treat immigrant groups as collectivities rather than individuals. Two trends assured that the framing of the immigrant issue in terms of national identity and/or ethnic danger to the French nation would endure. The first was the rapid decline of the Communist Party that first became evident in 1981, but in many respects had begun in the 1970s. Although the Communists had tried to exploit anti-immigrant sentiments during the presidential election campaign of 1980–81, the party organization had also defended immigrant rights at the local level, and effectively mobilized them as voters, and potential voters, at election time. As the party organization declined during the 1980s, it was not replaced by an effective Socialist machine. This meant that even when the Communists moved away from their position of 1980–81, the Left in general would be less effective in reframing the issue. The second was the rise of the anti-immigrant National Front in the mid-1980s, often in the very areas in which the Communists were declining. This assured that the racialized approach to immigration would become a more permanent part of the discourse of French electoral politics. Thus, a broad range of policy-making elites participated in developing a frame of reference for policy on immigration that focused on questions of identity, and that rested on assumptions about non-European immigration that militated against easy integration. Britain In many ways the British case is most interesting in this context. After 1905, British law differentiated between aliens and Empire/ Commonwealth citizens. Although alien entry was restricted, the entry of Empire/Commonwealth citizens was not. Like the United States, but

A transatlantic comparison

369

unlike France, this immigration consisted largely of families, and therefore had an impact by the 1960s well beyond the labor market. The first moves towards immigration restriction were taken during a period of widespread (if declining) need for immigrant labor, well before the economic crisis of the 1970s, and well before restrictive actions of other countries in Western Europe. Beginning in 1961, first the Conservatives, then in 1964 Labour, openly favored the immigration restriction of non-white Commonwealth immigrants (although the 1962 legislation applied to all Commonwealth immigrants). Although Labour opposed the Commonwealth Immigrants Act of 1962, it actually extended restriction when it came to power in 1964, and then passed even more restrictive legislation in 1968. Subsequent legislation between 1971 and 1992 served to break down the distinction between aliens and Commonwealth citizens, and the acceptance of the “patrial” standard sharply restricted entry of non-white (New Commonwealth) citizens into the United Kingdom (the first restrictive legislation passed in 1962 is analyzed in Hansen 2000; a good summary of the development of legislation can be found in Geddes 2003, chapter 2). There is a vast literature on immigration control by both United Kingdom and United States scholars that relates public opinion to the restrictionist policies first initiated by governments in the 1960s. Most of this literature agrees that restrictionist policies were first introduced in response to rising public opinion against non-white Commonwealth immigration. Indeed, the key element that is cited is the reaction to the riots in the Notting Hill area of London and Nottingham in 1958. Yet, as Money points out, there had been race riots in Liverpool in 1948, as well as Birmingham and Deptford (London) in 1954, without a public policy reaction (Money 1999: 98). There is now a body of literature on immigration control in Britain that goes one step further, and argues, either explicitly or implicitly, that public opinion on this issue was manipulated by racist political elites into opposing immigration. This opposition, in turn, was then used to justify the imposition of increasingly restrictive immigration controls (Paul 1997; Spencer 1997; Miles and Phizacklea 1984, especially chapter 1; Carter, Harris, and Joshi 1987. Much of this literature is summarized in Hansen 2000: 10–16). In his study of British immigration policy, Randall Hansen argues persuasively that public opinion appears to have needed little manipulation in its opposition to immigration and its support for restrictive legislation. Indeed, as in France and the United States, political elites who favored immigration could draw little comfort from low levels of mass support, and there is no indication that variations in

370

Immigration and the Transformation of Europe

policy were driven one way or the other by public opinion (Hansen 2000: 14). At least until the early 1960s, there was considerable support among business and political elites for increasing, not decreasing, labor market immigration. As late as 1965, the government predicted labor shortages of two hundred thousand annually. Declining industries, especially the northern textile companies, relied on immigrant labor maintain their competitive advantage. And the British government itself was actively involved in recruiting immigrant labor for the London Transport and the National Health Service. The British Hotels and Restaurants Association also actively enlisted Commonwealth immigrants. The newly arrived immigrants found housing in inner city areas that were vacated when the native workforce moved to more desirable suburban locations. (Money 1999: 80)

On the other hand, among policy-making elites, there was a rising tendency to frame the immigration issue in terms of identity and race. As in France, what most concerned policy-makers well before the initiation of restrictions were racial tensions and the perceived problems of integrating non-white immigrant groups. Although there was clear conflict within each political party, even before 1962, about the need for immigration restriction from New Commonwealth countries, the core of the dialog in each case turned on racial differences and race relations. The Tories, who had previously argued that the experience in the Commonwealth demonstrated the viability of race relations, argued in 1965 that they “ . . . reject the multi-racial state not because we are superior to our Commonwealth partners but because we want to maintain the kind of Britain we know and love” (Foot 1965: 124, as cited in Money 1999: 74). Labour, which first challenged the 1962 legislation because, as Gaitskell said, it carried racial overtones, then linked their support for even tighter immigration restriction with support for the Race Relations Act of 1965 (Hansen 2000: 136–41). This frame of the immigration issue certainly reflected public opinion, but more importantly, it defined the way that the issue would be politicized. Although the dialog around third world immigration in Britain among political elites was similar to the dialog in France, and the framing of the issue was similar, there were some important differences. In Britain, the issue involved the re-definition of citizenship and the reduction of rights of Commonwealth citizenship. In Britain, this debate took place within and between political parties while in France, the initial discussion took place within the administration. On the other hand, the racialization of the immigration issue in both countries, and its framing in terms of a challenge to identity, was based far more on anticipated problems of

A transatlantic comparison

371

integration, rather than on existing problems. However, this kind of frame sharpened the perceived differences between immigrants and natives – and even other immigrants. The United States The definition of the political issue of immigration in the United States has been deeply linked to a complex history of racism, and appears to have little to do with labor market needs. Until the mid-1890s, the focus of discussion among political elites had been almost entirely on limiting the entry of categories of undesirable immigrants. The major exception, however, was Chinese exclusion. A decade-long racist campaign in California – led by the early labor movement in the 1870s – resulted first in the success of the Democratic Party in the state, and then in the successful passage of the Chinese Exclusion Act of 1882. The campaign also served as the impetus for the establishment of the AFL (American Federation of Labor) as a national trade union organization (see Mink 1986, chapter 3). Nevertheless, the well-organized Chinese community learned to use the court system to prevent strict enforcement of the Chinese Exclusion Act see Salyer 1989; (my thanks to Professor Martin Shapiro for this reference). The racial arguments that were developed for this campaign in California and then used in Washington, most unambiguously by the Democrats, were then broadly extended and given a scientific base by Republican restrictionists in the 1890s. As extended to apply to the new immigrants from eastern and southern Europe, these racial frames defined desirable immigrants as those from northern Europe. Discussions of immigration were focused on the problem of assimilation of Europeans, and on an effort to define the content of American citizenship, and both were tied to more highly focused scientific discussions of race based on eugenics, and on a new nationalism based on racial and religious type. The impetus for the redefinition of the immigration issue in nationalist and racial terms, came from congressional leaders and intellectuals. After Chinese exclusion in 1882, the first indication of a more general change in approach towards immigration during the 1890s came with the emergence of a movement to impose literacy tests on new immigrants. The movement was organized and led by the Immigration Restriction League, founded by Boston intellectuals in 1892, and for which Senators William E. Chandler and Henry Cabot Lodge were the chief spokesmen. Chandler was the chairman of the Senate Immigration Committee, which had been established in 1889. Literacy tests, argued Chandler, were the most effective means of restricting the entry of certain races, alien to

372

Immigration and the Transformation of Europe

American nationality. “No one,” he said in 1892, “has suggested a race distinction. We are confronted by the fact, however, that the poorest immigrants do come from certain races” (Mink 1986: 101). The 1890s marked the beginning of a debate on the relationship between “Americanism” and immigration that would last until the imposition of strict immigration restriction in 1924. In many ways, this debate would continue through the cold war, and would endure into the present. The political debate focused on national integration, race, and the danger of “alien races” (Henry Cabot Lodge’s phrase) for American democratic institutions. The framing of the core problem of European immigration as “racial” in content is evident in the Senate report that accompanied the Immigration Act of 1891, and was a particular concern of patrician advocates of restriction in the Immigration Restriction League linked to Congress. The illiteracy test will affect almost entirely those races whose immigration to the United States has begun within recent times and which are most alien in language and origin to the people who founded the 13 colonies and have built up the United States. . . .

Racial concerns made it possible for Republican restrictionists – based almost entirely in the Boston area – to distance themselves from the business orientation of their party at the turn of the century, and finally to link their political agenda to that of southern Democrats, who, with some reluctance, accepted this frame of the new immigration. This definition of the problem of immigration from Europe did not become widespread beyond Congress until the turn of the century, when patrician concerns about race, supported by new areas of scientific enquiry, gave form to growing popular nativism. Henry Cabot Lodge, the most important congressional leader of the political movement against the new immigration before World War I, made considerable efforts to build his case with scientific support for racial differences (Lodge 1891, cited in Higham 1963). Lodge’s thinking (and writing), in fact, was part of an ongoing dialog that included at least one university president (Francis Walker of MIT), one of the leading sociologist of the day (Franklin Giddings, the first professor of sociology at Columbia University), and, ultimately, the president of the United States, Theodore Roosevelt. In a wide-ranging debate on the ability of the United States to absorb the new tide of immigrants, the focus was increasingly on race and racial superiority or inferiority. Walker argued that the superior Anglo-Saxon racial strains were being overwhelmed by inferior European strains, in part because of declining native birth-rates, while Giddings contended that we had nothing

A transatlantic comparison

373

to worry about, since the superior Anglo-Saxon races would dominate here as they dominated there. (Giddings was the authority of an important sociology text first published in 1896. The text devoted relatively little space to questions of race, but where it did, it reflected the author’s somewhat ambivalent social Darwinism that emphasized the survival and dominance of “superior” white races. See Degler 1991: 17–18.) Roosevelt, while rejecting the case for restriction, concluded that we may have something to worry about. He then initiated a campaign for more children (and against birth control) to prevent “race suicide.” After analyzing the reaction to the campaign in the popular press, the historian John Higham concluded that: In the end, the whole discussion probably caused more race-thinking than reproduction. At least it brought to a wider audience the racial pessimism previously confined to a limited group of upper-class intellectuals. (Higham 1963)

By the time of World War I, race thinking had become well-established among virtually all political actors, and became integrated into political thinking through the actions and writing of government agencies. By the 1960s, forty years later, this definition of the problem of immigration had disappeared. We were a “nation of nations,” argued President Kennedy, in a famous speech in 1962. In fact, the national quota basis of the legislation had been somewhat undermined between 1945 and 1965 by the exigencies of the cold war. Nevertheless, the legislation could not be changed without challenging the assumptions behind it. The change in thinking developed through a process which began with ethnic organization, by the recognition of the legitimacy of a multiethnic America that was portrayed by government propaganda during World War II, and that was reinforced by the emergence of what Martin Kilson has called “Black neo-ethnicity” in the 1960s (1975, chapter 8). Indeed, the ideal of neo-ethnicity, a “nation of nations,” began to emerge at the same time that intermarriage among the children and grandchildren of European immigrants was sharply on the rise, and when important indicators of ethnic “memberships” were on the decline (organizational membership and language ability above all). Government programs in the 1960s that effectively “created” minorities “ . . . by ascribing to them certain characteristics that serve to justify their assignment to particular societal roles,” represented an attempt to deal with a racial crisis, not immigration or assertions of multiculturalism, but had unanticipated responses (see Glazer 1954: 158–73; Ireland 1994: 10–11). Within a decade, the impact of “re-ethnicization” had effectively challenged the melting pot understanding among policy-making elites. By 1965, the racial criteria

374

Immigration and the Transformation of Europe

of the early part of the twentieth century no longer had support among any sector of the political elite, although the quota system continued to be defended within the judiciary committees in terms of cultural assimilation (King 2000: 237; on the new ethnicity and multiculturalism see chapter 9). Thus, at a time when political leaders in France and Britain were developing race-based criteria as a frame for restrictive legislation, their counterparts in the United States were abandoning these criteria as a basis of exclusion, and developing a multicultural frame more consistent with the civil rights revolution. Moreover, at a time when their European counterparts were attempting to sharply reduce immigration, they were seeking a more “rational” basis (a better frame) for continuing legal entry into the United States. Efforts to reform the system were first pursued during the Truman administration, continued under Eisenhower and Kennedy, and finally implemented under Johnson, framed by both the Civil Rights Act of 1964 and the Voting Right Act of 1965. In practical terms, the system had been undermined after 1945 by special legislation that progressively went around its assumptions. By the postwar period, only one in three immigrants entered the United States under the national origins system (from House Report No. 745, 89th Congress, First Session, August 6, 1965, p. 11, cited in King 2000: 242). Nevertheless, what held the reformers together was fundamental opposition to the racial and discriminatory basis of the quota system, and support for a new system that was more consistent with the values of the 1960s. What is most striking, however, is that given the long history of racism in the United States, the issue of immigration did not become racialized after 1965, even when it became apparent that the primary result of the new legislation was a rising wave of third world immigration. For a moment in time in the mid-1990s it appeared as if the pattern would be repeated, that in a somewhat different form the politics of immigration in the United States would reproduce the racialized pattern of Europe, and, indeed, our own heritage from the early part of the century. Many of the elements seemed to be in place for a reframing of the immigration issue: a political reaction in which California was at the cutting edge, an upsurge of public opinion that followed, and the beginning of a resurgence of a movement that resembled the eugenics movement of the turn of the century. Dorothy Nelkin wrote at the time that: The immigration discourse of the mid-1990s is assuming an ominous but familiar tone. We hear, for example, that “natural” laws support “territorial integrity,” that certain groups are “genetically inferior,” that immutable biological differences

A transatlantic comparison

375

underlie social distinctions, and that immigration will weaken the American “gene pool” and result in “race suicide.” Once again, arguments about race relations in America (The Bell Curve) have been linked to immigration (Brimelow, Rushton, and the statements issued by FAIR). (Nelkin and Michaels 1998: 33)

It appeared that California was once again serving as a harbinger of changes in national policy. (Indeed, the emerging dialog had real repercussions. In 1997, the now forgotten United States Commission on Immigration Reform, the Jordan Commission, had recommended that legal immigration be cut by a third; in 1994, the State of California passed a referendum, Proposition 187, that would limit access of even the children of illegal immigrants to schools, hospitals, and welfare services; and federal legislation in 1996 limited some welfare state benefits to legal immigrants. Finally, public opinion seemed to be moving sharply towards support for at least limited restriction.) However, in many ways, the lessons of California were far different from the anti-immigrant storm that had emerged out of California a century before (see above). As it evolved, the campaign in California focused most intensely on the economic burden of the new immigration, and when the economy improved during the next five years, the identity campaign gained little support. Moreover, although the campaign in California began as a challenge to immigration, it unexpectedly encouraged the political mobilization of the Mexican-Americans, who then began to vote in larger numbers. As more immigrants and ethnic Mexicans began to vote, the focus of the immigration issue for political leaders began to change (see below), not only in California, but throughout the country. Political space and arenas of decision-making In each of our cases, the framing of the problem of immigration has been directly related both to questions of political space, as well as to the arenas of decision-making within which immigration policy is developed – but not, however, in the same ways. Thus, the politicization of immigration issues is related to the distribution and concentration of immigrant populations, but in ways consistent with how the issues have been framed. France Jeanette Money, presents interesting evidence that as early as 1971 the Right-wing national majority began to take immigration control and the “problem” of immigrants seriously for electoral purposes at the local

376

Immigration and the Transformation of Europe

level, as part of a strategy to defend their majority against the resurgent Left. She demonstrates that there was a high proportion of potential swing electoral constituencies in departments with high concentrations of immigrants (generally, where the Left was strong), and that the Rightwing majority targeted anti-immigrant appeals to those constituencies. These appeals were widespread during the municipal elections of 1971, the legislative election campaign in 1973, and were even more widespread during the presidential election campaign in 1974. Thus, although immigration was certainly not an important national issue in these campaigns, it was of far greater importance at the local level, in swing constituencies controlled by the Left, and seen as vulnerable by the Right. There is no question that immigration issues were important to Communist and some socialist mayors in towns with large immigrant populations during this period. Communist mayors in particular anticipated political reaction to North African immigration, and right-wing rivals saw this as an issue that could be exploited. Nevertheless, there was no clear evidence of pressure from the voters themselves, and attempts to mobilize voters around immigration were not successful. By the late 1970s, local Communist governments were trying to limit access of immigrant residents to local services, and were pressing the national government to impose local quotas on the immigrants resident in these towns. This was capped by an anti-immigrant campaign by the Communist Party during the presidential election of 1981. However, this campaign faded quickly when it failed to resonate among voters, and after the PCF lost support in the towns that had been the focus of this campaign (Schain 1985). In fact, in the 1973 legislative elections the Right lost support in constituencies where the Left was strong, many of which were those swing constituencies with high immigrant populations that had been targeted by the Right. Thus, although there was certainly political pressure from some local governments in areas of high immigrant concentrations, this is quite different from electoral or public pressure. Furthermore, it is important to differentiate between voters’ concerns, and the priorities of voters’ concerns when they actually vote. Indeed, as late as 1984 (at the point of the breakthrough of the National Front), only 6 percent of the French electorate gave immigration as a motivation for the electoral choice. Even after the question of immigration was clearly on the political agenda, opposition to immigration tended to vary considerably by locality, but not in the expected way. If we examine the variation of anti-immigrant public opinion in 1996 by geographic concentrations of immigrants among different towns (communes), for example, we find

A transatlantic comparison

377

Table 14.1 Percentage of respondents who claim there are “too many Arabs in France”∗ % Immigr in Commune/ Party ID

None

−1–4%

−5–10%

10%+

Total

PC PS UDF RPR FN

66.7% 75.9 83.3 80.0 100.0

70.0% 54.5 70.0 85.0 95.5

31.6% 46.0 92.3 81.3 92.0

25% 26.5 72.7 71.4 85.7

41.2% 47.5 77.3 80.0 92.0

Total

74.2

64.4

62.8

46.9

60.9

∗ Does

not include smaller parties and non-identifiers. Source: CSA survey 9662093, November, 1996, Q5246/RS12/ETR.

that opposition declines as the proportion of immigrants increases (see Table 14.1). This would appear to support some of the ideas of contact theory developed in the American context: either that more extensive contact tends to reduce opposition and conflict; or that in areas where there are high concentrations of immigrants, the French population most strongly opposed to immigration tend to move somewhere else. On closer examination, however, it appears that this inverse relationship most accurately reflected the sentiments of respondents who identify with the Left, rather than with the Right. In fact, we can infer from Table 14.1 a growing gap between voters of the Left and Right as the proportion of immigrants increases within communes. Where the immigrant population is more than 10 percent, the gap between Left and Right is, on average, more than 50 percent; compared to 14 percent where there is no immigrant population. Indeed, it was the National Front that succeeded where other parties had failed to politicize the issue of immigration, and to mobilize large numbers of voters for whom immigration was a political priority. In 1986, well before the more general mobilization of working class voters in favor of the FN in 1993, there was an impressive electoral breakthrough for the National Front in the core Communist bastions, all towns in departments with indicators of “high anti-immigrant pressure” analyzed by Money, that had maintained Communist local governments since at least 1947. The combination of anti-immigrant sentiments and concentrations of immigrants tended to magnify support for FN and diminish support for the traditional Right. However, the success of the National Front did not simply mobilize public opinion; it changed it by politicizing it. In 1984, what most clearly

378

Immigration and the Transformation of Europe

Table 14.2 The motivations of voters: 1984–97∗ (percentage of party voters voting for these reasons) %: ->

Law and order

Immigrants

Unemployment

Social inequality

84

88

93

97

84

88

93

97

84

88

93

97

84

88

93

97

PC 9 PS 8 Rt 17 FN 30 TT 15

19 21 38 55 31

29 24 37 57 34

28 29 43 66 35

2 3 3 26 6

12 13 19 59 22

16 19 33 72 31

15 15 22 72 22

37 27 20 17 24

59 43 41 41 45

77 71 67 64 68

85 83 72 75 75

33 24 7 10 16

50 43 18 18 31

52 40 23 26 32

46 47 21 25 35

∗ Since several responses were possible, the total across may be more than 100%. For 1988, the results are for supporters of presidential candidates nominated by the parties indicated. Sources: Exit Poll, SOFRES/TF1, June 17, 1984, Le Nouvel Observateur, June 22, 1984; ´ de l’opinion, Cl´es pour 1987 (Paris: Seuil, 1987: 111); Pascal Perrineau, and SOFRES, Etat “Les Etapes d’une implantation e´ lectorale (1972–1988),” in Nonna Mayer and Pascal Perrineau (eds), Le Front National a` d´ecouvert (Paris: Presses de la FNSP, 1988: 62); Pascal Perrineau, “Le Front National la force solitaire,” in Philippe Habert, Pascal Perrineau, and Colette Ysmal (eds), Le Vote sanction (Paris: Presses de la FNSP/Dept. d’Etudes Politiques du Figaro, 1993: 155; CSA, “Les Elections legislatives du 25 mai, 1997,” Sondage Sortie des Urnes pour France 3, France Inter, France Info et Le Parisien: 5.

differentiated the voters for the National Front from those of the more established Right (as well as other parties) was the priority that they gave to the issue of immigration (see Table 14.2). What is more striking, however, is how the priorities of the National Front and its voters appear to have influenced the priorities of voters of other political parties. Relatively few voters, aside from those that supported the National Front, considered either immigration or law and order to be a strong priority at the time that the FN achieved its electoral breakthrough. By 1988, however, these issues ranked alongside social inequality, and far higher than concerns about the environment, corruption and the construction of Europe; only concern with unemployment ranked higher. In this context of politicized public opinion the relationship between attitudes and concentrations of immigrants at the commune level becomes more important. In those areas where the established parties of the right were not successful in stimulating transfers of votes from the Left by raising anti-immigration issues in 1971–1973, the FN was far more successful in luring transfers from the established Right and then attracting new voters from the potential working class constituency of the Left.

A transatlantic comparison

379

Table 14.3 Circumscriptions in France (m´etro) with 10% + immigrant population (1999) by the political party of the d´eput´e

Right Left Total

Paris (ville)

Suburbs of Paris

%

Provinces

%

9 11 20

24 22 46

66 77 71

17 10 27

34 23 29

Total 50 43 93

100% 100% 100%

Source: INSEE Circonscriptions l´egislatives: r´esultats du recensement de la population de mars 1999.

Since the 1970s, the political geography of immigration in France has changed substantially. Areas of immigrant population have become more concentrated, while the political domination of these areas by the Left has been reversed. By 1999, only 17 percent of the electoral circumscriptions (districts for National Assembly elections) in France had immigrant populations of more than 10 percent. More than 70 percent of these circumscriptions were in the Paris region, with most of these in the suburbs. However, a majority of them were represented by the Right – everywhere except in the city of Paris itself. This shift is indirectly related to the success of the National Front, because of its growing ability to mobilize working class voters; even where the National Front was eliminated or withdrew in the second round in elections in the 1980s, most of these voters never shifted back to the Left. Although the normal vote of the Left has declined by only a few percentage points since the 1970s, it has declined far more in the 32 d´epartements in which there is the highest concentration of immigrants than in other d´epartements. The turning point – argues Pierre Martin – was the European elections of 1984, the first breakthrough of the National Front (Martin 1998: 154). Martin’s analysis is supported by Nonna Mayer’s work, where she argues that, “ . . . la pr´esence de populations e´ trang`eres exerce un effet sp´ecifique sur le vote FN, ind´ependant des charact´eristiques sociales et culturelles de l’´electeur” [1999: 258–9], and see Table 14.3). In this way, immigrant presence provides a key to the realignment of the party system. Table 14.1 indicates that higher concentrations of immigrant populations offer the possibility of politicization, but in which way? How should the policy problem be framed, and how should policy be developed? In France, the definition of the problem has long been framed as that of ethnic danger. Locally, however, there have been variations in the way the problem has been framed. Some towns governed by the Left

380

Immigration and the Transformation of Europe

(generally the Socialists) have seen immigrants as a political resource, and Table 14.1 reflects this orientation. On the other hand, towns governed by the Right have generally been prone to see immigrants as a challenge to French identity. Table 14.1 reflects this as well. However, with so few circumscriptions with concentrations of immigrants more than 10 percent, and with most of these circumscriptions dominated by the Right, there appears to be little incentive to frame the problem of immigration in more positive terms. On the one hand, this has meant that the political advantage of a more positive approach to immigration nationally is limited, since there are few electoral constituencies in which there are large numbers of immigrants who could represent a potential gain in electoral support. On the other hand, the political will to do so is substantially reduced as these areas have become dominated by the Right, which tends to see immigrants as a problem rather than as an electoral potential (see Table 14.3). In France, the pressure that drove the immigration issue towards restriction in the 1970s came from localities dominated by the Left, pressures that were moderated by their national parties. From the early 1980s, the issue was driven by constituency-level competition between the established Right and the National Front. Britain In contrast to France, where immigrant concentration was strongest in localities dominated by the Left, at least during the period when immigration restriction was initiated, two-thirds of the 106 constituencies with at least 5 percent foreign-born population were held by the Conservatives after the 1959 parliamentary elections. About one-third of these were classified by the Tories as swing constituencies. Jeannette Money argues that this provided an incentive for the Conservatives to initiate legislation for immigration restriction as a response to voter sentiment in key constituencies. Nevertheless, as in France, the political gain nationally was minimal, and the Tories lost the 1964 election. There was also clear electoral pressure in some Labour-represented constituencies with high immigrant populations, where Butler and Stokes estimate that Labour lost three seats and was prevented from winning others. In Aston (Birmingham), the sitting Labour candidate lost his seat in 1964 after he refused to shift his stand on immigration – even though the party itself supported the Tory legislation of 1962 that it had vehemently opposed at the time. And then, of course, there was the shock of Smethwick, a safe Midlands constituency where Patrick Gordon Walker – Labour’s shadow foreign secretary – was overwhelmed by his

A transatlantic comparison

381

anti-immigrant Conservative opponent; he then lost again in a by-election a few months later, when the party gave him a safe seat to bring him back into the cabinet. Anthony Messina argues that, “The extent to which the electoral outcome at Smethwick altered the major parties’, and especially Labour’s, perception of the race issue cannot be overstated.” Ten years later, Labour politician Richard Crossman would summarize the problem: Ever since the Smethwick election it has been quite clear that immigration can be the greatest potential vote-loser for the Labour party if we are seen to be permitting a flood of immigrants to come and blight the central areas in all our cities. (cited in Messina 1989: 34–6)

Nevertheless, there is at least some evidence that the electoral pressure played differently for the Tories than for Labour; that far from attempting to outbid one another to “ . . . shift their policy positions in response to changing community preferences,” each party was seeking to mobilize different kinds of voters in areas of high immigrant concentration. In a survey taken among MPs in 1969, the differences between the two major parties could not have been more stark, with Labour MPs strongly opposed to restriction, and more opposed in constituencies of high immigrant concentration. The commitment of Tory MPs, on the other hand, was to go much further down the road of restriction, even after the most restrictive legislation had already been passed – and even during a period when policy agreement between Labour and the Conservatives appeared to be strong. By 1969, well over half of Tory MPs were prepared to agree with the most radical anti-immigrant propositions, with the percentage increasing with immigrant concentration (see Table 14.4). Therefore, immigration appeared to be a Conservative issue that divided Conservative politicians, far more than an issue that divided the Left. There is plenty of evidence of electoral pressure at the constituency level, but the impact of that pressure was different for each of the major parties. For the Tories, the pressure from MPs was for a commitment to greater restriction; Labour MPs representing constituencies with large numbers of immigrants tended to favor less restriction. What seems to have happened in Britain is that national parties assumed relatively moderate national positions on immigration restriction, in reaction to more radical pressure from their constituency organizations. The political dynamics were also different. In France the positions of the national and local parties of the Right were similar; while the local parties of the Left appeared to be more consistently favorable to restriction than the national parties, which were far more conflicted.

382

Immigration and the Transformation of Europe

Table 14.4 MPs’ attitudes on immigration and repatriation in 1969 and 1982 (percentage) 1969: Britain must completely halt all coloured immigration, including dependents, and encourage the repatriation of coloured persons now living here.

Labour MPs from high immigrant areas Labour MPs from low immigrant areas Conservative MPs from high immigrant areas Conservative MPs from low immigrant areas

Agree

Disagree

Strongly disagree

Don’t know

Total

4%

29%

66%

–

100%

6

38

54

2

100

50

7

36

7

100

37

45

13

5

100

Source: Frasure 1971.

However, although the political geography of immigration has changed substantially in France since the 1960s, the change in Britain seems somewhat more limited. As in France, the areas with an immigrant concentration have remained small and concentrated in a few areas. However, in contrast to the 1960s, we can estimate that there has been a substantial growth in the number of electoral constituencies with more than 10 percent New Commonwealth (NCW) immigrants (Money estimates that there were about 33 constituencies with 10 percent or more NCW immigrant and alien immigrants [1999: 79]). We can tentatively estimate this change from the most recent census data. Table 14.5 summarizes the results of localities in which the non-white population is greater than 10 percent, together with estimates of the party affiliation of the MP elected in 1997. Of the 387 local authorities for which we have census results, 59 (15.2 %) have non-white populations of 10 percent of more, of which twenty-five are in London, four are in the Manchester area, and eleven in the remainder of the Midlands. However, there has probably been a shift in political representation in these areas. In the 1960s Money reports that 17 percent of the electoral constituencies had immigrant populations of 5 percent or more, and of these, two-thirds were represented by Tories. On the other hand, Frasure’s estimate of representation during this period, based on different estimates, gives the advantage to Labour (Money 1999: 85; Frasure 1971: 203).

A transatlantic comparison

383

Table 14.5 British local authorities with 10% or more non-white population, by political representation

Labour Conservatives Lib-Dem Total

London

Outside

Total∗

11 4 2 17 (London + W. Midlands = 69%)

24 3 0 27

35 7 2 44

∗ There

was no reliable political data for 15 localities. Source: Census results for England, Scotland and Wales for 2001; results of elections for 1997, House of Commons: Total constituencies = 630.

By the 1990s, however, the advantage was clear. Of the forty-four local authorities for which we can reasonably estimate political representation, thirty-five are represented by Labour. They are in the very same areas cited by Frasure. This would appear to increase the incentives for a more positive orientation towards immigrants (or non-whites), as well as incentives for their electoral mobilization at the local level. The opportunities, however, remain relatively limited. The United States The United States case is quite different from the European cases, both with regard to the political geography of immigration, and with regard to the relationship between political geography and policy outcomes. Immigrant populations have been concentrated in the United States, but far less than in France and Britain. In 1990 six states (California, New York, Florida, Texas, New Jersey, and Illinois, in that order) accounted for almost three-quarters of the foreign-born population in the United States, with California and New York accounting for almost half. Politically, immigrant populations in these states gained considerable importance – or at least potential importance – during the following decade. In each of them, congressional districts (CDs) with 10 percent or more of the populations born abroad accounted for half or more of the CDs by 2000. Thus, with two exceptions, every CD in California had a population born abroad greater than 10 percent, as did two-thirds of those in NY, and more than half in Texas. In each state, the local political dilemma was how to deal with the question of immigration and immigrants. One option was presented in California in 1994, where a sustained effort was made to mobilize growing negative public opinion against undocumented immigrants. In an environment of high immigration

384

Immigration and the Transformation of Europe

pressure and rising local unemployment, Proposition 187 – which would limit access of these immigrants and their families to state services – qualified for the 1994 ballot, supported by Republican Governor Pete Wilson who had previously supported immigration from Mexico. Wilson was re-elected and the initiative passed in a campaign that grew increasingly anti-immigrant as it wore on. During the same time period, both Democratic senators from California introduced immigration control legislation in the United States Senate. During the next two years, President Clinton toughened the patrols along the Mexican border, supported legislation that restricted the rights of legal aliens, and established the Commission on Immigration Reform, the Jordan Commission, which quickly recommended a reduction in annual immigration limits. Finally, Pat Buchanan became the most prominent political leader in favor of immigration restriction after his Republican primary victory in New Hampshire in 1996. Thus, the California conflagration spread quickly to national politics. A second option, however, was presented in other states with high immigrant concentration. Although there was some anti-immigrant activity in Florida and Texas during this period, it was not supported by state governments; and there was virtually no such activity in other areas of immigration concentration, specifically New York. In addition, although there were certainly attempts to define the immigration problem in racial terms by certain pundits, intellectuals and politicians (notably Pat Buchanan), even as framed in California and Washington, it was essentially a welfare issue, that is, competition for limited resources. Surprisingly, given the history of immigration and race in the United States, attempts to racialize the immigration issue did not resonate with public opinion. In Mich`ele Lamont’s study of American and French workers (researched during this period in the New York and Paris metropolitan areas), in contrast to their French counterparts, American workers (white and black): . . . do not define “people like us” in clear opposition to immigrants. In interviews, most were indifferent to them. Others held relatively positive views of this group. A few had more negative views and focused almost exclusively on language issues as opposed to moral character issues. This pattern holds for white and black workers. (Lamont 2000: 212–13)

The construction of the immigration issue in welfare terms, meant, above all, that electoral pressures were defined in a different way than in France or Britain. As the economic constraints began to ease in California (and nationally) after the mid-1990s, attitudes towards immigrants and immigration began to change as well (see Table 14.6).

33 7

39

13

1965%

Sources: Roper Poll, 1953; Gallup Poll, 1965–2002.

“Should immigration be decreased” “Increased”

1953%

7

42

1977%

7

49

1986%

Table 14.6 Opposition to immigration in the United States

6

65

1993%

7

65

1995%

10

44

1999%

14

41

2001%

8

58

2001%

12

49

2002%

386

Immigration and the Transformation of Europe

However, the change of attitudes seemed to be substantially linked to the resurgence of a second political option for dealing with immigration, an electoral option that gave priority to the mobilization of potential immigrant voters. The politics of immigration in both France and Britain was built around the mobilization of sentiment against immigrants. However, the other impact of immigration is the potential of citizenship, its impact on the distribution of votes among political parties, and the speed with which this impact is felt. Kristi Andersen demonstrates that the party realignment that took place in the United States between 1928 and 1936 was essentially related to a new electorate of immigrants and their children voting for the first time in large cities in the United States, rather than voters switching from the Republican to the Democratic Party (Andersen 1979). More recently, in areas of high immigration, where immigrants are voters or are perceived as voters or potential voters, the electoral pressure has increasingly moved political elites towards a more favorable position on immigration. This dynamic appears to have taken hold in the United States, first in localities like New York and California (Waldinger and Bozorgmehr 1993), and more recently at the national level. Almost a decade after Proposition 187, it is clear that the impact in California was to mobilize new immigrant voters around the Democratic Party. Wilson lost his second race for governor, and Orange County, long a conservative Republican bastion, is now increasingly competitive, thanks to the incorporation of Latin American immigrants and their children into the electorate (two of the six congressional districts that are all or in part in Orange County are solidly Democratic, with 60 percent or more of the vote in 2000 and 2002). On the national level, immigration is at an historic high, interest in restriction in Congress seems to have faded, and public opinion, in contrast to most countries in Europe, has moved against restrictionism. Pat Buchanan played no significant role in the presidential election of 2000 (if we ignore the “chad” fiasco in Florida), and the conservative Republican President of the United States is generally favorable to immigration and immigrants – often in Spanish! Finally, the AFL–CIO (American Federation of Labor – Congress of Industrial Organizations) has announced that it will no longer oppose even illegal immigration, and would make a major effort to organize new immigrant workers. Certainly the most interesting aspect of this change has been the decision by the President to court the Latino and prospective Latino vote. Existing studies show that Latinos (with the exception of Cubans) are strongly Democratic in orientation, and become more so with increasing education and tenure in the United States. As governor of Texas,

A transatlantic comparison

387

Table 14.7 US Congressional districts 10%+ born outside of the United States (2000) after Congressional elections of 1998 New York/Cal

%

Remainder of US

%

Republican CDs(50%+ 1998) Democratic CDs(50%+ 1998)

21 48

38 50

34 48

62 50

55 100% 96 100%

Total

69 (46%)

46

82

54

151 100%

Total

Source: Bureau of the Census, Congressional District Data Book CDs = 435.

President George W. Bush had some success in attracting Latino voters, and the President seems to feel that not to make this effort would be to surrender the electoral future to the Democrats (Gimpel and Kaufman 2001). Indeed, this gamble had some payoff in the 2004 presidential election. At least some of this shift in orientation can be attributed to the different kinds of political geography of immigration in the United States as compared to Europe. Immigrant populations are concentrated in certain areas of the country, but from a national perspective, these areas are crucial in presidential elections. Moreover, though limited, these areas are also far more widespread than in Europe (see Table 14.7). More than one-third (35 percent) of the congressional districts in 2000 had immigrant populations of 10 percent or more, and, although they tend to be concentrated in relatively few states, they are spread across 22 states. This distribution of CDs with a high proportion of immigrants is about twice as great as France or Britain, and provides a reasonable measure of the potential electoral gains. While these gains can be particularly important for the Democrats, since two-thirds of these CDs have Democratic representation, the challenge is also quite real for the Republicans who represent the other third. With this number of CDs at stake, neither party can afford to ignore the electoral potential of immigrant populations. Thus, compared to France and Britain, the electoral stakes are far more important in the United States. While the mobilization of immigrant citizens and ethnic voters has become central to American party competition at the national level, it has been marginal and episodic in France and Britain. Miriam Feldblum (1999: 43) analyzes efforts by the French Socialist and Communist parties to attract the new immigrant vote in the 1980s. Le Monde, December 3, 2003, documents disappointment with the Left among immigrant voters, and attempts by the Right to attract their support.

388

Immigration and the Transformation of Europe

In Britain, there are studies that document the success of a small number of non-white ethnic candidates (overwhelmingly Labour) in British local elections in the 1980s, as well as a small number of alliances between ethnic organizations and local authorities, “ . . . a handful of authorities” (see Studlar and Welch 1992: 147–9; Rath and Saggar 1992: 210–13). The evolution of this dynamic, which appears to have taken hold first in localities like New York, may be related to the local access structure. In part, because the structure of politics is quite different in Los Angeles and New York, a larger immigrant population was isolated from the political system in LA for a longer period of time. The combination of non-partisan elections, a highly centralized system and a small city council with large districts gave Mexican-Americans little leverage in local government. Whereas New York’s ethnic demography leads to electoral competition, Los Angeles’ demography has the opposite result. Heavy immigrant densities make the Mexican-American districts into rotten boroughs, where only a small proportion of the adult population votes. Thus, in the recent mayoral election, the most densely Mexican council district turned out one fifth as many voters as did the heavily white districts in the San Fernando Valley or the west side. In New York’s last mayoral election, by contrast, the turnout in immigrant Crown Heights was more than two-thirds the turnout on the affluent, largely white upper east side. (Waldinger and Bozorgmehr 1993)

In effect, the proliferation of electoral offices in New York during the 1990s, and the greater openness of the system, provided a structure that incorporated immigrants more rapidly into the political system. Of course this also made it less tempting to use immigrant population as the object of an anti-immigrant campaign. Despite these constraints, however, the impact of new voters – even in California – is now being felt. Conclusion This chapter began with an attempt to understand differences between policy on immigration in Europe and the United States. To understand these fairly striking differences, we focused first on how the question of immigration has been framed and politicized on either side of the Atlantic. I have argued that the divergence is related to differences in political geography. The electoral dynamic that has been increasingly important in the United States case has made it more difficult to define the immigration question in exclusionary terms. This dynamic, in turn, has also had an impact on public opinion, and support for a more open immigration policy. In France and Britain, the exclusionary framing of the immigration question has been challenged less than in the United States by the political

A transatlantic comparison

389

geography of immigration. Although in each case political geography has provided a different political opportunity structure for the development of policy, even under similar circumstances, the outcomes tended to vary in specific ways. In Britain, the radical orientation of Tory MPs in constituencies with relatively high percentages of immigrants reverberated throughout the party, and tended to encourage the move towards exclusion by consensus at the national level in the 1960s, since each national party developed a compromise towards the center. Since then, there has been little incentive to move in a different direction. In the French case, control by the Left of areas of immigrant concentration resulted in a substantial problem at the local level, and pressures for exclusionary policies, at the same time that national parties pressured for policies that work to the benefit of immigrants already in the country. In more recent years, the combination of conservative political representation of immigrant areas, and the pressure imposed by the National Front, has meant that there has been very little incentive to develop a more open policy orientation on immigration. Thus, the development of immigration policy on either side of the Atlantic is related to political forces that are quite different, particularly if we focus on issue framing and the political geography of immigration. The politics of immigration in the United States tend to favor more open immigration policies, in contrast to Europe, where the politics of immigration tend to create pressures that favor the maintenance of immigration restriction. On the other hand, there are now other pressures in Europe which tend to favor more open immigration policies. The Tampere summit of European heads of state in 1999, recognized two widely discussed needs in Europe for immigrant labor: labor market needs in such areas as technology, agriculture, construction, and services; demographic needs posed by pressures on the welfare state (see Commission of the European Communities 2000). It has been difficult to raise this issue at the national level because of the challenge of the extreme right in a number of European countries. Nevertheless, there have been some tentative movements to redefine the immigration problem. In France, just after the split in the extreme right National Front in 1999, former Prime Minister Alain Jupp´e proposed a more open immigration policy that would take into consideration emerging labor and demographic needs. His proposals might have been taken more seriously, had the National Front not begun to show increased strength in the months after they were made (see, for example, the failed trial balloon floated by Alain Jupp´e reported in Le Monde, October 1, 1999). Nevertheless, the issue has not disappeared, and has continued

390

Immigration and the Transformation of Europe

to be publicly discussed in government reports (for example, a widely publicized report of the French Economic and Social Council, supported by MEDEF – the French employers’ association – recommended that France “ . . . open our frontiers to controlled immigration,” and estimated a need for an increase of 10,000 foreigners per year; see Le Monde, November 8, 2003). In Germany, the SPD/Green Government persevered for three years, and finally passed legislation (in a compromise with the CDU opposition) in June 2004 that would formally open the country to the legal immigration of highly skilled workers from outside of the EU, for the first time since the 1970s. (The legislation is a follow-up of a five-year green-card program that was initiated in 2000 to attract highly skilled information technology specialists. Permits were granted for up to five years, without possibility of permanent residency or naturalization.) Harmonization of these initiatives at the European level – the mandate of Tampere – is, however, limited by the fact that few EU countries have anything resembling an immigration policy of any kind (beyond a policy of restriction). Nevertheless, through its reports and recommendations, the European Commission has emerged as an important agenda-setting force (Commission of the European Communities 2000). Therefore, for different reasons from the United States, and through a very different process, Europe appears to be edging towards a more open immigration policy. Just how far it goes will depend on the economic trade-offs, public opinion, EU processes, and other factors examined by the other chapters in this volume – and most directly, on the political geography that connects them to policy.

 Andersen, Kristi 1979 The Creation of a Democratic Majority, 1928–1936. Chicago, IL: University of Chicago Press. Bleich, Erik 2003 Race Politics in Britain and France: Ideas and Policymaking since the 1960s. New York: Cambridge University Press. Calvez, Correntin 1969 “Le Probl`eme de travailleurs e´ trangers.” Journal Officiel de la Republique Française, Avis et Rapports du Conseil Economique et Social, March 27: 315. Carter, Bob, Clive Harris and Shirley Joshi 1987 “The 1951–55 Conservative Government and the Racialization of Black Immigration.” Immigrants and Minorities 6(3): 335–47. Commission of the European Communities 2000 “Communication from the Commission to the Council and the European Parliament on a Community Immigration Policy.” COM(2000) 757 final. Brussels: Commission of the European Communities. November 22. http://europa.eu.int/eurlex/en/com/cnc/2000/com2000 0757en01.pdf.

A transatlantic comparison

391

Degler, Carl N. 1991 In Search of Human Nature: The Decline and Revival of Darwinism in American Social Thought. New York: Oxford University Press. Feldblum, Miriam 1999 Reconstructing Citizenship: The Politics of Nationality Reform and Immigration in Contemporary France. Albany, NY: State University of New York Press. Foot, Paul 1965 Immigration and Race in British Politics. Baltimore, MD: Penguin Books. Frasure, Robert C. 1971 “Constituency Racial Composition and the Attitudes of British MPs.” Comparative Politics 3(2) (January): 201–10. Freeman, Gary P. 1995 “Modes of Immigration Politics in Liberal Democratic States.” International Migration Review 29(4) (Winter): 881–902. Freeman, Gary P. 1989 “Immigrant Labour and Racial Conflict: The Role of the State,” in Migrants in Modern France: Population Mobility in the Later Nineteenth and Twentieth Centuries, Philip E. Ogden and Paul White (eds). London: Unwin Hyman. Geddes, Andrew 2003 The Politics of Migration and Immigration in Europe. London: Sage. Giddings, Franklin Henry 1896 The Principles of Sociology: An Analysis of the Phenomena of Association and of Social Organization. New York: Macmillan Company. Gimpel, James G. and Karen Kaufmann 2001 “Impossible Dream or Distant Reality? Republican Efforts to Attract Latino Voters.” Washington, DC: Center for Immigration Studies, August. www.cis.org/articles/2001/ back901.html. Glazer, Nathan 1954 “Ethnic Groups in America: From National Culture to Ideology,” in Freedom and Control in American Society, Morroe Berger, Theodore Page and Charles H. Abel (eds). New York: Van Nostrand. Hansen, Randall 2000 Citizenship and Immigration in Post-War Britain: The Institutional Origins of a Multicultural Nation. Oxford: Oxford University Press. Higham, John 1963 Strangers in the Land: Patterns of American Nativism, 1860– 1925. New York: Atheneum. Ireland, Patrick R. 1994 The Policy Challenge of Ethnic Diversity: Immigrant Politics in France and Switzerland. Cambridge, MA: Harvard University Press. Kilson, Martin 1975 “Blacks and Neo-Ethnicity in American Political Life.” Chapter 8 in Ethnicity: Theory and Experience, Nathan Glazer, Daniel P. Moynihan and Corinne Saposs Schelling (eds). Cambridge, MA: Harvard University Press. King, Desmond S. 2000 Making Americans: Immigration, Race, and the Origins of the Diverse Democracy. Cambridge, MA: Harvard University Press. Lamont, Mich`ele 2000 The Dignity of Working Men: Morality and the Boundaries of Race, Class, and Immigration. Cambridge, MA: Harvard University Press. Lodge, Henry Cabot 1891 “The Distribution of Ability in the United States.” Century Magazine 42(5) (September): 687–95. http://cdl.library. cornell.edu/cgi-bin/moa/moa-cgi?notisid=ABP2287-0042-165. Lowi, Theodore J. 1969 The End of Liberalism: Ideology, Policy, and the Crisis of Public Authority. New York: Norton.

392

Immigration and the Transformation of Europe

Martin, Pierre 1998 “Qui vote pour le Front national français?” in L’extrˆeme droite en France et an Belgique, Pascal Delwit, Jean-Michel De Waele, and Andrea ´ Rea (eds). Brussels: Editions Complexe. Mayer, Nonna 1999 Ces Français qui votent FN. Paris: Flammarion. Messina, Anthony M. 1989 Race and Party Competition in Britain. Oxford: Oxford University Press. Miles, Robert and Annie Phizacklea 1984 White Man’s Country: Racism in British Politics. London: Pluto Press. Mink, Gwendolyn 1986 Old Labor and New Immigrants in American Political Development: Union, Party, and State, 1985–1920. Ithaca, NY: Cornell University Press. Money, Jeannette 1999 Fences and Neighbors: The Political Geography of Immigration Control. Ithaca, NY: Cornell University Press. Nelkin, Dorothy and Mark Michaels 1998 “Biological Categories and Border Controls: The Revival of Eugenics in Anti-Migration Rhetoric.” International Journal of Sociology and Social Policy 18 (5–6): 35–63. Paul, Kathleen 1997 Whitewashing Britain: Race and Citizenship in the Postwar Era. Ithaca, NY: Cornell University Press. Rath, Jan and Shamit Saggar 1992 “Ethnicity as a Political Tool in Britain and The Netherlands,” in Ethnic and Racial Minorities in Advanced Industrial Democracies, Anthony M. Messina, Luis R. Fraga, Laurie A. Rhodebeck and Frederick D. Wright (eds). New York: Greenwood Press, 210–13. Salyer, Lucy 1989 “Captives of Law: Judicial Enforcement of the Chinese Exclusion Laws, 1891–1905.” Journal of American History 76 (June): 91–117. Schain, Martin A. 1985 “Immigrants and Politics in France,” in The French Socialist Experiment, John S. Ambler (ed). Philadelphia, PA: Institute for the Study of Human Issues. Schattschneider, Elmer E. 1960 The Semisovereign People: A Realist’s View of Democracy in America. New York: Holt, Rinehart and Winston. SOPEMI (Systeme d’Observation Permanente sur les Migrations) 2004 Trends in International Migration: SOPEMI – 2003 Edition. Annual Report, 2003 edn. OECD Emerging Economies. Paris: OECD. Spencer, Ian R. G. 1997 British Immigration Policy since 1939: The Making of Multi-Racial Britain. New York: Routledge. Studlar, Donley T. and Susan Welch 1992 “Voting for Minority Candidates in Local British and American Elections,” in Ethnic and Racial Minorities in Advanced Industrial Democracies, Anthony M. Messina, Luis R. Fraga, Laurie A. Rhodebeck and Frederick D. Wright (eds). New York: Greenwood Press, 147–49. Taguieff, Pierre-Andr´e 1988 Le force du pr´ejug´e: essai sur le racisme et ses doubles. Paris: La D´ecouverte. Viet, Vincent 1998 La France immigr´ee: construction d/une politique, 1914–1997. Paris: Fayard. Waldinger, Roger and Mehdi Bozorgmehr 1993 “From Ellis Island to LAX.” Unpublished manuscript. Department of Sociology, UCLA. 1993. August.

15

“Useful” Gastarbeiter, burdensome asylum seekers, and the second wave of welfare retrenchment: Exploring the nexus between migration and the welfare state1 Georg Menz

Introduction: Managing migration, managing welfare Migration has re-emerged as a highly politicized, divisive, and contested issue in European politics over the past twenty-five years or so. Much of the public debate has focused on the economic ramifications of migration and a commonly rehearsed claim underlying calls for restrictive policy concerns: the “drain” on welfare systems that migrants constitute. The phenomenon of “welfare state chauvinism” is by no means unique to its Scandinavian origins, and throughout Europe populist xenophobes portray migrants as welfare scroungers. More sophisticated opponents of liberal migration point to the difficulties in labor market integration experienced by the second and third generation of Europe’s ethnic minorities. Meanwhile, pragmatic policy-makers regard migration as one possible tool in redressing Europe’s demographic deficit, the shifting age pyramid and the looming pension shortfalls. Despite the burgeoning literature on migration to Europe (Guiraudon and Joppke 2001; Geddes 2003; Lahav 2004), few scholarly efforts link discussions of the future of European welfare systems with immigration (for an exception see Bommes and Geddes 2000). This chapter addresses this gap in the literature by focusing on the following key question: how do the four main types of European welfare state regimes – liberal, Bismarckian, Scandinavian, Mediterranean – interact and cope with migration? In pursuing this line of inquiry and 1

The country sections on Germany, Italy, and the United Kingdom draw on extensive background interviews on migration policy design conducted with representatives of national ministries of interior affairs during 2004–05. I am grateful to the British Academy for providing the financial support to carry out these interviews and to Craig Parsons and participants at the two Luxembourg conferences associated with this book project for helpful comments on earlier drafts of this chapter.

393

394

Immigration and the Transformation of Europe

focusing on the mechanisms of inclusion (or lack thereof) by various welfare regimes in Europe, I explore both how migration affects these various welfare regimes and conversely how they affect migrants. In addressing these questions, the analysis draws on empirical material from the United Kingdom, Germany, Sweden, and Italy, representing the four “families” of “welfare capitalism” (Esping-Andersen 1990, 1996; Ferrera 1996). I develop two major hypotheses. Firstly, I argue that to the extent that the receipt of welfare benefits is based on employment and the welfare system is financed through employment-related contributions, governments will be particularly concerned to ensure high labor market participation rates, including among migrants. However, relatively low rates of labor market activity present a potentially serious challenge to such welfare regimes. Individuals who not only fail to contribute through employmentbased contributions, but may also be eligible for the lowest tier of transfer payments defy the logic of such work-based regimes. Aside from the assumptions made about levels of labor market participation embedded in the logic of the welfare regime, such a challenge may be compounded if a welfare system is largely financed through employment-related charges, rather than through general taxation. Fiscal policy is much more comprehensive in its coverage and the tendency to reduce direct personal taxation and rely more heavily on indirect taxation ensures that even transfer payment recipients and undocumented residents contribute to fiscal revenue. Thus I suggest that the Bismarckian welfare system experiences the potentially most serious challenge from immigration, followed by the Scandinavian model. By contrast, the minimalist Mediterranean and liberal welfare regimes, which are financed through fiscal revenues (or indeed rely partially on private charitable providers and family networks), appear to be somewhat less tested. Secondly, I argue that the return to the active management of migration witnessed across Europe over the past five years is in part due to the incorporation of an economistic competition state logic, but is also due to policy-makers being eager to engineer and steer immigration much as they did in the active labor recruitment programs in German-speaking Europe prior to the 1973 oil shock. This re-invented gastarbeiter paradigm owes its appeal to the insight that managed migration through active labor migration recruitment for select niches of the labor market is preferable to the status quo ante of the period between the late 1970s and the late 1990s, during which the “front door” was sealed but the humanitarian “side entrances” of family reunion and political asylum remained open (see chapter 12 in this volume, and also Menz 2006). European policymakers realize that they cannot close these side doors entirely due to international legal obligations, but they have attempted to limit access

Migration and the welfare state

395

through these avenues while cautiously re-opening the principal entrance. Active labor market recruitment of the 2000s thus goes along with limiting alternative access routes for immigration; it also attempts to counteract the allegedly negative and “draining” effects of these somewhat more “eclectic” and less regulated immigration channels. Inspired by the experience of classic immigration countries in North America and Australasia, European governments seek to select the young migrants with desired language and professional skills that are likely to integrate rapidly into the labor market (and presumably society as a whole) and to impose limited costs on the welfare systems. Europe’s demographic deficit, with its implied pension crunch, adds a certain urgency to this sort of recruitment. By contrast, asylum seekers (and other types of migrants entering under regulatory categories European governments are wary of accepting) are commonly portrayed as constituting a “burden,” as they commonly lack language skills that would secure their rapid integration into the labor market and their professional skills may not be adequately recognized or transferable. Its moral virtues aside, this line of reasoning can be sustained with reference to empirical evidence from all four case studies. However, many of the problems migrants face in their integration into the labor market are associated with failure or the lack of policy. Highly-skilled asylum seekers may therefore find themselves working in low-skilled jobs. Before turning to the four case studies, it is worth noting some implications of these two hypotheses. I submit that while any simple general charge of a “welfare drain” imposed by migration cannot be sustained – as the economic analysis by Bruecker et al. (chapter 5 in this volume) suggests – there do exist unfortunate interaction effects between migration and welfare systems, especially within the contribution-based Bismarckian model. While policy-makers strive to increase labor market participation regardless of the welfare system, the logic of the Bismarckian and the Scandinavian system compels them to pursue this aim with particular vigor. Such activation efforts that attempt to increase “employability,” often summarized by the United States term “workfare,” may not be entirely bad for migrants, since such active labor market policy may bring about language and professional skill training programs targeted at populations that were previously completely neglected, as will be discussed below. But they do generate tension, since “workfare” carrots usually come with hefty sticks attached. In keeping with the findings of the earlier chapters, I stress that welfare regimes cannot be said to exercise significant “magnet effects” not least because of the very severe restrictions on eligibility for any transfer payments. As we shall see, while such eligibility is commonly limited to

396

Immigration and the Transformation of Europe

documented, permanent residents, more generous provisions are always linked to participation in the labor market, a tie most pronounced in the Bismarckian system. The liberal welfare system offers a truly minimal bottom tier “safety net,” while in the Mediterranean system, access even to this lowest level may prove impossible for migrants. Just as the design of welfare regimes never incorporated immigrants’ concerns – and in fact was commonly specifically geared towards reserving access for permanent residents – neither the first wave of welfare state retrenchment under the conservative governments of the 1980s (Pierson 1994) nor the second wave of center-left active labor market policy of the 1990s and early 2000s specifically took migrants into consideration. Nevertheless, the re-design of European welfare states, informed by the neoliberal competition state logic as well as the need to address so-called “new social risks” (Bonoli 2004), has significant consequences for resident migrants eligible for transfer payments. In fact, the re-invention of the gastarbeiter paradigm and Europe’s throwing down the gauntlet to its former colonies in North America and Australasia in the competition for skilled human resources (Menz 2002) bestow renewed importance to concepts such as “employability” and “usefulness” for migrants old and new. While xenophobic rhetoric often links migrants with abuses of welfare state services, cuts have not usually been strongly linked to implicitly racist discourse. This difference to the United States may wane in the future, but currently signs of an Americanization of European welfare policy (Freeman 1986) due to decreasing solidarity based on perceived racial differences are scarce. Reagan’s mythical Cadillacdriving black welfare mom may have equivalents among Europe’s far right parties, but such a racialized image is as of yet less influential among policy-makers. In sum, the rediscovery of labor recruitment and more actively managed migration is partly motivated by the will to counteract the effects of passive humanitarian migration channels, while also being informed by a business-friendly competition state logic. Behind this backdrop certain welfare systems are potentially more challenged by migration than others. The minimalist southern and liberal model offers very little, or indeed nothing, by way of a social net to immigrants, while the Scandinavian and Bismarckian models face more serious challenges. Nowhere will policy-makers welcome without hesitation migrants that will potentially draw on transfer payments, even if limited eligibility means a relegation to the lowest tier of income support. Across the board, policy-makers will seek to strive for high labor market participation rates. But how and how much they do so varies with the interaction of migration and their kind of welfare regime.

Migration and the welfare state

397

Migration and European welfare systems Scholars take two broad views of the net impact of migration on welfare systems. Skeptics highlight that further migration may cause the “Americanization of European welfare politics” (Freeman 1986). Popular support for welfare transfer payments and solidarity more generally might decline in light of an ethnically more diverse population, where the lower socioeconomic stratum is composed of immigrants and their descendents – or where the situation is successfully portrayed in this fashion. While the empirical evidence does not support this position, the key assertion of migration skeptics cannot be dismissed as readily: while immigrants themselves may benefit from their relocation through higher lifetime earnings and remit payments to relatives and friends back home; they may also “fail,” and may thus, sooner or later come to be dependent on welfare state benefits. Indeed, the social problems experienced not by first generation migrants themselves but by second and third generation descendants – low levels of labor market participation, high level of dependency on transfer payments, low social mobility, and low levels of educational achievement – are often invoked as drawbacks of a passive or permissive migration policy, and have thus underpinned calls for active recruitment of “desirable” migrants and a more restrictive stance towards the “undesirables.” Low-skilled migrants could be seen as constituting a (future) liability, as economic restructuring and the outsourcing of manufacturing has led to a decline in labor market demands for such employees. Linguistic and cultural differences along with difficulties with the recognition of foreign degrees and diplomas might impair rapid labor market integration, thus causing dependency on transfer payments. In most countries, some types of migrants – particularly asylum seekers and political refugees – are commonly at least temporarily barred from employment. Such a ban can, of course, easily turn into a self-fulfilling prophecy and create dependence on transfer payments. Similarly, while proof of economic self-sufficiency is a commonly required prerequisite to family reunion, new arrivals are often female, underage, or retirees. Family reunion still constitutes the largest channel of legal migration in Europe. This position is not entirely compelling, as it neglects the failures of educational systems in ensuring acceptable standards of educational achievements for descendants (as developed in chapters 8 and 9 in this volume) and fails to realize that the proportionally high net dependency of migrants on transfer payments is conditioned not only by socioeconomic profile but also by more complex interrelated factors. Migrants have been disproportionately affected by deindustrialization and

398

Immigration and the Transformation of Europe

“precarization” processes. The shift towards the tertiary sector implies the need for higher linguistic and professional skills even for low-skill low-wage service sector jobs than was true for low-skill, or unskilled manufacturing jobs. The “precarization” of European labor markets creates more patchwork biographies, less job security and stability, temporary and part-time employment, implying more potential occasions for taking recourse to the social net. Piore’s (1979) earlier concept of a dual labor market with extreme variations in wage levels, working conditions, safety, and stability draws attention to the fact that there is a structural necessity for migrant labor in some sections of the labor market. This is not only true in countries with large informal economies such as Italy, but also resonates well with the needs of a bifurcated service sector labor market in “global cities” such as London and Frankfurt (Sassen 1991). This concept serves as a caveat against visions of a rejected migrant underclass, reminding us that even advanced postindustrial economies require individuals willing to carry out unattractive and poorly remunerated jobs. Even when such a need is not officially acknowledged, it tends to be unofficially accepted in a cavalier enforcement of migration regulation or in attempts to actively manage this labor market sector. By contrast, as Brucker ¨ et al. discussed in chapter 5, migrants may actually provide a source of support for European welfare systems. Not only are migrants often not eligible for transfer payments, they may make a net contribution. While the politics of retrenchment have been perhaps less severe in continental Europe than in Britain, with very uneven reforms by the New Right in the 1980s (Pierson 1994), some scholars contend that the future of European welfare states is nonetheless menaced by the public spending constraints imposed by European Monetary Union, the downward spiral and race to the bottom in social spending presumably implied by globalization, and last but not least the inverting demographic pyramid. Migration may help solve the conundrum of greater numbers of recipients of state pensions and health care services, yet shrinking proportions of active members of the labor force (Pierson 1998, 169ff.). The need for additional transfer benefits will arise to counteract old age poverty, partly caused by new social risks such as more unconventional family structure patterns and departures from lifelong stable full-time employment patterns and/or currently punitive mechanisms with regards to time away from labor market participation dedicated to childcare that cause gaps in pension eligibility. Migration may indeed provide a contribution – though certainly not a panacea – towards resolving Europe’s impending pension and social security crisis. While the base of contributors is shrinking and the

Migration and the welfare state

399

number of recipients keeps expanding, migrants to Europe are often young, healthy, and fertile. Without any migration, the population of the European Union (EU) is set to shrink by 17.3 percent to 311 million by 2050. However, important regional differences exist across Europe: While in a zero migration scenario, population decline by 2050 could be expected to reach 27.4% in Germany, 24.3% in Austria, 24.8% in Spain, 29.6% in Italy, 14.6% in Portugal, and 8.3% in Britain, in this scenario, non-migrant population is projected to increase by 1.1% in France and 26.5% in Ireland. The phenomenon of a declining and ageing population is not limited to the pre-2004 EU. In fact, economic and social insecurity and disintegrating systems of childcare and social assistance have led to even steeper declines in fertility across central and eastern Europe. Thus, projected decline for all of Europe (20.4) marginally exceeds the figure for the pre-2004 EU.2 To illustrate the demographic impact of the fall of communism the German case proves instructive: birth rates in East Germany that used to comfortably outpace West German rates throughout the mid-1970s and 1980s tumbled from an average of a lifetime total of 1.5 births per woman between fifteen and forty-five years of age in 1990 to 0.8 in the mid-1990s and have since remained below the West German rate of 1.4 (Der Spiegel 2004). The second wave of welfare reform, competition states and managed migration Following the 1980s wave of welfare state retrenchment, most pronounced in the United Kingdom, a second wave of welfare reform continued the restructuring and redrawing of labor and social market policy in the 1990s and 2000s under the leadership of the “reformed” Third Way Social Democratic parties across Europe (Blair’s New Labour, Jospin’s Socialists, Schroder’s ¨ New Center Social Democrats, Kok’s Purple Coalition). While critics accuse these policies of continuing the neoliberal logic of Thatcherism, in principle the target, discourse, and ambition of these reforms were not aimed at “rolling back the state” but rather at creating an “enabling” state that offered a “hand up, not a hand out.” Still, this embrace of active labor market policy, often influenced or even 2

Based on demographic factors, the argument to support migration so as to sustain welfare state systems appears quite powerful. But as Paul Demeny developed in chapter 2, two obvious caveats limit the extent to which migration can provide a panacea for the demographic challenge to the welfare state. First, migrants adopt the fertility rates of their host societies within one generation and will inevitably age themselves. Second, for migration to have a significant effect in the near term on European demographic profiles would require immigration on a scale that is hard to imagine politically.

400

Immigration and the Transformation of Europe

shaped by policy transfer from the United States (Daguerre and TaylorGooby 2004), combines carrots with sticks. The often mandatory character of job (re)training measures imposed on transfer payment recipients by Social Democratic state managers is presumably sweetened by the prospect of labor market integration. For migrants, the emphatic embrace of activating supply-side measures has been a very mixed blessing. Such measures are never tailored to accommodate migrants or resident ethnic minorities. Migrants are much more likely to be on the receiving end of the “sticks” implied by the logic of the second wave of welfare retrenchment of the 1990s: tighter eligibility criteria, lower absolute levels of transfer payments, more pressure and demands on the individual recipients. The embrace of active labor market policy by center-left governments in Britain, France, Germany, The Netherlands and Sweden can be seen as the reconfiguration of state activity and philosophical self-definition that obviously conditions public policy. Coloring these reforms is the logic of the competition state, described by Cerny (1997) as internalizing the demands of more internationally active and mobile capital, embracing marketization and openness, and pursuing policy in line with the purported realities of globalization. Active labor market policy, the embrace of “flexicurity,” and the pursuit of precarization under the auspices of labor market deregulation are all portrayed as reasserting Europe’s competitiveness vis-`a-vis North American and East Asian competitors. Raising labor market participation rates is a crucial component in ensuring more amiable business conditions and promoting such competitiveness, as expressed in the EU’s Lisbon Agenda. The competition state agenda views the struggle over attracting migrants and human resources as a new arena for competition with other potential locales of emigration (Menz 2002). But the 2000s approach of active migration management, informed by an economistic logic of heeding business demands for select labor migrants and the presumed failures of past humanitarian migration channels, dovetails with a more repressive policy towards migrants whose “employability” and labor market skills are more difficult to assess. Tight eligibility criteria are obvious limits to any “drain” migrants can possibly constitute. Indeed, Pierson (1998: 107) points out that as the provision of welfare services became a right of citizenship, rather than a threshold of exclusion from these rights for the recipient in the early twentieth century, governments were even more eager to impose some form of residency requirement on beneficiaries of social policy. Leibfried reminds us that up until the early 1990s, the submission of a claim for social assistance and thus the admission to a lack of self-sustainability constituted grounds for deportation from Germany, including EU nationals (2000: 196). To this day, the receipt of transfer payments in Europe is limited to

Migration and the welfare state

401

legal permanent residents, including citizens of other EU Member States, thus effectively barring “welfare shopping” or “tourism.” Welfare states as a migration magnet? It is not only tight limits on eligibility that undermine claims of welfare provision acting as migration incentives. Potential migrants’ level of information about the potential benefits offered by members of various welfare regimes can also not be assumed to be particularly detailed or sophisticated. Migrants arriving in Europe as part of the renaissant gastarbeiter programs or as part of family reunion schemes have to demonstrate financial self-sufficiency in order to qualify and can therefore not be assumed to be selecting countries of reception based on reputed levels of transfer payments. The only possible groups of migrants for which such considerations may be relevant are political refugees and asylum seekers. However, owing to institutionalized pan-European efforts to impede physical access for these groups, including carrier sanctions, safe third country regulations, and the bans on “asylum shopping” implicit in the Dublin Conventions, it is close to impossible for individuals arriving in Europe clandestinely to consciously select any given country of reception. Other chapters contained in this volume similarly negate any correlation between levels of transfer payments and levels of inward migration (see chapters 3, 5, and 11). Table 15.1 juxtaposes the gross number of migrants attracted between 1970 and 2000 with the type of welfare regime. Both the highest and the lowest levels of increase in relative terms are recorded by members of the Bismarckian group, while “generous” Scandinavian Sweden attracted fewer migrants than “miserly” liberal Ireland. While the small number of cases n renders any generalization methodologically problematic, there is no consistent correlation between the two variables “amount of inward migration” and “type of welfare regime.” Factors traditionally identified in the literature – cultural, linguistic, and colonial-historical ties, ethnic networks, and geographical proximity – play a much more important role than the generosity of a welfare system or the perception thereof. High levels of variation within such welfare families stress the idiosyncratic nature of migration incentives and further dilute any claims to a straightforward correlation. Managing migration in liberal welfare states: The case of the United Kingdom While the United Kingdom is still somewhat of a hybrid case – colored by remnants of the more generous postwar Beveridgean settlement, including notably universal health care, free at point of delivery – the neoliberal

402

Immigration and the Transformation of Europe

Table 15.1 Increase in total number of migrants in percent from 1970 to 2000, arranged by welfare state family

Increase in percent

Increase in absolute figures (in 1,000s)

Bismarckian: Austria Germany France Belgium Netherlands Luxembourg

314 168 3 22 396 104

587 4,738 1,067 199 1,315 100

Liberal/Anglo-Saxon: Ireland UK USA

84 28 167

180 1,083 25,286

Scandinavian: Denmark Sweden Finland

140 69 271

187 458 102

Mediterranean/Southern: Italy Spain Portugal Greece

67 185 99 394

721 886 131 444

Source: Figures from UN Population Division, World Population. Prospects: The 2000 Revision cited in Grimlat 2003.

retrenchment of the 1980s and early 1990s has moved the country closer to the ideal-typical liberal model. Esping-Andersen (1990) summarizes the principal components of this model as relying on “means-tested assistance, modest universal transfers, or modest social-insurance plans” and “entitlement rules are . . . strict and often associated with stigma; benefits are typically modest” (1990: 26). Legal migrants have full access to British welfare services in theory. Asylum seekers are granted the right to seek employment if no status decision has been made for six months, refugees that are granted “exceptional leave to remain” may also take up employment. During the period of asylum claim processing, claimants have the right to apply for subsistence and emergency accommodation. In 2003, of the new arrivals in the United Kingdom, 318,630 were students, 65,805 individuals were granted permanent residency rights as part of family reunion measures, 119,180 had been granted work permits, 49,405 were asylum seekers, and

Migration and the welfare state

403

11,340 refugees were granted exceptional leave to remain (author’s calculations from Great Britain Home Office 2004). Education and housing aside, transfer payments include unemployment benefits, income support, sickness benefits, state pension, housing benefits, and child and other family-related benefits. Impact of migrants on the welfare system in the United Kingdom Evidence from the 1999 Labour Force Survey (cited in Gott and Johnston 2002) indicates that distribution of claims submitted partly reflects the demographic profile of foreign-born residents who tend to be mainly of working age (20–65). Hence, claims for state pension and sickness benefits are less common among foreign-born residents than among their United Kingdom-born counterparts, with claim rates amounting to 7 versus 9% for pensions, and 5 versus 7% for sickness benefits, respectively. However, higher than average claims for income support (7 vs 5%), housing benefits (7 vs 5%) and child benefits (19 vs 16%) highlight the low socioeconomic status of some migrants; a higher than average rate of economic inactivity (30 vs 22%) and unemployment (6 vs 4%) point to problems in terms of labor market participation. An examination both of educational attainment levels and gross weekly pay levels reveals that while non-United Kingdom-born individuals broadly resemble the rest of the population, they are slightly more likely to have enjoyed tertiary education or to have received “no” or an unrecognized “other” education. While non-United Kingdom-born individuals are more likely than others to command wages in excess of £2,000 per month, more than half earn less than £1,200 monthly, indicating a certain dichotomy in socioeconomic status amongst migrants. Gott and Johnston (2002) calculate the total drain of migrants on benefits and state services as £28.8 billion for 1999–2000, including public expenditure on education and health care, social security, housing and other general public expenditure. However, this calculation is based on the costs imposed by the general non-United Kingdom-born resident population, constituting 8.4 percent of the total United Kingdom. Compared with a tax and social security contribution of £31.2 billion, this amounts to a net current contribution of £2.6 billion to public coffers. Another recent study calculates the net contribution to income tax revenue made by foreign-born United Kingdom residents as 10.2 percent of overall contributions, despite constituting only 8.7 percent of the population and earning on average 15 percent more in average wages than their United Kingdom-born counterparts. Extending Gott and Johnson’s study, the authors calculate that between 1999–2000 and 2003–2004 the

404

Immigration and the Transformation of Europe

fiscal contribution made by foreign-born individuals has increased from £105 annually for every £100 contributed by native-born residents to £112 (Sriskandarajah, Cooley, and Reed 2005). A certain bifurcation within the qualification and earnings profile of migrants is apparent in the United Kingdom, embodied by the London boroughs of Kensington and Chelsea on the one hand and Tower Hamlets on the other. Both boroughs claim high rates of non-United Kingdomborn residents, yet they stand at opposite extremes of the income distribution. While non-United Kingdom-born residents overall make a positive fiscal contribution, this masks the heterogeneous unequal income distribution amongst migrants and the consequent need for welfare transfer payments by those of lower socioeconomic status. Regional unemployment is particularly concentrated in northeastern England and parts of London (Office for National Statistics 2005), United Kingdom regions with high rates of migrant populations. While migration cannot be said to constitute a drain on the fairly minimal United Kingdom welfare system, further policy attention needs to be directed at promoting the labor market participation of migrants where needed and desired. The generally younger-than-average-age profile of non-native born United Kingdom residents implies limited claims to state pensions, a scenario likely to change slightly. The empirical evidence suggests that social expenditures on foreignborn individuals do not exceed the fiscal contributions made by this group in the United Kingdom. In some sense, the minimal level of welfare provisions offered and funding via taxation help ward off serious systematic challenges to the welfare system by migration. However, there is evidence of a socioeconomic two-tiered nature of migration, with migrants overall being somewhat overrepresented among recipients of transfer payments. Dealing with migration in a Bismarckian welfare regime: The German case The Bismarckian welfare state is fundamentally financed through employment-related charges and eligibility is primarily linked to (previous) employment status, with limited redistribution and limited provisions made for individuals with little or no employment history. The very nature of this welfare family makes it, therefore, particularly dependent on high employment rates. In 2003, a net sum of 102,696 migrants arrived in Germany, a total of 50,563 applications for asylum were received, 72,000 ethnic Germans from eastern Europe were accommodated, 15,542 Jewish

Migration and the welfare state

405

Russians arrived, 76,000 individuals entered as part of family reunion measures, and 2,285 actively recruited IT sector employees were admitted. The key avenue towards integration into the German welfare state is through employment and benefits that are available based on permanent residency status and steady (verfestigte) residence, rather than citizenship. Prior to a 1993 legislative change that significantly reduced levels of benefits and permitted payment “in kind,” social assistance was also available for asylum seekers. Thus, the great challenge for the Bismarckian system always lay in the integration of new migrants and resident non-Germans into the labor market. Not only did the pool of contributors need to remain stable or even increasing, but the number of non-active recipients of transfer payments also had to remain limited. While there can be little doubt that the country has benefited greatly from the contributions of the “first wave” of postwar migration that was generally composed of male, blue-collar, low-skilled southern Europeans, following the Anwerbestopp of 1973, the nature and composition of immigration has changed dramatically. The new migrants commonly arrived under the auspices of political asylum and family reunion. Resident migrants aged. While early migrant workers were generally employed, contributing to the systems of social security, and take-up of transfer payments was minimal, the picture had changed by the late 1980s. Between 1980 and 2003, the absolute number of foreigners claiming social assistance has risen from 70,523 to 624,472, representing an increase of non-Germans among recipients of social assistance from 8.3 to 22.3 percent. However, the absolute number of German welfare recipients has also risen dramatically from 780,629 to 2,142,740 (Bundesamt 2004). Impact of migrants on the welfare system in Germany The rapid and dramatic increase in non-German recipients of transfer payments since the 1980s is not only an expression of economic malaise, it also illustrates two worrying developments. First, many new immigrants did not manage to enter the labor market and remained reliant on social assistance, their lack of an employment history barring them from access to the more generous unemployment compensation. One might be tempted to suspect that this is particularly true of migrants entering through humanitarian channels such as family reunion and political asylum, an assumption supported by the rapid rise of asylum applicant numbers throughout the 1980s. However, data on welfare recipients do not distinguish between different categories of non-citizen recipients. Second, the second and third generation of migrants is commonly

406

Immigration and the Transformation of Europe

obtaining poor secondary, and no tertiary education, obstructing their entry into employment. Marshall (2001: 53) cites government figures according to which 20 percent of foreign students in 1994 left secondary school without any sort of degree, 44 percent with the basic Hauptschule degree, and only 9.5 percent managed the completion of all thirteen years and an Abitur (see also Maurice Crul’s examination of education and labor market participation for Turkish immigrants in chapter 9). Given the dearth of job opportunities for the unskilled, and an alleged structural shift towards a “knowledge economy” there is a serious risk of the abandonment of the second and third generation of descendants of migrants and labor market integration problems for poorly qualified newcomers. Females with a migration background face considerable challenges to meet the already low Bismarckian levels of female labor market participation. Thus, while labor market participation differs little between male German citizens (78.9%) and male non-Germans (77.6%), there is a significant gap for females (64.7/50.7%). Non-German citizens are also slightly overrepresented among the unemployed (OECD data cited in Harris/Coleman 2003). Incidentally, France, as another Bismarckian welfare state, shares very similar problems. Unemployment among non-EU citizens in France stood at 32.6 percent in 1994, more than twice the national average of 12 percent (Vaillant 1996), and has continued to remain at this unfavorable ratio since. While the earlier migrant generation is affected by the structural economic shift towards the predominance of the tertiary sector, newcomers outside of the carefully managed channels of new labor recruitment schemes often face difficulties in obtaining adequate language and professional skill training or in having their existing qualifications recognized. More disconcertingly, the educational system is not succeeding in offering the level of skill acquisition required for labor market integration to the German-born second- and third-generation descendants of migrants. Discrimination and racism among employers contribute to the problem. The potential challenge, especially by asylum seekers, has informed two legislative changes. One was informed by the ideology of deterrence, while the second more appropriately aimed at integration. The aim of the 1993 reform of the benefits for asylum seekers law (Asylbewerberleistungsgesetz) was to replace the previous provision of social assistance to asylum seekers with much less generous asylum seeker benefits, partly rendered in kind or through vouchers. The government also used temporary bans on employment as a deterrent towards asylum seekers in 1987, but discovered that this, in turn, imposed higher social assistance burdens in

Migration and the welfare state

407

1991, only to re-impose them in 1997. That same year another change to the original 1993 bill meant further cuts in absolute levels. Henceforth, total net expenditures for asylum seekers were reduced from €2.8 to 1.5 billion between 1994 and 2002. This 46 percent reduction in payment level compares with a 37.6 percent reduction in beneficiaries (Bundesamt 2004). More active integration measures were made with the 2005 Immigration Law. Policy-makers realized the importance of language skills, offering 4,579 German language classes at a cost of €17.5 million and €31.5 million on general integration projects at the local level (Bundesamt 2004). However, given that 45% of all participants in 2003 were Turkish, 73% female, and 45.6% registered their profession as being housewives, it would appear as though most beneficiaries of these courses are long-term foreign residents, rather than recent migrants. Some recent changes to labor and social policy have important ramifications on the welfare state–migration nexus. The 2005 abolition of the secondary tier of unemployment assistance by the Schroder ¨ government’s so-called Agenda 2010 means that once the temporarily limited eligibility for unemployment compensation expires, formerly employed migrants can claim social assistance only. Informed by the ideology of workfarism, this legislative package also permits the imposition of sanctions on social assistance recipients deemed not to “cooperate” in the search for employment or unwilling to accept the offer of temporary jobs remunerated at the hourly rate of one Euro (US $1.40). Since non-citizens are strongly represented among social assistance recipients, one might expect policy design to take the particular challenges of this group into consideration in imposing workfarist requirements to actively search for employment. However, this is not the case, and it remains to be seen to what extent re-training measures on offer will succeed in supporting the labor market (re)integration of non-citizens. Dealing with migration in the Scandinavian welfare regime: The case of Sweden The Social Democratic Scandinavian welfare states, exemplified by the paradigmatic example of Sweden, sought to “de-commodify” labor by offering very generous levels of state-sponsored welfare provision services based on a social contract that included a very high labor market participation rate, including notably by women. Such de-commodification entailed welfare “services and benefits . . . upgraded to levels commensurate with even the most discriminating tastes of the new middle classes” (Esping-Andersen 1990: 27).

408

Immigration and the Transformation of Europe

Swedish immigration has been much less pronounced than in the larger European states, with postwar migrants coming mainly from neighboring Nordic countries and Germany. Sweden’s migration regime has been described as “integrationist” (Castles 1989). It aimed at permanent migration, much of it informed by humanitarian concerns, rather than temporary labor migration. Most migrants arrived under the terms of a traditionally fairly accommodating asylum regime whose capacity was really only tested from the mid-1980s onwards. Whereas between 1950 and 1985 a total of 90,000 refugees were recognized in Sweden, between 1985 and 1991, 32,250 newcomers applied for political asylum, rising to 245,500 between 1995 and 1999 (Hammar 1985, Regeringskansliet 2001). The Aliens Act of 1989 drastically re-defined the asylum regulations in Sweden, limiting its scope and coverage. Since 1990, Sweden has experienced an average annual net migration surplus of about 20,000. Family reunion accounts for about 50 percent of all new arrivals (author’s calculations, based on Regeringskansliet 2002). The social contract between residents and the state with an implicit obligation to labor market participation obviously also extends to migrants. While the majority of non-European migrants entered Sweden through humanitarian channels, rather than as part of specifically targeted labor recruitment programs, low labor market participation rates among the non-native born population indicate shortcomings in the promotion of employment as a gateway to integration. Relatedly, these low rates also point to high public expenditures on transfer payments. Given the implicit expectation inherent in the welfare state design that work provides one central integrative avenue, the difference between the labor market participation rates for Swedish and non-Swedish males (78/63.1%) is striking and even more pronounced for females (74.2/60.3%) (OECD data in Harris and Coleman 2003). These rates mask the distinction between European and non-European migrants, with the latter commanding an average employment rate for both sexes of only 55 percent (Regeringskansliet 2003). As in other European countries, the labor market integration of second-generation migrants, and female migrants in particular, seems to encounter difficulties. Public policy-makers began to respond somewhat belatedly, but have recently sought to address the problems outlined above. In its amendments to the Aliens Act of 1997 and 2000, conditions for receipt of refugees have been clarified and the position of female migrants has been improved. The 1997 bill, entitled “Sweden, the future and diversity – from immigration policy to integration policy,” attempted to enact a number of integrative measures, laying the foundation for anti-discrimination legislation, offering training courses for new migrants, facilitating the

Migration and the welfare state

409

recognition of foreign qualifications, and attempting to discourage segregation in housing. The Swedish government remains committed to working towards the societal integration of asylum seekers from a very early age; its 2001 Schooling for Asylum-Seeking Children Bill (Asyls¨okande barns skolgng m.m.) explicitly clarified the right to free public schooling and childcare for the children of asylum seekers. This legislative act both re-affirms the basic integrationist philosophy of Swedish migration policy and demonstrates a commitment to pursue the path of societal integration through labor market integration. Impact of migrants on the welfare system in Sweden In Norway and especially Denmark, the far Right has successfully exploited “welfare state chauvinism.” The Danish 2002 Aliens Act has dramatically limited both eligibility for, and levels fertile grounds for the electoral fortunes of the Far Right, dramatically of welfare benefits for non-citizens or increasing the duration of residency as a requirement such rhetoric and policies have not (yet?) attracted electoral support in Sweden. The shortcomings in successful labor market integration and a comparatively generous accommodation of political asylum seekers do imply a cost, of course. Sweden’s Migration Board reception system accommodated 20,162 individuals in 2001 at an annual cost of about 1.5 billion SEK (about US $218 million). Asylum seekers may also claim a daily allowance of 71 SEK (US $10) and a 500 SEK (US $71) monthly housing allowance if living outside of the center, but only 24 SEK (US $3) if living inside (Regeringskansliet 2002). The 2003 government budget allocated 7.7 billion SEK for spending on refugees and asylum seekers or roughly 1 percent out of a total budget of 703.5 billion SEK (US $100 billion), but this figure pales against total spending on social services including pensions of 323 billion SEK. Sweden’s recently acquired role of an immigration country, attracting more asylum seekers and refugees and fewer Nordic and EU labor migrants, poses a significant challenge to the structure of its work-based welfare state to the extent that the integration of newcomers into the labor market does not proceed smoothly. Of course, the general increase in recipients of social assistance in the 1990s dramatically highlighted that the underlying assumption of near full employment was difficult to maintain during a severe recession. While numbers had surpassed the 100,000 barrier for the first time in the early 1990s, it had increased to 140,000 by the middle of the decade. There are some indications that the relatively recent immigrants have encountered much greater difficulty in successfully integrating into the labor market and hence were more likely

410

Immigration and the Transformation of Europe

to apply for assistance. By contrast, long-term foreign-born residents have about the same income level as the native born, according to government data (Regeringskansliet 2003). Sweden’s new migrant population faces the challenges of fulfiling the obligations implied by a work-based welfare state, reinforced by the shift towards active labor market policies in the late 1990s, while often struggling with inadequate language and professional skills. Thus, the Swedish government has initiated a series of (re-)training programs in 2000, aimed particularly at the young and the long term unemployed. These offers go along with disciplinary measures in the form of benefit cuts for individuals who do not actively search for employment and do so even beyond their home region and profession. However, despite much fanfare, active labor market policy has not delivered on its promise of returning 70 percent of all participants back to full employment and has not yet been rolled out to cover all of its potential target areas due to financial constraints. While the basic contract of the Swedish welfare state may hold, with the country boasting the highest labor market participation rates for either gender in Europe, the integration of immigrants, especially of recent arrivals has been more problematic. As in the German case, immigrants are falling into the remit of active labor market policies that seek to redress unemployment. Unlike the German case, migration-legislation reforms suggest that some of these challenges have been understood and measures are being put into place to address them. The Swedish government has realized that any such expectation presupposes an active helping hand and appears at least rhetorically willing to aid migrants in their social and labor market integration as well as underwrite its humanitarian asylum and refugee policy financially. Dealing with migration in the southern welfare regime: The Italian case The southern model of welfare is a curious blend of elements of liberal minimalism colored by conservativism. Transfer payments provided are generally only marginal, individuals rely on support from kin and private charities, often related to the Catholic Church. At the same time, just as in the Bismarckian system, pension and health care systems are generally linked to participation in the labor market. The role of the central state in welfare provision has thus been relatively insignificant, whereas private actors such as charities and employers play a predominant role, while local and regional governments have traditionally also provided limited social assistance.

Migration and the welfare state

411

Along with other Mediterranean states, Italy became a country of immigration only very recently. Though political asylum has been available in theory since the inception of the postwar Italian republic in 1946, it has never emerged as a major channel of immigration. Immigration from outside of Italy was practically non-existent until the late 1980s. The first significant numbers of immigrants to arrive came from former Italian colonies Somalia and Ethiopia, followed by arrivals from Albania and Yugoslavia. Only over the past few years have the sources of immigration become more diverse. Regions of origin now include the Middle East and East Asia (Iraq, China, Sri Lanka), the Indian subcontinent, the Maghreb, and sub-Saharan Africa, as well as Latin America (Peru, Brazil, Dominican Republic). Numerically, major source countries of immigration include Albania (142,066), China (60,075), the Philippines (65,353), Egypt (32,841), Yugoslavia (36,823), Morocco (159,599), Tunisia (45,680), and Romania (68,929) out of a total foreign population of 1,388,153 million, including some 151,798 EU citizens (absolute figures as of December 31, 2000, Italian Ministry of Interior). Italian authorities were relatively ill-prepared for hosting the arrivals. In light of the obvious need for some type of migration regulation, the first major piece of legislation was the 1986 legge 943, amended in 1990 by the so-called legge Martelli. Since then, important legislative milestones include the 1995 legge 489 and the 1998 Law on the Regulation of Immigration and the Living Conditions of Foreigners (Consolidation Act). While the 1986 law was modelled on the restrictive regimes put in place by France in Germany after the oil shock, along with cracking down on illegal immigrants, it did create annual quotas for legal immigration (Sciortino 1999). The 1990 law permitted for post-hoc French-style regularization. While both initiatives seem to suggest the contours of a relatively liberal asylum and migration policy (Contel and Di Biase 1999), it is important to consider a wide gap between theory and practice in the Italian case. At first glance it would appear as though Italy has been relatively successful in integrating its recent migrants. In fact, labor market participation among foreigners is slightly higher than among native Italians for both males (73.6/87.7 percent) and females (46.6/50.7 percent) (Harris and Coleman 2003)! However, these figures need to be understood in the context of the particularities of the Italian welfare state and migration regulation. Legal migration to Italy is limited almost exclusively to specifically recruited labor migration from outside the EU. The Italian government has assumed a much more restrictive stance, commencing with the 1998 Consolidation Act, and culminating with the Bossi-Fini Law of

412

Immigration and the Transformation of Europe

July 20, 2002 (legge: n.189). This law continues the trend towards discerning between “useful” labor migrants and undesirable refugees. Italy has negotiated bilateral labor contracts with those countries that have proved particularly cooperative either in preventing outward migration and/or accepting deported nationals. These countries include Albania, Tunisia, Morocco, and Romania. These “privileged countries” (paesi privelegati) in 2000 received annual quotas of 6,000 and 3,000 for the latter three countries respectively for “subordinate,” “autonomous,” and “insertion into the labor market.” In addition to these four countries, there was an annual quota of 45,000 workers for non-EU citizens in 2000, which has remained roughly steady since. In practice, it is close to impossible to be recognized as a political asylum seeker in Italy, and only under pressure to implement recent EU directives in this field has Italy started to design a specific regime of asylum accommodation and legislation. Also, by pursuing a two-tier strategy vis-`a-vis migrants – very restrictive towards undocumented “undesirable” aliens, more permissive towards migrants arriving legally and/or undocumented but gainfully employed, legal or not – in practice Italian migration policy seems to dovetail with a pan-European trend towards making such distinction. Another factor contributing to high labor market participation rates among non-native born in Italy is the recent character of the immigration phenomenon, thus permitting little possibility for policy failure with respect to a second or third generation. The almost exclusive actively-managed character of documented migration to Italy obviously also implies that welfare dependency of such individuals must remain minimal. The structure of the Italian welfare state has historically been centered around employment – it offered very generous pensions, but very modest unemployment benefits. Until the late 1990s, the receipt of these benefits used to be limited to 180 days and there was no nationwide system of social assistance, with some minimal provisions offered by regional and local governments and often religious charities. Health care, while free to the user, is limited to Italian (and EU) citizens. The 1997 Prodi reforms, along with the 2000 law on integrated intervention systems (legge 328/00) have sought to address Italy’s potential pensions crisis by curtailing eligibility, raising the retirement age, and lowering pay-out levels (Ferrera and Gualmini 2000; Fargion 2001). However, these measures, along with the establishment of a national system of means-tested unemployment compensation and social assistance have proven highly controversial. At the risk of caricaturing the Italian welfare system, it would appear fair to highlight its fixation on pensions with a state of underdevelopment in most other areas of social policy, such as housing, public education, personal

Migration and the welfare state

413

social services, and transfer payments. For migrants, the implications are clear: there are scant resources that can be drawn on, eligibility limitations aside. Impact of migrants on the welfare system in Italy With the level of welfare provisions being generally low with the exception of pensions and health care, and, most importantly, the character of documented immigration to Italy thus far, migrants necessarily make a positive contribution to fiscal revenues and cannot be said to constitute a burden. Despite the marginal character of its welfare state, Italy has been very successful at integrating its documented migrants into the labor market. In light of low fertility rates and a worrisome demographic scenario, the attraction Italy has exerted towards immigrants in the late 1990s seems to create a positive-sum equilibrium. However, the high rate of labor market participation is certainly also caused by a restrictive migration policy limiting access to state welfare provision by possessing an embryonic asylum and refugee policy. A substantial segment of Italy’s immigrants enter the country in unofficial fashion and/or enter the labor market outside of officially sanctioned channels, for example, by overstaying tourist visas. This secondary labor market is particularly pronounced in low-skill service sector activities including agriculture, tourism, and gastronomy and attracts significant interest from the Maghreb countries. While difficult to quantify for obvious reasons, this population of workers contributes directly or indirectly to fiscal revenues, but is barred from any access whatsoever to stateprovided welfare benefits. While the Berlusconi government of the early 2000s has proven eager to launch high profile “crackdowns” on visible forms of economic activity by undocumented migrants, especially in petty commerce and street prostitution, it appears less willing to alienate service sector employers by taking serious action on the gray economy. The state management of migration and welfare My two central hypotheses can now be assessed in light of the empirical evidence from the four case studies. A cross-national comparison reveals both commonalities and differences. The obvious caveat of eligibility limitations aside, as we have seen, (relatively) low levels of labor market participation and (relatively) high uptake rates of transfer payments are a common occurrence in three of the four cases. The exception to this pattern, Italy, is both a new immigration country and relying almost exclusively on managed

414

Immigration and the Transformation of Europe

actively-recruited migration. This pattern indicates problems in the labor market integration of migrants accepted for humanitarian reasons and the second-generation descendants of migrants. It reveals shortcomings in the educational facilities made available for immigrants, but is, of course, also correlated with deindustrialization processes. Official statistics on transfer payment receipt among the second generation and new arrivals, to the extent that they are available, also do not control for their lower socioeconomic profile, which is an important caveat to consider. Declining labor market participation is a potentially serious menace either to the employment/contribution-based Bismarckian system or the universal Scandinavian regime. It is partly for this reason that Germany and Sweden, along with the United Kingdom, have embraced so-called active labor market policy in the early 2000s, offering a combination of carrots and sticks to prod and cajole the recipients of social assistance into paid employment by offering theoretical (“life long learning”) or practical (“on the job”) training and poorly remunerated work placement, while imposing sanctions on individuals not willing to accept these obligations. A positive side-effect of this obsession with the labor market participation rates is the package of programs made available for asylum seekers and refugees in Sweden and to some extent the integrative measures now offered in Germany, while punitive components and programs not specifically designed for migrants do not appear promising in promoting labor market integration. The challenge of migration to the welfare state is particularly pronounced in an employment-based and financed welfare state. This hypothesis is supported most clearly by the German case, where dramatic increases in the number and proportion of non-native assistance recipients signal a rising cohort of non-contributors and indeed beneficiaries. As we have seen, German policy-makers have responded largely defensively in a number of ways, first by ending access rights for asylum seekers, and second by eliminating the secondary form of unemployment compensation. This means less reliance on the employment contributionbased component of the welfare state and higher dependence on social assistance, financed through general fiscal revenues. A somewhat milder challenge emerges in tax-financed welfare systems. However, where the level of services offered assumes high labor market participation rates, as in the case of the Scandinavian model, migration may become costly to the extent that it does not result in the standard high labor market participation rates of these countries. The liberal model deliberately offers low levels of social welfare services; hence high labor market participation rates are not as crucial to its sustainability. This was historically also true of the southern model. Italy’s more recent policies

Migration and the welfare state

415

indicate more encompassing structures, but its limited and carefully targeted labor recruitment programs have produced a highly active migrant population. It needs to be mentioned, however, that this “success story” comes at the expense of a henceforth incoherent asylum policy and on the back of a substantial unregulated segment of the labor market. The second hypothesis would seem to be most strongly supported by evidence from the German case, and to a lesser extent from the United Kingdom and Italy. By contrast, in Sweden the partial shift to a competition state rationale does not appear to have influenced migration policy design significantly. Active migration management seeks to steer migration flows, distinguishing between economically useful and burdensome migrants. Government in the three former cases have (re-)discovered the attractions of active labor market recruitment that offers to address labor market shortages, the demographic deficit and related social policy quagmires, and certain “push” and “pull” factors between sending and recipient countries. The return to this paradigm proceeds under the auspices of the embrace of business and investor-friendly competition state logic that seeks to “internalize globalization” (Soederberg, et al. 2005). While the perception of being in control is important in active migration management to public policy-makers, partly to placate populist xenophobia, subordinating migration management to the dictate of competition state primates of optimizing the conditions for investment entails, of course, a significant loss of control in policy-making autonomy. The undeniably uncomfortable specter of a disproportionate segment of transfer payment recipients being of non-native origin ought to motivate a serious assessment of past failures and shortcoming in promoting linguistic, labor market, and thus societal integration. However, such processes seem to be underway only in Sweden. Elsewhere it seems that public policy-makers have drawn a very different lesson from past failures. Evidence of failed migrants is interpreted to showcase the shortcomings of the “passive” migration management that followed the mid-1970s end to recruitment in Western Europe. In other words, the humanitarian channels of migration, including family reunion and political asylum, seem to have produced less “desirable” human resources than the postwar era of labor migration recruitment or post-hoc regularization. By contrast, active labor market recruitment, launched in competition to and possibly emulation of classic emigration countries, seems to produce more model migrants, as evidenced by the Italian case. The problem with deriving this conclusion from the data in public policy terms is the production of a false dichotomy between the “useful” and the “burdensome,” which is not only ethically questionable, it also incorrectly assumes low skill levels on the part of “humanitarian”

416

Immigration and the Transformation of Europe

migrants, without giving rise to active labor market integration programs for such groups or proactive measures aimed at the recognition of foreign professional qualifications and the fight against discrimination. State migration management is prone to interfere with existing migration flows, either by limiting, re-routing, criminalizing, or, in some cases, legalizing them. While difficult to quantify, undocumented labor migration and employment in unofficial segments of the labor market generates significant economic activity. It may contribute to indirect fiscal revenue, but, more likely, it will cause a net loss to welfare systems based on employment or direct taxation. Migration cannot be used as a panacea to the implications of the turning demographic pyramid, but neither does it constitute a nail in the coffin of the European welfare state. Policy-makers are beginning to realize that migration may pose a challenge to welfare regimes where it leads to high levels of social assistance receipt and low levels of labor market participation. However, the lessons learned seem to deny the benefits of humanitarian migration or realize past mistakes in integration efforts and concentrate on economically “useful” labor migration. Even on its own terms, such policy may prove unsuccessful and may well replicate the same problems the earlier gastarbeiter era produced.  Bommes, Michael and Andrew Geddes 2000 Immigration and Welfare: Challenging the Borders of the Welfare State. London: Routledge. Bonoli, Giuliano 2004 “Modernising Post-war Welfare States. Explaining diversity in patterns of adaptation to new social risks.” Unpublished manuscript. University of Fribourg, Fribourg. Bundesamt fur ¨ Migration und Fluchtlinge ¨ 2004 Migration. Nurnberg: ¨ Bundesamt fur ¨ Migration und Fluchtlinge. ¨ Castles, Stephen 1989 Migrant Workers and the Transformaton of Western Societies. Ithaca, NY: Center for International Studies, Cornell University. Cerny, Philip G. 1997 “Paradoxes of the Competition State: The Dynamics of Political Globalization.” Government and Opposition 32 (2) (Spring): 241–74. Contel, Michele and Rosaria De Biase 1999 “Italy,” in Asylum and Migration Policies in the European Union, Steffan Angenendt (ed). Berlin: Research Institute of the German Society for Foreign Affairs, 228–42. Daguerre, Anne and Peter Taylor-Gooby 2004 “Neglecting Europe: Explaining the Predominance of American Ideas in New Labour’s Welfare Policies since 1997.” Journal of European Social Policy 14(1) (February): 25–39. Der Spiegel 2004 “Land ohne Lachen” [A country without laughter] 2004, January 5, P 38. Esping-Andersen, Gøsta 1990 The Three Worlds of Welfare Capitalism. Cambridge: Polity Press.

Migration and the welfare state

417

Esping-Andersen, Gøsta 1996 Welfare States in Transition: National Adaptations in Global Economies. London: Sage Publications Inc. Fargion, V. 2001 “Italy: Moving from the Southern Model,” in International Social Policy: Welfare Regimes in the Developed World, Peter Alcock and Gary Craig (eds). Basingstoke: Palgrave, 183–202. Ferrera, Maurizio 1996 “The ‘Southern Model’ of Welfare in Social Europe.” Journal of European Social Policy 6(1): 17–37. Ferrera, Maurizio and Elisabetta Gualmini 2000 “Italy: Rescue from Without?” in Welfare and Work in the Open Economy: Diverse Responses to Common Challenges in Twelve Countries, Vivien A. Schmidt and Fritz W. Scharpf (eds). Oxford: Oxford University Press, 351–98. Freeman, Gary P 1986 “Migration and the Political Economy of the Welfare State.” Annals of the American Academy of Social and Political Sciences 485(1) (May): 51–63. Geddes, Andrew 2003 The Politics of Migration and Immigration in Europe. London: Sage. Gott, Ceri and Karl Johnston 2002 “The Migrant Population in the UK: Fiscal Effects.” Research Development and Statistics Directorate Occasional Paper, No 77. London: Home Office, Communications Development Office. www.homeoffice.gov.uk/rds/pdfs/occ77migrant.pdf. Great Britain Home Office 2004 Control of Immigration: Statistics, United Kingdom 203. London: Stationery Office. Grimlat, Joseph-Alfred 2003 “Des scenarios d’immigration pour une Europe viellissante [Immigration scenarios for an aging Europe].” Esprit (December): 92–101. Guiraudon, Virginie and Christian Joppke (eds) 2001 Controlling a New Migration World. New York: Routledge. Hammar, Tomas 1985 European Immigration Policy: A Comparative Study. New York: Cambridge University Press. Harris, Nigel and David Coleman 2003 “Does Britain Need More Immigrants? A Debate.” World Economics 4(2) (April–June): 57–102. Italian Ministry of the Interior 2000 “Rapporto Immigrazione (Report on Immigration).” Accessed April 5, 2004. Available from www.stranierinitalia.it. Lahav, Gallya 2004 Immigration and Politics in the New Europe: Reinventing Borders. Cambridge: Cambridge University Press. Leibfried, Stephan 2000 “Towards a European Welfare State?” in The Welfare State Reader, Christopher Pierson and Francis G. Castes (eds). Cambridge: Polity Press, 190–206. Marshall, Barbara 2001 The New Germany and Migration in Europe. Manchester: Manchester University Press. Menz, Georg 2002 “Patterns in EU Labour Immigration Policy: National Initiatives and European responses.” Journal of Ethnic and Migration Studies 28(4) (October 1): 723–42. Menz, Georg 2006 The Future of European Migration. Boulder, CO: Lynner Rienner. Office for National Statistics 2005 “Labour Force Survey.” 28/2005. London: Office for National Statistics.

418

Immigration and the Transformation of Europe

Pierson, Christopher 1998 Beyond the Welfare State? The New Political Economy of Welfare. Cambridge: Polity Press. Pierson, Paul 1994 Dismantling the Welfare State? Reagan, Thatcher, and The Politics of Retrenchment. Cambridge: Cambridge University Press. Piore, Michael J. 1979 Birds of Passage: Migrant Labor and Industrial Societies. Cambridge: Cambridge University Press. Regeringskansliet 2001 Sweden in 2000 – A Country of Migration: Past, Present, Future. Stockholm: Regeringskansliet. Regeringskansliet 2002 “Swedish Integration Policy for the 21st Century.” 2001/02:129. Stockholm: Regeringskansliet. June. Regeringskansliet 2003 “Sweden’s Action Plan against Poverty and Social Exclusion 2003–2005.” S2003:032. Stockholm: Regeringskansliet. July. Sassen, Saskia 1991 The Global City: New York, London, Tokyo. Princeton, NJ: Princeton University Press. Sciortino, Giuseppe 1999 “Planning in the Dark: The Evolution of Italian Immigration Control,” in Mechanisms of Immigration Control: A Comparative Analysis of Euroepan Regulation Policies, Grete Brochmann and Tomas Hammar (eds). Oxford: Berg, 233–60. Soederberg, Susanne, Georg Menz and Philip G. Cerny 2005 Internalizing Globalization: The Rise of Neoliberalism and the Decline of National Varieties of Capitalism. Baskingstoke: Palgrave. Sriskandarajah, Dhananjayan, Laurence Cooley and Howard Reed 2005 “Paying Their Way: The Fiscal Contribution of Immigrants in the UK.” London: Institute for Public Policy Research, April 27. www.ippr.org.uk/publications and reports/publication.asp?id=280. Vaillant, Emmanuel 1996 L’immigration. Toulouse: Milan.

16

The European Union dimension: Supranational integration, free movement of persons, and immigration politics Adam Luedtke1 Introduction

European integration (the transfer of policy-making authority from national governments to Brussels) presents a dilemma for national immigration authorities. The walls and borders that divided east and west have slowly crumbled and the lines between insiders and outsiders are increasingly blurred. The European Union (EU) has expanded its membership, and will eventually allow 500 million persons to move freely across its national borders. With little or no coordination on security and immigration issues, Europe will experience an unprecedented challenge. Can nation-states construct free-trade zones – allowing free movement of persons, services, and goods – without common immigration policies? Or is a common immigration policy the inevitable product of the functioning of regional economic cooperation, despite the national pressure to maintain domestic control over this sensitive issue? There is a third option. Europe’s states could try to avoid the explicit construction of strong common immigration policies – and such policies might not be strictly inevitable in functional terms – but they might come about through unintended institutional processes. Looking at the history of European integration, one sees that in other policy areas (such as gender equality or environmental protection), Member State governments did not originally anticipate the degree to which the EU’s central institutions (the European Commission, Court of Justice, and Parliament) would eventually gain policy-making authority as the EU evolved (Stone Sweet and Sandholtz 1998; Stone Sweet and Caporaso 1998; Cichowski 1998; Stone Sweet 2000; Stone Sweet et al. 2001). Thus, when the real possibility of EU-level immigration policy “harmonization” arose in the 1

The author wishes to thank the following sources for research support: The Research Fellowship Program of the German Marshall Fund of the United States, and the Chris Piening Memorial Scholarship, from the European Union Center at the University of Washington.

419

420

Immigration and the Transformation of Europe

run-up to the 1992 Maastricht Treaty, the Member States decided to tread very carefully. The EU’s involvement with immigration in the Maastricht Treaty was kept to a minimum, weak, “intergovernmental” bargain, whereby any one Member State could veto any proposed EU immigration policy, and such policies (if passed) would only be non-binding targets, as opposed to “hard,” judicially enforceable laws. This development makes sense, of course, given immigration’s high degree of political salience, and given the worries of national governments that electorates would punish them for relinquishing control over entry into national territory (a central facet of the Westphalian state). And yet events in the last decade have proven surprising. Despite open Member State worries over the possibility of the EU taking control over immigration policy, and despite conscious national attempts to head off EU control through institutional innovations and blockages, the draft European constitution made immigration an area of full EU control.2 In the meantime, the EU has also passed binding immigration laws in a variety of areas, which now commit national governments to implementing EU immigration policy. What factors caused this dramatic result, and what does this mean for the future rights and freedoms of the EU’s immigrants? The first section of this chapter will detail the evolution of the EU’s immigration regime, from Maastricht’s modest beginnings, to the ambitious European Constitution. Since “immigration policy” is a very broad topic, the second and third sections will focus on one key policy area, in order to consider these evolutions at an finer-grained level: the rights and freedoms of Europe’s nearly 20 million “third-country nationals” (TCNs), who are legally resident in an EU Member State, but do not hold citizenship in any Member State. Despite calls by the European Commission, the Parliament, and several of the Member States, TCNs do not possess the same rights as EU citizens to move freely and take up employment in any Member State. Therefore these immigrants, though legally resident in the EU, cannot participate fully in the common market. The second and third sections will detail the political struggle over granting free movement rights to TCNs, looking at the key roles played by the European Court of Justice and the European Commission in the face of strong opposition by powerful Member States. This political struggle will be analyzed in the context of the broader debate over the EU’s evolution, between the theories of “supranationalism” and “intergovernmentalism” 2

Though with one significant exception – Germany succeeded in inserting a clause into the draft Constitution stating that Member States shall retain responsibility for setting levels (numbers) of immigrants who can enter their country. Below I discuss where the debate stands given the failure of the Constitution.

EU immigration politics

421

(Stone Sweet and Sandholtz 1998; Moravcsik 1998). The empirical evidence for the analysis is based on secondary sources, and interviews I conducted with political actors in Brussels during 2003 and 2004. The key question I will address is whether Member States have been able to consciously dictate EU immigration measures, or whether EU political processes have moved immigration policy in an unforeseen direction. The fourth section concludes the chapter, analyzing the extent of TCN free movement rights as they currently stand, as well as their prospects of future expansion, in the broader perspective of the increasing degree of EU control over immigration policy in the twenty-first century. I aim to analyze whether recent debates over TCNs showcase a powerful logic of unintended consequences, and point the way towards further development of a centralized European immigration regime. The evolution of a European Union immigration regime Europe’s national governments are under constant political pressure on the issue of immigration. On the one hand, some European employers call for more immigrant workers, and human rights groups call for a higher level of immigrant rights and freedoms. On the other hand, however, electorates punish political parties deemed “soft” on immigration, and many voters turn towards radical Right parties, who openly advocate strong anti-immigration stances (Givens and Luedtke 2005). In some countries, radical Right parties have won governing power, while in others they simply push mainstream parties into taking more restrictive stances on immigration (Sniderman et al. 2002; Schain, Zolberg, and Hossay 2002; Givens 2002). Either way, anti-immigrant pressures have become an incredibly powerful motivator for Europe’s politicians. In such a highly-charged political situation, where national governments are often accused of “losing control” over immigration, one might not expect to see national governments willingly giving up what control they do have to the EU. This is especially so given that the European Commission (the EU’s quasi-executive branch) and the European Parliament have openly expressed pro-immigrant opinions, and have called for a more pro-active and egalitarian immigration policy, which respects human rights and serves the needs of business (European Commission 2000; European Parliament 2001). And yet this section will detail how the logic of European integration seemingly pushes towards EU control over immigration. Member States have apparently found it impossible to have a free-travel zone (and a common external border), without also agreeing on a common policy of who to let in to this zone, and what rights and freedoms they should enjoy once they are there. This is how

422

Immigration and the Transformation of Europe

immigration differs from other high-salience policy areas, such as defense and foreign policy. Immigration policy has a clear link with the common market and the free-travel zone. To put it bluntly, Italy’s frontiers are now Sweden’s frontiers. Of course, in the run-up to the 1992 Maastricht Treaty, Europe did not yet have a free-travel zone, and the single market existed mainly on paper. Yet even at this early stage, Member States had already seen limited advantages to coordinating and cooperating on immigration policy, though keeping the final say for each national government. As early as 1976, a resolution by the EU’s Council of Ministers (the legislative body representing national governments) encouraged Member States to develop common immigration policies, in consultation with the Commission (Papademetriou 1996). Economic malaise and anti-immigrant politics throughout the 1970s seemed to prevent any further movement towards a common policy, but in the mid-1980s, as both economic growth and European integration gathered speed, the Commission set forth a series of “Guidelines for a Community Policy on Migration” (European Commission 1985). These guidelines not only emphasized the importance of free movement for EU citizens (which was by now a right well-established in EU law, though little-used by EU workers), but also called for “equality of treatment in living and working conditions for all migrants, whatever their origins” (cited in Papademetriou 1996: 20). In other words, the Commission called for TCNs to be given free movement rights, commensurate with EU citizens. In the same year (1985), the Commission also released a Decision that required Member States to provide advance notice of any measures they intended to take towards TCNs in the areas of entry, residence and employment. However, even this small, non-binding step was unwelcome as far as some Member States were concerned, and they took their objections to the European Court of Justice (ECJ), arguing that the Commission had over-stepped its authority by seeking to play a role in immigration policy. The ECJ reached a compromise decision; voiding parts of the Decision, but letting the remainder stand (Papademetriou 1996). This political struggle over immigration policy-making authority foreshadowed the much larger struggles a decade later. In 1986, the beginnings of the single market became reality, with the Single European Act’s declaration of the “four freedoms”: freedom of movement for goods, services, capital, and (most importantly for our subject), persons. Predictably, the Commission interpreted the category of “persons” expansively, arguing that it should cover legally-resident TCNs. However, Member States argued that it should only cover EU citizens. Worried about the Commission’s expansive position, Member

EU immigration politics

423

States adopted a declaration affirming their right to control immigration policy, under the Single European Act (Papademetriou 1996). The following years saw Member States cooperating only minimally on immigration, using intergovernmental fora outside of the EU’s institutional structure, without binding national governments to any “hard” commitments. But the early 1990s brought new opportunities for EU control, during the intergovernmental conference leading up to the Maastricht Treaty. Given that free movement for EU citizens would soon become a reality, the Commission, the European Parliament, and some Member States argued that the existing structure of immigration cooperation was both inefficient (a multitude of overlapping bodies producing no binding commitments) and undemocratic (meeting behind closed doors, with no parliamentary and judicial oversight). These same actors called for immigration to be brought under EU jurisdiction, to make it more open, democratic and effective (Hix 1995). Heeding Member State reluctance, however, the Luxembourg Presidency of the EU in 1991 proposed an institutional structure that would keep foreign/defense policy and immigration as separate “pillars” of the Community, outside of the control of the Commission, Parliament and Court of Justice. This draft was met with resistance by several of the more pro-EU Member States, one of which, The Netherlands, submitted an alternate draft Treaty that brought immigration policy under EU control. But this draft was defeated by the more skeptical Member States, including the United Kingdom, Ireland, Denmark, Greece, and France. The resulting Maastricht Treaty thus prevented immigration policy from becoming “supranationalized” in four respects: r it allowed Member States the right of initiative to propose new EU-level measures (in “normal” EU decision-making, it is only the Commission who can propose new measures); r it allowed the Parliament only the right to be “consulted” over decisions, but gave it no veto or amendment power; r it prevented the ECJ from having legal jurisdiction over immigration; and r it allowed any Member State to veto a proposed measure. Thus, the Maastricht Treaty was a victory for intergovernmentalism, in that it did little to change the already-existing structures for immigration cooperation (Papademetriou 1996). Five years later, however, the debate was re-opened. In the run-up to the 1997 Amsterdam Treaty, three key developments pushed immigration cooperation back onto the agenda. First, the Amsterdam Treaty incorporated the Schengen agreement (on free travel) into the EU’s institutional structure. Thus, external borders would become common

424

Immigration and the Transformation of Europe

borders, obviously lending new salience to immigration cooperation. Second, the modest cooperation already achieved on immigration (e.g., agreements over common standards on political asylum, to prevent “asylum-shopping” among Member States), was seen by national governments as a success, in that it allowed them to crack down on immigration at the EU level, where they were relatively free of pressure by proimmigrant NGOs and courts (Guiraudon 2000; Guiraudon and Lahav 2000; Givens and Luedtke 2004). Increasing problems with illegal immigration and rising numbers of political asylum seekers throughout the 1990s gave these issues even more pressing salience. And finally, most participants agreed that the “Third Pillar” structure was relatively inefficient, given the plethora of intergovernmental groups that lacked the power to forge binding commitments (Geddes 2000). These factors pushed all but the most reluctant Member States (the United Kingdom, Ireland, and Denmark) to re-think their opposition to EU control. Thus, the Amsterdam Treaty achieved a partial supranationalization of immigration policy authority. It was agreed that after five years, the Commission would gain the sole right of initiative, the Parliament would gain the power of “co-decision,” the unanimity requirement (national veto) in the Council would disappear, and decisions would thus be taken by a majority vote. It was also agreed to give the European Court of Justice (ECJ) jurisdiction over immigration, though with a special exception, in that only high courts could refer cases to the ECJ. Since the three most skeptical Member States (the United Kingdom, Denmark, and Ireland) were not participants in the Schengen agreement, they were allowed to “opt out” of this new decision-making structure, thus allowing them to drop their objections to supranationalization (Geddes 2000). As the five-year transition period neared its close, it was overshadowed by the Convention on the Future of Europe and the resulting draft Constitution. Secure in the knowledge that EU control had allowed them to be “tough” on immigration over the past five years by passing highly restrictive measures (most notably the various steps to reduce the number of asylum seekers, which even the United Kingdom had embraced!), Member States surprisingly agreed to further expand EU control. Not only would the Commission get the sole right of initiative, the Parliament get co-decision, and the Member States lose the national veto (as agreed at Amsterdam, after five years), but the ECJ was also given full jurisdiction over immigration (in that any national court could request an ECJ ruling). The only full (opting-in) participant to openly express strong skepticism was Germany, who succeeded in inserting a compromise clause into the Constitution that Member States shall retain the ultimate control over determining the quantitative levels of immigration

EU immigration politics

425

in their countries. This clause tempered the worries of other states which had previously opposed EU control. It now looks like the draft Constitution has failed, however, and in the meantime, the five-year transition period has ended. Accordingly, Member State governments agreed in late 2004 on a blueprint for EU immigration policy, the Hague Programme, which will lay the groundwork until the eventual passage of a new treaty. In recent interviews with political actors in Brussels, I learned that the Hague Programme is seen by many as a setback for supranationalism, in that it prevents the ECJ from having full jurisdiction over immigration (again, only national high courts can refer cases), and, more importantly, it moves all areas of immigration policy-making to co-decision and majority voting, except for one key area: legal migration. In other words, other areas of immigration policy (most notably illegal immigration and political asylum) will become “normal” EU policy areas, with full EU control, but control over legal migration will retain aspects of intergovernmentalism (unanimity voting in the Council, and no co-decision power for the European Parliament). Thus, EU immigration policy faces two possible futures. If the draft Constitution (or a potential revised version) passes, then full EU control will become a reality (though with the caveat that Member States can set numerical levels of immigrants). But if the Hague Programme continues to be the guiding structure, then national governments will have retained control over the most sensitive area of immigration: legal migration. Illegal immigration and political asylum, while sensitive in their own right, raise less controversy when delegated to Brussels, because they are seen in terms of restriction and control; that is, how can the EU help to reduce the numbers? Legal migration is more sensitive for national governments because these are decisions over who to let in. Member State governments have few problems with using the EU as a way to coordinate exclusion; coordinating inclusion is a different matter altogether. What does the evolution of EU control over immigration policy mean for the rights and freedoms of Europe’s almost 20 million third-country nationals? Will TCNs gain the right to participate in the EU’s single market, by moving freely across borders to take up employment? This issue offers an illuminating test case for analyzing the possibilities and effects of EU control, because the battle lines are clearly drawn. The European Commission, Parliament, and a small group of Member States want TCNs to have free movement rights, while a larger and more powerful group of Member States is opposed to taking this step. Thus, a focus on this area can tell us whether states have been able to consciously dictate EU immigration measures, or whether EU political processes have moved immigration policy in a new direction.

426

Immigration and the Transformation of Europe

The next two sections analyze the political struggle over TCN freemovement rights, beginning with the role of the European Court of Justice, and moving to the Commission. Both sections will be analyzed in the context of the debate between supranationalism and intergovernmentalism. For supranationalists, the EU tends to gain increasing power despite the intentions of Member States (Stone Sweet and Sandholtz 1998), while for intergovernmentalists, Member States always retain the final say (Moravcsik 1998). Given that the European Commission, Parliament, and Court of Justice have advocated the expansion of immigrant rights and freedoms, while most national governments have preferred a restrictive line, the debate between supranationalism and intergovernmentalism is of direct importance to the future of Europe’s TCNs. Third-country nationals and the European Court of Justice Until recently, developments in EU immigration policy seem to have provided strong verification for intergovernmentalism, especially because the Member States consciously excluded the ECJ (normally the vanguard of supranationalism) from jurisdiction over a key policy area. This is the issue of legally-resident TCNs: workers that do not possess free movement rights in the EU, due to not having citizenship in their Member State of residence. Many analysts were skeptical about TCNs ever gaining free-movement rights, despite the fact that the European Parliament, Commission, and some of the more federally-minded Member States consistently called for the granting of these rights throughout the 1990s (Papademetriou 1996). This is because the Parliament and the Commission both lacked the political clout to get such rights past the Council, who still voted by the unanimity rule where immigration is concerned. Thus, it seemed that the ECJ was the only institution that could grant TCN rights, due to the special political status of European law on the free movement of workers, and of a legacy of successful rights claims within this law. However, the institutional changes after Amsterdam (detailed above) gave the Commission more clout in the area of immigration, resulting in the passage of a directive in 2004 that gives TCNs a limited form of free-movement rights. This seemed to make the Court’s role less crucial, and to enhance the Commission’s ability to expand immigrant rights, though the ECJ may still play a vital role in interpreting the new rules. Looking at TCN cases that have already come before the Court, as well as the views of other analysts, I will now highlight some key characteristics of immigration policy which prevent the ECJ from gaining as much

EU immigration politics

427

jurisdiction over it as it has over other policy areas. After covering the ECJ’s role, I will then analyze the European Commission’s power and the new “Long-Term Residents Directive,” a piece of binding EU law passed in 2004, to determine its effect on TCN free-movement rights. These two sections will attempt to argue that TCN rights have been expanded (or the institutional groundwork laid for future expansion), against the express wishes of most of the Member States. Since the institutional arrangements for making immigration policy in Brussels are currently unclear, given the confusion surrounding the European Constitution, I will close by pointing to some initial signs that show promise for increased future oversight by the Court of Justice and the Commission. Undeniably, the ECJ has gained some jurisdiction over the rights of TCN immigrants, through its power to legislate on the free movement of workers. Free movement policy (for EU nationals) is by now strongly “supranationalized,” with EU jurisdiction. This is a fait accompli. However, all EU immigration policy vis-`a-vis legal immigrants, which does not deal with the movement of EU nationals (those who are citizens of Member States), is still decided in a quasi-intergovernmental fashion, meaning that the EU’s central organizations (the EP, the Commission and ECJ) have little “competence” over this area. However, in the words of Geddes (2000), “this does not mean that the two [free movement and immigration] are disconnected and can be analyzed separately” (32). They are inextricably linked in the political realm, because, paradoxically, “freer movement for European Union citizens has brought with it tighter controls on movement by non-European Union citizens [TCNs]” (32). At the same time, “free movement for people has . . . created legal and political sources of power and authority [like the ECJ] that have implications for TCNs” (43). In other words, by giving the Court the power to regulate free movement, the Member States unintentionally granted the Court a limited say over immigration policy as a whole. I will now give evidence of the Court making use of this role, by building a legal basis to potentially grant TCNs free movement rights against the wishes of national politicians. If this happens, then we can say that “spillover” has occurred. Spillover is a key causal mechanism of supranationalization, whereby the EU expands its power into new areas, unforeseen by the national governments who initially empowered it. In 1992, the Single European Act’s ambitious goals on free movement came to concrete fruition in the Maastricht Treaty, with immigration matters sidelined in its “Third Pillar,” as detailed above. Again, the Third Pillar was seen as widening the EU’s “democratic deficit” by handing more power to Member State executives (ministries that deal with immigration), who were to have no judicial oversight. Overall, the Maastricht

428

Immigration and the Transformation of Europe

process strongly followed the lines expected by intergovernmentalist analyses, and the Court played the very role that its Member State principals would have wanted, striking down the Commission’s only two regulations on immigration, both of which dealt with harmonized visa policy (Geddes 2000: 100). Thus, at this point there seemed to be little sign of independent influence from supranational actors, as EU institutions were locked out of the process and the Court made sure to reinforce this state of affairs. The Amsterdam Treaty changed things only marginally vis-`a-vis ECJ jurisdiction, with its compromise decision whereby the ECJ was only given jurisdiction when national courts of “final instance” requested rulings (Guild 1998). And again, the 2004 Hague Programme retained this restriction. What does this mean for eventual ECJ jurisdiction over TCN matters? In other policy areas, the ECJ’s growing jurisdiction has depended on using the law itself as a shield and a mask, relying on the isolation of the legal realm’s “technical discourse” to overcome Member State reluctance, which in turn depends upon the consistency and coherency of legal reasoning and the EU’s evolving case law (Mattli and Slaughter 1998). The mechanism of “direct effect,” which allows any national court to request an ECJ ruling, is the key component in this process (Weiler 1994). Aware of this requirement, the Member States (consistent with intergovernmentalism) appeased Britain and Denmark by denying full direct effect on TCN matters. “Therefore coherence will be much slower dependent first on the adoption of measures which will regulate TCNs . . . and it will take much longer for interpretive questions to reach the Luxembourg Court” (Guild 1998: 619). That is, if spillover is to happen, it may be a longer, more drawn-out process than with other areas of European law. In fact, in the six years since Amsterdam, only two cases have been referred to the ECJ, and both were ruled inadmissible because they did not come from final courts. Further, Amsterdam explicitly denied the ECJ jurisdiction in all areas of border controls (including TCN entry) that “relate to the maintenance of law and order and the safeguarding of internal security” (Monar 1998: 141), which may well be interpreted to cover all areas of immigration policy in the current political climate. This might seem like an additional victory for intergovernmentalism, but the ECJ actually has the power, under Amsterdam, to determine for itself the criteria for “law and order” and “internal security,” meaning that the Court will likely not rule itself out of the picture. This, of course, seems more consistent with supranationalism. The legal coherence and consistency problem, however, remains an obstacle for supranationalization. “Coherence,” in this case, means no

EU immigration politics

429

national discretion over legal matters, since national discretion is an inherent obstacle to integration. Under the Amsterdam Treaty, TCNs now have the coherent, consistent right to short-term travel across the EU’s internal borders without frontier checks. However, there is a clear lack of rights, protections and freedoms in three other areas of TCN travel, which are still subject to full Member State discretion (with no “floor” of standards/procedures that could pave the way for rights-claiming). These are: visa-free travel, standards for the issuing of visas, and external border checks. In these three areas, Member States are free from the ECJ’s watchful eyes, and can apply the utmost national discretion in their policies. TCNs who feel that they are unjustly required to have visas, are unjustly denied visas, or are abused at the EU’s external borders have no formal recourse to the ECJ (Guild 1998: 617). This state of affairs pushes towards intergovernmentalism, since it maximizes the ability of national politicians to capitalize on the public’s fears over immigration as a national security threat. But how did the supranational institutions attempt to counter this marginalization? The European Parliament, keen to extend ECJ jurisdiction (as well as its own competence) challenged national discretion over visa policy in the Court of Justice itself, and lost (Parliament vs Council [1997] cited in Guild: 617). This seemingly confirms the hypothesis of Garrett, Kelemen and Schulz (1998), who argue that the Court is reluctant to overstep its delineated role on issues where it feels it could lose political legitimacy in the eyes of its Member State creators. Thus, TCNs remain only the “objects” of EU policy and law, and cannot be “actors” in their own legal right since they have no right of direct effect (bringing ECJ-justiciable claims to national courts). Under intergovernmentalism, there is no legal framework, which in turn means no legal coherence or consistency, which in turn means that law cannot function as mask or shield to everyday politics. Indeed, there is not even a clear, stated, guiding goal for TCN policy in the Treaty (unlike with free movement), so the Court has no guiding principles to invoke in future rulings. In the absence of guiding principles and legal coherence, there would seem to be scant possibility for spillover. Intergovernmentalism, again, seems to be confirmed by the Member States retaining discretion in order to minimize domestic political costs. However, it was mentioned earlier that there are some openings that might allow for spillover by letting the ECJ rule on TCNs, against Member State wishes, through the mechanism of free movement law. The first example of these openings regards TCN family members of EU nationals who exercise their freedom of movement in another Member State. According to the ECJ, TCN family members of EU citizens are entitled to the same residence, work, and welfare rights as Member State

430

Immigration and the Transformation of Europe

nationals. However, this right is not directly held by the foreign family member as a “human” right, but is instead derived from the rights of EU-national economic migrants, meaning that the ECJ has intentionally limited the legal standing of TCNs to economic matters (Guiraudon 1998). This, again, might confirm Garrett et al.’s self-limiting hypothesis, despite the fact that it is a small opening for TCN spillover. Another possible expansion of immigrant rights, however, comes with the issue of same-sex spouses. I interviewed several officials who mentioned that free movement rights could also be extended to TCN same-sex partners of EU citizens, against the wishes of many Member States, if the ECJ interprets “spouse” to include same-sex partners. While covering only a small number of immigrants, this is an example of how supranational control can move policy away from national preferences. One British official admitted to me that ECJ competence over immigration and asylum is a “big worry” for the United Kingdom, because of the tendency of the judiciary to exert its political independence. Another opening, however, has provided additional potential TCN legal spillover. The TCN rights in this case are also derived from economics, but are potentially farther-reaching than those applying to family members. This opening concerns the status of TCN workers who are employed by EU firms performing services in another Member State. In the Rush Portugesa decision,3 the ECJ found that if some of a company’s employees are TCNs, “Member States cannot refuse them entry to protect their own labor market on the grounds that immigration from nonEuropean Union states is a matter of national sovereignty” (Guiraudon 1998). In other words, the “four freedoms” (labor, capital, goods, services) appear to trump national discretion over immigration policy, meaning that TCNs gain rights if they are included in the protective cradle of EU economics (being employed by EU firms). Again, however, TCNs have no direct rights under Rush Portugesa, but instead have indirect rights that stem from discrimination against their employers, not against them. Being that this decision came against Member State wishes, however, it seems a small but significant step towards legal spillover where TCNs are concerned (Guiraudon 1998). The third opening that should be mentioned is perhaps the farthestreaching one in terms of the potential for ECJ “activism” to bring TCNs into the EU fold. This concerns the legal status of “Association Agreements” between the EU and third countries, which provided a quasi-legal framework for supranational immigration policy. “These Agreements have direct effect and have established rights of free 3

Rush Portugesa Ltda v office Nationale d’Immigration C-11 237/91 (1990) ECR I-3461.

EU immigration politics

431

movement and the transferability of social entitlements for some TCNs,” applying to large numbers of Turkish and North African migrant workers (Geddes 2000: 149). In other words, TCNs from these associated countries must be treated equally to native workers. Guiraudon (1998) argues that ECJ “activism” allowed for some legal spillover in this area, against Member State wishes, since “member states clearly intended association law to be incomplete in the sense that no individual rights could be inferred” (Hailbronner and Polakiewicz 1992: 57). The ECJ was therefore activist, contrary to intense political pressure from the Member States (and thus contrary to the self-limiting hypothesis of Garrett et al.), because it ruled that “nationals of associated contracting states had directly enforceable rights that . . . had to be upheld by national courts” (Guiraudon 1998: 660). In the 1987 Demirel case, the ECJ granted itself jurisdiction over the entry and residence of Association workers. In the Sevince case,4 the ECJ went even further by arguing that a “right of residence” could be inferred from Association Agreements, since the already-established rights of employment would be “useless” without rights of residence (Guiraudon 1998: 660). And in the Kziber case,5 the Court went further still, by vindicating the application of a Moroccan living in Belgium for unemployment benefits, effectively bringing TCNs into the cradle of “social Europe.” These rulings seem to showcase a growing level of supranational influence, because the Court decided to ignore the difference between EU nationals and TCNs; “judges applied the principles of Community law rather than the limited framework of Association law . . . the Court ignored principles such as reciprocity: there are no unemployment benefits in Morocco” (Guiraudon 1998: 660). These rulings made the Member States “furious,” according to Guiraudon, and even pro-EU observers found the ECJ’s legal reasoning here to be “dubious” (660). Was there a loss of legitimacy for the Court, and did this cause the Court to retreat, as Garrett’s hypothesis might suggest? The “judicial capital . . . which is involved each time that a court breaks with the past and makes a new development” (Weiler 1993) may indeed have eventually seemed “prohibitive” for the Court. In a 1995 case, (cited in Guiraudon 1998) the ECJ retreated from its earlier activism and ruled against the plaintiff, a Turk, whose incapacity to work led to his residence permit not being renewed (Guiraudon 1998: 660). And since then, the Member States have been careful to exclude the right of free movement from new Association Agreements with eastern European countries, perhaps 4 5

Sevince v Staatssevetaris van Justitue, C-237/91 (1990) ECR I-3461. Office National de L’Emploi v Kziber, C-18/90 (1991) ECR 1-199.

432

Immigration and the Transformation of Europe

rendering the Court’s early activism a moot point (though also supporting the notion that they recognize the Court as a powerful actor to be wary of). However, Geddes (2000) has pointed to the existence of a significant political opportunity here, emphasizing the fact that NGOs have “argued that if these rights are extended to some TCNs because of relevant Association Agreements, then it is unsustainable for other legally resident TCNs to be excluded” (149). Indeed, the fairness of this arrangement might be called into question by political actors, and this “leads to a free movement dynamic with potential spillover effects” (149). It seems that the Court, while hesitant for the moment, has preserved itself a future space whereby it could extend Association-derived rights to all TCNs by virtue of the principles of legal coherence and consistency, not to mention normative fairness. Whether or not it acts in this area will be a good future test of the intergovernmentalism/supranationalism debate. An additional political opening should be mentioned here, and this is the Treaty provision prohibiting discrimination on a wide variety of grounds, including race (but, notably, not nationality). Guild (1998) argues that this anti-discrimination principle (as a guiding norm) “is capable of providing a further justification . . . for assimilation of the position of legally resident third country nationals to the position of their Member State national colleagues” (619). What would be the legal logic? The Court may in the future rule that treatment of TCNs “may need to be equivalent to that of Member State nationals if it is to avoid the risk of being challenged as . . . discriminatory on the basis of race” (619). This would admittedly involve a rather large leap in legal reasoning by the Court, but the fact that it holds competence over the anti-discrimination provision means that it might in the future take the activist path, Member State objections or not. The Commission and the Long-Term Residents Directive As stated above, throughout the past decade the European Commission consistently advocated for TCNs to have free movement rights. But given the intense political salience of the issue, several Member States were reluctant to take this step. And because of the institutional mechanisms of the Third Pillar (unanimity voting in the Council) it seemed unlikely that the objections of these Member States could be overcome, in potentially drafting a new directive to let TCNs participate in the single market. Thus, all pro-immigrant eyes initially turned to the ECJ as a potential force for change. While this potential has not yet been fully realized, new hope for immigrants and their advocates came from the direction of the

EU immigration politics

433

Commission in 2003, in the form of the so-called “Long-Term Residents Directive” (LTRD). This section will analyze the political evolution of the LTRD, with a view to whether it is an example of successful harmonization, and whether it is (currently or potentially) expansive towards immigrant rights in a way that diverges from national preferences. As stated before, the Commission has been a consistent defender of TCN rights. At the Tampere Council in 1999, the Commission pushed the Member States into declaring their agreement with the principle of equality between (legally-resident) TCNs and EU citizens. This is an example of supranational leadership, in which the Commission used its role as agenda-setter to reach agreement on broad (but not yet binding) policy goals (Pollack 1998; Uçarer 2001; Hooghe 2001). The Tampere conclusions read: The European Union must ensure fair treatment of third country nationals who reside legally . . . A more vigorous integration policy should aim at granting them rights and obligations comparable to those of European Union citizens. The legal status of third country nationals should be approximated to that of Member States’ nationals. A person, who has resided legally in a Member State . . . should be granted in that Member State a set of uniform rights which are as near as possible to those enjoyed by European Union citizens. (European Council 1999)

Taking these conclusions as its mandate, the Commission then proposed an ambitious directive in 2001 that would not only harmonize the status and rights of LTR TCNs across the EU, but would also grant these recognized LTRs roughly the same free movement rights as EU citizens. Under the guidelines of the proposed directive, TCNs would be able to move freely from their Member State of residence to a second Member State, as long as they were taking up work, study or vocational training (or if they had adequate financial resources to support themselves). Further, they would not lose their status as a “worker” (and thus not be deportable), even if they had an illness, accident, took up vocational training, or lost their job. In the latter case, they would even have a guarantee of unemployment benefits. On the restrictive side, however, the proposed directive did mandate that TCNs would have to apply for a residence permit in the second Member State within three months of moving there, and could be required to present proof of employment or resources, and could also be expelled for reasons of “public health,” “public policy,” or “domestic security.” However, if the second Member State expelled the TCN, (s)he would have the right to return to the first Member State of legal residence. Predictably, this proposal was attacked from two sides. Some Member States felt that it was too generous and expansive, while some

434

Immigration and the Transformation of Europe

pro-immigrant groups felt that it was too restrictive towards TCN rights, and did not go far enough to equalize their status with EU citizens. This political conflict provides a good test of supranationalism, because the Commission expressed a commitment to defending the principle of TCN equality, against the preferences of some of the more powerful Member States. Thus, I will now evaluate the two-year political battle that raged in the Council over the Long-Term Residents Directive, to determine if the outcome deviated (or may deviate in the future) from national preferences in any meaningful way. How much did the final Directive, adopted in January 2004, diverge from the Commission’s proposal (in terms of restricting TCN rights), and do TCNs now have the right of free movement, against Member State wishes? Pro-immigrant groups had already attacked the proposed 2001 LTRD for being too restrictive towards TCNs. The United Kingdom-based group Statewatch fretted that the proposed LTRD did not explicitly refer to a right to take up economic or non-economic activities under the same conditions as EU nationals. Instead, it created a separate regime with less-clear legal protections. Statewatch argued for making explicit reference to legislation on free movement rights of EU citizens: “at present, while such a right could possibly be inferred from the proposed Directive, it could also be argued that it is not included in the absence of express wording. The best way forward would be to incorporate references to legislation on the free movement rights of EC nationals in place of a separate regime for long-term residents” (Statewatch 2003). The United Kingdom-based Immigration Law Practitioners’ Association raised similar objections, in that they recognized the right to reside in a second Member State, but found the lack of any express reference to the right to undertake economic or non-economic activities in another Member State (Immigration Law Practitioners’ Association 2003). The ILPA also raised potential legal problems with problematic language in the Directive (which could lead to inadequate legal bases for TCN rights) in several other areas: unemployment benefits in the second Member State, vocational training, identity documents, proof of resources, family reunification, conditions for expulsion, judicial remedies, a three-month deadline for the state to process applications, a provision that application fees be no higher than those for nationals to get identity cards, and a provision that TCNs cannot work while waiting for their permit in the second Member State, which is not in line with the rules for EU citizens. On all of these issues, the ILPA felt that TCN free-movement rights either lacked an adequate legal basis (key for future ECJ rulings that might take rights in a more expansive direction), or else did not do enough up front to equalize TCN free-movement rights with those enjoyed by EU nationals.

EU immigration politics

435

Though the Commission was certainly sympathetic to these positions, it felt that its hands were tied to some degree, and that its proposed directive was an acceptable “first draft” to begin the difficult negotiations with the more recalcitrant Member States in the Council. Any further expansion of TCN rights might alienate the Member States to an insurmountably large degree. Thus, work began in the Council to turn the Commission’s proposal into a real directive that would begin to harmonize TCN status and rights across the EU. It is important to note that the United Kingdom, Denmark, and Ireland (normally the most Euro-skeptic Member States on issues of immigration) played virtually no role in the negotiations over the LTRD, since these states were opting out of the EU’s immigration policy. Thus, other Member States did not have the “usual suspects” to hide behind on this issue (which had worked quite well under unanimity voting, for obvious reasons – Member States could preserve their Euro-federalist reputations while letting the United Kingdom take the heat for blocking harmonization), and had to make their objections clear. And indeed, a newly skeptical Member State stepped up to take the United Kingdom’s place, in the form of Germany. Over the two years that the Directive was negotiated, Germany consistently played a blocking role, preferring to restrict the free movement rights of TCNs. This provides an excellent test of supranationalism, because intergovernmentalists would predict that Germany would get its way, and that any successful harmonization would reflect German preferences. Supranationalists, on the other hand, would predict that Germany would eventually lose control over the policy. Germany took a strong line on the issue of TCN free movement. One German official with whom I spoke even denied that the Tampere Conclusions imply equal treatment between TCNs and EU nationals. Referring to the use of the word “approximate,” he (and the German delegations to the Council) argued that the Tampere Conclusions do not imply that TCNs should have full free movement rights. Thus, instead of the draft directive proposed by the Commission, Germany preferred to let Member States give preference to their own nationals in the labor market – a clear violation of the principle of the free movement of labor. Germany was not alone in this position, however, and was joined by a group of likeminded Member States. In a Council working party meeting of December 2002, the Greek, Italian, and Austrian delegations openly opposed the equal treatment of TCNs as regards free movement rights (European Council 2002). At the same meeting, the German delegation joined these three countries in advocating that the text of the directive be changed so that Member States gained the right to give preference to their own

436

Immigration and the Transformation of Europe

nationals for jobs. France, Finland, and Sweden opposed this suggestion, and advocated full equal treatment for TCNs. As the draft directive was edited and made its way through the Council’s four legislative bodies dealing with immigration (the Working Party on Migration and Expulsion, the Strategic Committee on Immigration, Frontiers and Asylum, COREPER, and the Justice and Home Affairs Council), two camps coalesced around the issue of free movement rights. The German camp, which also included Austria, Luxembourg, Greece, Italy, Spain, and Portugal, resisted granting TCNs full free movement rights. I do not have the space here to fully analyze the reasons behind these national preferences (see Givens and Luedtke 2004 for an attempt to explain these), but in the German case it was clear that domestic politics was the catalyst. Germany had previously been quite Euro-federalist and cooperative on immigration issues (Hix and Niessen 1997), but (in addition to losing the United Kingdom as a “shield”) was now recalcitrant because of political pressure on the Social Democratic/Green government, coming mainly from the L¨ander (since the Bundesrat was controlled by the Christian Democrats). Edmund Stoiber, the Chancellor of Bavaria and possible future leader of the CDU/CSU, who is quite restrictive on immigration (and even wants competence to revert fully back to the Member States!), had been making a political issue of immigration. Since the L¨ander are responsible for paying social benefits and aid, there was a prominent worry that legal migration by TCNs might lead to an increase in social aid, for example, a TCN moving to Germany from another EU country and becoming jobless after two years. Since German social protection is more generous than in some Member States, the L¨ander worried that TCNs would engage in “benefit shopping,” and flock to Germany under the new rules. Since Stoiber and others had succeeded in politicizing this worry, the German delegations felt that they must protect the principle of favored treatment for nationals vis-`a-vis TCNs. We can probably assume that similar worries over national sovereignty6 were operative in Austria, Luxembourg, Greece, Italy, Spain, and Portugal, though Germany’s objections were the most prominent and important in terms of the success or failure of the Directive. The pro-TCN camp in the Council included The Netherlands, France, Belgium, Finland, and Sweden. These Member States, as well as the Commission, wrangled with Germany and the others regarding equal treatment for TCNs. A Commission official told me that the Commission 6

See Luedtke 2005 for an analysis of the impact of nationalism and national identity on the construction of an EU immigration policy.

EU immigration politics

437

got some leverage by reminding Member States of their grand promises (in the Tampere Conclusions), when it came time to draft legislation, thus confirming supranationalist arguments about Commission leadership and agenda-setting. However, the resulting Directive, finally adopted in January of 2004, must be seen primarily as a German victory and a setback for pro-TCN advocates. The same Commission official admitted to me that the resulting LTRD does not live up to the Tampere Conclusions in many areas. In fact, the final Directive is so restrictive towards TCN rights, that a British official told me that the United Kingdom is now seriously considering opting in! Though there were some victories for the pro-TCN camp in non-free movement areas, such as the duration of residence needed before gaining LTR status (Italy wanted it to be six years, but the Commission and the more expansive-minded Member States succeeded in keeping it to five years), on the whole the free movement section of the Directive does not offer much positive evidence for supranationalism (in that protectionist national preferences were safeguarded). Not only did Germany succeed in getting the national labor market preference inserted, Member States also gained the important rights to set numerical quotas on TCNs, and to require that TCNs comply with certain “integration” measures, including taking language classes. Importantly for assessing the impact of immigration policy harmonization on European integration in general, these and other departures from the principle of equal treatment for TCNs were allowed through multiple uses of the word “may” (instead of “shall”) in the language of the directive, so that Member States are not faced with hard legal obligations. Recall that a key part of the supranationalist argument is the coherence of the law, so that the ECJ can later use legal language to raise the standard of rights protection. With this past history in mind, Member States obviously wished to free themselves from further legal action by the ECJ, by agreeing to a Directive that contained weak legal language. Thus, the intergovernmentalist perspective still appears to be accurate in assessing current developments regarding TCN free-movement rights. However, there are some signs that supranational influence may increase in the future. One Council legal specialist with whom I spoke admitted that the Court may not be “impressed” with the text of the Directive, and particularly the removal of the word “right” (of residence) from the final draft. He wondered if the ECJ could in the future possibly infer a right, using one or more of the legal bases mentioned above. This same official told me that in view of the very minimal and restrictive

438

Immigration and the Transformation of Europe

harmonization contained in the LTRD, he sees this Directive as intermediate, and thinks that there will be a new version within ten years. Clearly, the ECJ, Commission, and the European Parliament still have some room for maneuver in bringing TCNs into the free-movement fold, against (some) national preferences. This is especially so if the institutional arrangements found in the European Constitution eventually become reality, given that these would grant the ECJ full jurisdiction over immigration, and would give the Parliament co-decision rights and mandate majority voting in the Council (as opposed to unanimity) on all aspects of immigration. Conclusion Why and how have EU organizations attempted to wrest institutional control away from Member States and move common policies in a more expansive direction vis-`a-vis immigrant rights? National governments initially agreed to limited harmonization of immigration policies in order to regain control over immigration flows and maximize electoral capital (Joppke 2000; Guiraudon 2000; Guiraudon and Lahav 2000; Givens and Luedtke 2004). Thus, the delegation of sovereignty was a calculated move to enhance policy effectiveness, in building a “Fortress Europe” to exclude migrants more effectively. But this control may now be slipping out of the grasp of Member States, and this has had (and will likely have in the future) positive effects in terms of the level of rights protection and freedoms for immigrants. This policy area provides a strong test of supranationalism’s power, because it appears to be a “least likely case” for three reasons: political salience is quite high, the transnational “exchangers” (namely immigrants and their employers and supporters) are politically weak and face enormous populist counter-pressure, and states are openly worried about the dangers of supranationalization in the field of immigration, given their past experiences with the EU, and have taken active steps to avoid it. Therefore, immigration is a strong test of supranational influence. There are good reasons to expect that national governments are intent on maximizing their own control, and although my analysis has shown only modest supranational influence, it also points to potential in the future, as expressed in the worries of some national governments themselves. This is a fairly strong and somewhat counterintuitive finding. Though the initial signs are not entirely promising, if supranational governance is consolidated even in this controversial, core area of sovereignty, then Europe will have made a radical break with traditional notions of statehood and territorial control.

EU immigration politics

439

 Cichowski, Rachel A. 1998 “Integrating the Environment: The European Court and the Construction of Supranational Policy.” Journal of European Public Policy 5(3): 387–405. European Commission 1985 “Council Resolution of 16 July 1985 on Guidelines for a Community Policy on Migration.” COM (85) 48. Brussels: European Commission, March 7. European Commission 2000 “A Communication from the Commission to the Council and the European Parliament on a Community Immigration Policy.” COM(2000) 757 final. Brussels: European Commission, November 22. http://europa.eu.int/eur-lex/lex/LexUriServ/site/en/com/ 2000-com2000 0755en01.pdf. European Council 1999 “Presidency Conclusions: Tampere European Council, 15 and 16 October.” Brussels: European Council. www.parliament.cy/ parliamentgr/101/conclusion tampere.pdf. European Council 2002 “Draft Council Directive Concerning the Status of Third-Country Nationals who are Long-Term Residents.” Interinstitutional File: 2001/0074 (CNS), Working Party on Migration and Expulsion. Brussels: European Council. European Parliament 2001 “Movement and Residence of EU Citizens. A5-0207/2000.” Official Journal of the European Communities 44(May 7): 189–93. http://europa.eu.int/eur-lex/pri/en/oj/dat/2001/c 135/ c 13520010507en01890193.pdf. Garrett, Geoffrey R., Daniel Kelemen and Heiner Schulz 1998 “The European Court of Justice, National Governments, and Legal Integration in the European Union.” International Organization 52(1) (Winter): 149–76. Geddes, Andrew 2000 Immigration and European Integration: Towards Fortress Europe? Manchester: Manchester University Press. Givens, Terri G. 2002 “The Role of Socioeconomic Variables in the Success of Radical Right Parties.” Chapter 7 in Shadows over Europe: The Development and Impact of the Extreme Right in Western Europe, Martin Schain, Aristide R. Zolberg and Patrick Hossay (eds). New York: Palgrave MacMillan, 137–58. Givens, Terri and Adam Luedtke 2004 “The Politics of European Union Immigration Policy: Institutions, Salience, and Harmonization.” Policy Studies Journal 32(1) (February): 145–65. Givens, Terri and Adam Luedtke 2005 “European Immigration Policies in Comparative Perspective: Issue Salience, Partisanship and Immigrant Rights.” Comparative European Politics 3: 1–22. Guild, Elspeth 1998 “Competence, Discretion and Third Country Nationals: The European Union’s Legal Struggle with Migration.” Journal of Ethnic and Migration Studies 24(4) (October): 613–25. Guiraudon, Virginie 1998 “Third Country Nationals and European Law: Obstacles to Rights’ Expansion.” Journal of Ethnic and Migration Studies 24(4) (October): 657–74. Guiraudon, Virginie 2000 “European Integration and Migration Policy: Vertical Policy-making as Venue Shopping.” Journal of Common Market Studies 38(2): 251–71.

440

Immigration and the Transformation of Europe

Guiraudon, Virginie and Gallya Lahav 2000 “A Reappraisal of the State Sovereignty Debate: The Case of Migration Control.” Comparative Political Studies 33(2) (March): 163–95. Hailbronner, Kay and Jorg ¨ Polakiewicz 1992 “Non-EC Nationals in the European Community: The Need for a Coordinated Approach.” Duke Journal of Comparative and International Law 3(1): 49–88. Hix, Simon 1995 “The 1996 Intergovernmental Conference and the Future of the Third Pillar.” Briefing Paper No 20. Brussels: Churches’ Commission for Migrants in Europe. Hix, Simon and Jan Niessen 1997 “Reconsidering European Migration Policies: The 1996 IGC and the Reform of the Maastricht Treaty.” Brussels: Migration Policy Group. Hooghe, Liesbet 2001 The European Commission and the Integration of Europe: Images of Governance. Cambridge: Cambridge University Press. Immigration Law Practitioners’ Association. 2003 “ILPA Response to Proposed EC Directive on the Status of Third Country Nationals Long-Term Residence.” London: ILPA. www.ilpa.org.uk/submissions/residents.html. Joppke, Christian 2000 Immigration and the Nation-State: The United States, Germany, and Great Britain. Oxford: Oxford University Press. Luedtke, Adam 2005 “European Integration, Public Opinion and Immigration Policy: Testing the Impact of National Identity.” European Union Politics 6(1): 83–112. Mattli, Walter and Anne-Marie Slaughter 1998 “Revisiting the European Court of Justice.” International Organization 52(1) (Winter): 177–209. Monar, Jorg ¨ 1998 “Justice and Home Affairs.” Journal of Common Market Studies 36 (Annual): 121–37. Moravcsik, Andrew 1998 The Choice for Europe: Social Purpose and State Power from Messina to Maastricht. Ithaca, NY: Cornell University Press. Papademetriou, Demetrios 1996 Coming Together or Pulling Apart? The European Union’s Struggle with Immigration and Asylum. Washington, DC: Carnegie Endowment for International Peace. Pollack, Mark A. 1998 “The Engines of Integration? Supranational Autonomy and Influence in the European Union,” in European Integration and Supranational Governace, Wayne Sandholtz and Alec Stone Sweet (eds). Oxford: Oxford University Press, 217–49. Schain, Martin, Aristide R. Zolberg and Patrick Hossay 2002 “The Development of Radical Right Parties in Western Europe.” Chapter 1 in Shadows Over Europe: The Development and Impact of the Extreme Right in Western Europe, 1st edn, Martin Schain, Aristide R. Zolberg, and Patrick Hossay (eds). London: Palgrave Macmillan. Sniderman, Paul M., Pierangelo Peri, Rui J. P. de Figueiredo, Jr., and Thomas Piazza 2002 The Outsider: Prejudice and Politics in Italy. Princeton, NJ: Princeton University Press. Statewatch 2003 “Statewatch Comments for House of Lords Select Committee on the European Union Proposed Directive on Long-Term Residents.” London: Statewatch.

EU immigration politics

441

Stone Sweet, Alec 2000 Governing with Judges: Constitutional Politics in Europe. Oxford: Oxford University Press. Stone Sweet, Alec and James A. Caporaso 1998 “From Free Trade to Supranational Polity: The European Court and Integration.” Chapter 3 in European Integration and Supranational Governance, Wayne Sandholt, and Alec Stone Sweet (eds.) Oxford: Oxford University Press, 92–133. Stone Sweet, Alec and Wayne Sandholtz 1998 “Integration, Supranational Governance and the Institutionalization of the European Polity.” Chapter 1 in European Integration and Supranational Governance, Wayne Sandholtz and Alec Stone Sweet (eds). Oxford: Oxford University Press, 1–26. Stone Sweet, Alec, Neil Fligstein and Wayne Sandholtz 2001 “The Institutionalization of European Space.” Chapter 1 in The Institutionalization of Europe, Alec Stone Sweet, Wayne Sandholtz, and Neil Fligstein (eds). Oxford: Oxford University Press. Uçarer, Emek M. 2001 “From the Sidelines to Center Stage: Sidekick No More? The European Commission in Justice and Home Affairs.” European Integration Online Papers (EIoP) 5(5) (May 17) http://eiop.or.at/eiop/texte/2001005a.htm. Weiler, J. H. H. 1993 “Thou Shalt Not Oppress a Stranger (Ex: 23:9): On the Judicial Protection of the Human Rights of Non-EC Nationals,” in Free Movement of Persons in Europe: Legal Problems and Experiences, Henry G. Schermers (ed). Dordrecht: M. Nijhoff, 248–71. Weiler, J. H. H. 1994 “A Quiet Revolution: The European Court of Justice and Its Interlocutors.” Comparative Political Studies 26 (January): 510–34.

17

The effectiveness of governments’ attempts to control unwanted migration Eiko R. Thielemann

Introduction In an increasingly interdependent world, rising numbers of asylum seekers and their highly unequal distribution across countries have meant that forced migration is now regarded as one of the key challenges facing nation-states today.1 This challenge is made even greater by the fact that one state’s policy decisions on the relative leniency or restrictiveness of its asylum regime will create externalities for other states and can thus lead to strained relations between states.2 As a consequence, forced migration has also come to be seen as a crucial challenge for international policy coordination, leading, for example, to rapid advances in the efforts of the European Union (EU) to provide for effective policies in this area. Policy-makers charged with finding an appropriate response to these challenges have been faced with two key questions: First, why have some states been faced with a much higher number of asylum applications than others? And second, what public policy measures, if any, can effectively influence the number of asylum seekers that a state receives? From a national perspective, the most frequent response to the first question has been to argue that if states’ asylum burden is disproportionate, then these countries’ asylum procedures are probably too lenient and their welfare provisions too generous in international comparison. By increasing the restrictiveness of their asylum policy, the argument goes, states will be able to redress the inequitable distribution of burdens, raising concerns 1

2

The largest part of the world’s 15 million asylum seekers in 2001 sought refuge in developing countries. However, from the early 1980s, the number of asylum seekers in Europe increased almost tenfold to 970,000 in 2001. In the period between 1985–99, Switzerland as the largest recipient of asylum seekers on average relative to its population size, was faced with 30 percent more asylum applications than Sweden, 40 percent more than Germany, six times as many as France and the UK, thirty times as many as Italy and 300 times as many as Portugal and Sweden (UNHCR 1999). Recent examples have been the strained relations between Denmark and Sweden following the introduction of highly restrictive asylum measures by the new conservative government in Denmark and the controversy about the Sangatte refugee camp which soured relations between France and Britain.

442

Government attempts to control migration

443

in some quarters, however, about a possible race to the bottom of protection standards. There have so far been few academic attempts (Bocker ¨ and Havinga 1998; Havinga and Bocker ¨ 1999; Holzer and Schneider 2002; Neumayer 2004; Hatton 2004) to use relevant theoretical models developed in the field of economic migration (Ranis and Fei 1961; Harris and Todaro 1970; Borjas 1990; Massey et al. 1993) to systematically analyze patterns of asylum flows in order to establish the importance of policy and other historical, economic or political migration pull factors that can explain why asylum seekers apply in a particular country. Regarding the second question, on the capacity of public policy in this area, there is still little consensus as to whether liberal states can control unwanted migration (Freeman 1994). The ‘transnationalist’ strand of the literature (Sassen 1996; Jacobson 1996; Soysal 1994) emphasizes systemic constraints that undermine the capacity of states to assert effective control in this area. In contrast, the more ‘state-centrist’ strand of the literature (Holzer et al. 2000; Guiraudon and Lahav 2000; Joppke 1997; 1998) argues that states have found new ways to regulate migration in an era of increasing interdependence, which enables them to retain much of their regulatory capacity in this area, even to the extent that their measures have undermined some of the more liberal aspects of the international migration regime. Largely missing from the literature have been quantitative studies that combine the systematic empirical analysis of the determinants of asylum seekers’ choice of destination country and the effectiveness of specific public policy instruments in regulating asylum flows. In an attempt to fill this gap, this chapter analyses UNHCR (United Nations High Commissioner for Refugees) and OECD data from twenty OECD countries for the period 1985–99 and shows many public policy measures aimed, at least in part, at deterring unwanted migration (“bogus asylum seekers”) and at addressing the highly unequal distribution of asylum burdens have remained limited in their effectiveness. The chapter argues that this is because the key determinants of an asylum seeker’s choice of host country are historical, economic and reputational factors that largely lie beyond the reach of asylum policy-makers – as emphasized in chapter 3 by Pedersen et al. and in chapter 11 by Jacqueline Andalls. It also suggests that the effectiveness of unilateral policy measures will be further undermined by multilateral attempts to harmonize restrictive policies and that current efforts such as those by the EU consolidate, rather than effectively address, existing disparities in the distribution of asylum burdens. To make this argument the chapter proceeds as follows: after a short overview of recent public policy responses, an analysis of the literature on migration will identify theoretically informed pull factors. From this,

444

Immigration and the Transformation of Europe

the chapter will generate a number of hypotheses. The next part develops and explains the model which is subsequently used to test the hypotheses empirically based on data for twenty OECD countries over the period 1986–99. The final section discusses the empirical results which call into question some widely-held assumptions about the underlying reasons for the unequal distribution of “asylum burdens” and the effectiveness of unilateral and multilateral deterrence measures. Setting the scene: Forced migration and public policy Since the mid-1980s, the issue of immigration and asylum has gained considerable prominence in OECD countries. The combination of heightened migration pressure and reduced willingness to accept inward migration has pushed the issue towards the top of the political agenda. As economic and political uncertainties increased in the 1990s, public opinion (often encouraged by electioneering politicians and a xenophobic media)3 became more and more wary about inward migration, which in turn produced more pressure on politicians for “decisive” action in this area. The important distinction between economic and forced migration was often lost in the process. Although states are generally free to decide on the number of economic migrants they are willing to accept, in the area of forced migration international obligations such as the Geneva Convention on the Status of Refugees impose important obligations on states. This, however, is not to dispute the fact that, with the door to legal immigration shut in most states since the early 1970s, a significant number of economic migrants have taken the “asylum route” as it has often constituted the only remaining avenue for third country nationals to legally settle in one of the OECD countries. In the 1990s, asylum applications, therefore, became a primary concern for policy-makers in all OECD states. Figure 17.1 shows the number of asylum applications filed in Europe and North America over the period 1980–99. However, policy-makers are not just concerned about the growth in the absolute numbers of asylum applications, over which they have only limited influence given a volatile international system and their international 3

The words “floodgates,” “swamped,” “scroungers,” “soft touch,” and “bogus” are frequently used by newspapers (and at times by politicians) in the context of asylum policy. Take the following examples from newspaper headlines in the United Kingdom: “Our land is being swamped by a flood of fiddlers stretching our resources – and our patience – to breaking point” (The Sun); “Hello Mr Sponger . . . Need Any Benefits?” (Daily Star). “Scandal of how it costs nearly as much to keep an asylum seeker as a room at the Ritz” (The Mail); “ . . . we resent the scroungers, beggars and crooks who are prepared to cross every country in Europe to reach our generous benefits system” (The Sun).

Thousands

0

100

200

300

400

500

600

700

800

1985

Northern America

1990

1995

Figure 17.1. Asylum applications in Europe and North America, 1980–99. Source: UNHCR.

1980

Europe

1999

446

Immigration and the Transformation of Europe

Table 17.1 Average number of asylum applications per year, 1985–99 (per thousand of population) Highest Switzerland Sweden Germany Denmark Netherlands Austria

Lowest 3.35 2.64 2.03 1.81 1.63 1.62

Belgium Norway Canada EU-15 France UK

1.41 1.23 0.97 0.96 0.55 0.46

Australia Ireland Greece US Finland Hungary

0.36 0.33 0.31 0.28 0.24 0.19

Spain Czech Rep. Italy Portugal Poland Japan

0.17 0.16 0.15 0.05 0.03 0

obligations. They are also concerned about the relative distribution of asylum applications between states, in particular when they feel that the policy measures adopted in neighboring states are at least in part responsible for their own asylum burden. When analysing the development of asylum applications across OECD countries, it soon becomes clear that the distribution of asylum applications has been highly unbalanced. Public attention was drawn to this when in 1992 Germany received over 438,000 asylum applications, which constituted more than 62 percent of all applications registered in Europe (UNHCR 1999). However, a focus on absolute figures is often misleading. When using the more meaningful measure of relative burdens, that is, one which takes account of differences in reception capacity,4 the unevenness in distribution becomes even clearer (see Table 17.1). It can be shown that since the mid-1980s some European countries, most notably Switzerland, Sweden, and Germany, have borne a much higher relative (per capita) burden than the EU average. Relatively high burdens have constituted a considerable domestic challenge in some countries. The inequitable distribution of burdens has also led to tensions between some OECD countries, particularly within the EU, as there was a feeling in some quarters that certain countries were introducing unilateral deterrence measures to deflect asylum applications towards other countries. Against the background of the serious collective action problems involved in this area, the German government in 1992 proposed a European-wide asylum burden-sharing system. The German proposal5 foresaw the distribution of asylum seekers across Europe according to indicative figures that were based on a distribution key composed of three 4 5

There are several possible criteria to measure reception capacity, the most common one being relative population size (per capita). Council Document 7773/94 ASIM 124.

Government attempts to control migration

447

criteria which were given equal weight (population size, size of Member State territory and GDP).6 The centerpiece of the German draft was the introduction of a compulsory resettlement mechanism which would have worked as follows: “Where the numbers admitted by a Member State exceeds its indicative figure [ . . . ], other Member States which have not yet reached their indicative figure [ . . . ] will accept persons from the first State.”7 This proposal, however, did not find the necessary support among other countries, with the United Kingdom in particular being strongly opposed to such a scheme.8 Since then, European states have instead agreed on a number of (limited) steps towards harmonizing asylum policy regulations across Europe, including joint visa regimes and a common refugee definition.9 At the same time, states have continued to undertake unilateral measures aimed at deterring asylum applications to their country. Such unilateral deterrence measures have covered the whole range of policy options available to policy-makers in this field, such as measures on access, status determination, and those concerning the integration of asylum seekers. These measures have been based on a number of apparently widelyheld assumptions. First, asylum seekers are assumed to be well informed (either through personal networks or their traffickers) about the relative openness and attractiveness of different destination countries’ asylum regimes. Second, they are expected to choose to apply to those countries which have the most attractive asylum policy package, in terms of access, determination and integration/welfare measures. In other words, “asylum shopping” is assumed to be a widespread phenomenon. Finally, there appears to be a belief that countries with relatively more attractive asylum policies will come to be regarded as a “soft touch” and will consequently have to cope with a disproportionately high number of asylum applications. Despite the fact that all of these assumptions are at least questionable, they have formed the basis for the introduction of unilateral deterrence measures, some of which have come to be viewed by both policy-makers (and academics) as having been highly effective. For example, the dramatic drop (by 71%) in the number of applications received by Germany between 1992 and 1994 has widely been 6

7 8 9

The form of the suggested redistributive mechanism followed the example of German domestic legislation, which stipulates a similar key for the distribution of asylum seekers among the German L¨ander. See 45 of the German Asylum Procedure Act (Asylverfahrensgesetz). Council Document 7773/94 ASIM 124. BMI, Pressemitteilung vom 1.12.1994, FAZ 27.1.1995, p 2; BT-Drs. 13/1070, 55; Integrationsbericht, p 92. For a comprehensive discussion see Noll 2000.

448

Immigration and the Transformation of Europe

attributed to the 1993 restrictions that were introduced to the German Basic Law and the corresponding legislation pertaining to foreigners (a conventional wisdom that will be fundamentally challenged later in this chapter). Some recent research (Cornelius, Martin, and Hollifield 2004; Holzer, Schneider, and Widmer 2000; Robinson and Segrott 2002), has already cautioned against being overly confident about the effectiveness of asylum policy in steering migration flows. The quantitative analysis of the Swiss case showed that Switzerland has had only limited success in influencing the inflow of asylum seekers between 1986 and 1995. The study showed that the Swiss government was partly successful in manipulating the relative number of refugees it recognized to achieve its deterrence objectives. However, the study concluded that deterrence measures can be expected to be unsuccessful “if the push factors in a region nearby to the receiving states reach a critical level” (Holzer, Schneider, and Widmer 2000: 1205). Qualitative research conducted in the UK draws even more skeptical conclusions. This research that was based on sixty-five interviews with asylum seekers found that most of the respondents knew very little about UK asylum policy before their arrival. The study found that they certainly did not have sufficient knowledge to make an informed choice based on the rational evaluation of reception conditions and welfare benefits on offer by several possible destination countries (Robinson and Segrott 2002: 46, 63). These studies, however, suffer from their focus on individual countries which makes generalization difficult. Moreover, qualitative research based on survey data, whereby asylum seekers are asked about their travel route, their preferences as to particular countries of destinations and why they applied in a particular country can be problematic. Although such studies might have a strong appeal (who else than the asylum seeker knows why s/he applied in a particular country?), research of this kind suffers from two difficulties in particular. First, survey analyses of the required kind is very costly, especially when one is interested in systematic comparative analyses across countries (let alone over time). Interviews are especially costly because they often have to rely on interpreters. Second, and potentially even more problematic, is the fact that asylum seekers might have a strong incentive to emphasize certain determinants over others, as they know that certain answers might compromise their asylum application or residence status.10 It is for these reasons that the systematic 10

This problem is certainly confounded when the research in funded by the national body which determines asylum cases, as was the case in the above study by Robinson and Segrott which was financed by the United Kingdom Home Office.

Government attempts to control migration

449

quantitative analysis in this chapter, despite its own limitations,11 hopes to make a contribution to the existing literature. Which host country? Migration theory and pull factors Although still an under-theorized area of study, one can identify a number of prominent theories and models of international migration (Kritz et al. 1992; Massey et al. 1993; Meyers 2000). One of the most commonly known theoretical migration models is the “push-pull model.” In its most abstract form, the push-pull model suggests that there are push factors in countries of origin that cause people to leave their country, and positive or pull factors that attract migrants to a receiving country. Although this model cannot simply be transferred to the area of forced migration,12 it still offers a number of insights for research on the direction of asylum flows and their distribution between host countries. A review of the theoretical literature on migration13 produces five categories of pull factors – economic, historic, political, geographic, and policy related – which will be introduced in turn below. Economic factors Economic theories of migration have only limited applicability in the area of forced migration, in which displaced persons often will have little or no time for deliberations of “utility maximisation.” However, despite the important substantive differences between economic and forced migration, economic considerations can still be expected to play a role in the 11

12

13

The most imminent problems with quantitative data analysis stem from the aggregation of data and difficulties stemming from the incongruence of national definitions. For an extensive discussion see Crisp (1999) and the chapter on “Sources and Comparability of Migration Statistics” in SOPEMI 2002: 269–73. In the area of forced migration, pull factors are not assumed to be the driving forces behind persons leaving their country, and push factors are often assumed to be limited to persecution of the kind listed in the Geneva Convention (UN Convention on the Status of Refugees 1951, as amended by the 1967 New York Protocol). The Convention defines a refugee as a person who “owing to a well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group, or political opinion, is outside the country of his nationality, and is unable to or, owing to such fear, is unwilling to avail himself of the protection of that country.” Refugees must therefore be seen as distinct from economically motivated labor migrants, as the former are assumed to move involuntarily, while for the latter there is an element of choice in their migration decision. In practice, such classifications are less clear than is often assumed, as political and economic causes frequently join forces in producing movement, and freedom of choice is rarely absolute and might be limited in both types of migration. One objective of this review of the general theoretical literature on migration is to analyze to what extent hypotheses derived from this literature (that deals primarily with economic migration) can be usefully extended to the area of forced migration.

450

Immigration and the Transformation of Europe

area of forced migration. Like other migrants, asylum seekers will often face financial and other constraints which will influence their choice (limited as it might be). In many cases, even forced migrants will have some choice as to their country of destination and can therefore be expected to take economic consideration into account. Moreover, we do of course know (Kunz 1981; Zolberg et al. 1989) that in a world of high crosscountry income differentials and highly restrictive admission policies, persecution is not the only push factor behind the large number of asylum applications of recent years. Given the above, economic factors must be included when analyzing the incidence of asylum applications across the OECD. Neoclassical economic migration theory (Ranis and Fei 1961; Harris and Todaro 1970; Todaro 1989) explains the decision to migrate as one of income maximization in which wealth differentials and differences in employment opportunities constitute important pull factors. International migration is expected to be determined by geographic differences in the supply and demand of labor. Ultimately, in this view, it is wage differentials which can explain movements from low-wage countries to high-wage countries. In its micro-economic extension (Sjaastad 1962; Borjas 1990), rational actors (be it individuals or larger units such as families or households) decide to migrate in the expectation of a positive, often monetary, net return from migration. In this framework, the decisive factor is income differentials as well as the probability of employment in the destination country. In other words, migration decisions can be seen as being guided by processes of income maximization and risk minimization. Historical ties, networks, and path dependency Historical ties between countries of origin and destination countries often lead to transport, trade and communication links which tend to facilitate movements of people from one country to the other. Material links are often accompanied by ideological or cultural links. Colonial legacies often explain why administrative and educational systems in third world countries mirror those of a past colonial power and often continue to be reflected in migration flows long after independence (Fassmann and Munz ¨ 1992). For example, Pakistanis and Bangladeshis learn English, are raised in a British-style education system and keep up links with the UK through the Commonwealth. Language ties, communication links, and cultural networks that are responsible for the diffusion of particular consumption patterns, can be responsible for channeling international migration to particular destination countries (Massey et al. 1993: 446–7).

Government attempts to control migration

451

Moreover, the fact that migrant or refugee communities have been established in certain destination countries as a result of historic ties, will often lead to the growth of migrant networks that may foster future migration flows. Such networks are sets of interpersonal ties between earlier migrants already resident in a destination country and potential migrants in countries of origin that are based on family ties, friendships or shared community origin. Such ties can significantly reduce the costs and risks of migration, thus channeling migration flows in the direction of earlier migration flows. By passing on information about access to a particular country and its employment opportunities, they constitute a form of social capital (Hugo 1981; Taylor 1986). Once migration connections have been established, the presence of relatives, friends, and/or others from the same community of origin may form a strong incentive to choose a particular destination. Migration may thus be seen as a selfsustaining diffusion process (Massey et al. 1993), which governments will find difficult to control. Following the same line of reasoning we could also expect a certain degree of path-dependency from one period of migration to the next. Such a process is likely to be the result of two dynamics. First, there will be a reduction of costs and risks for migrants as they can rely on the support of personal networks. Second, there will be certain persistence of existing migration routes and patterns, as agents and traffickers will have incurred sunk costs by investing in the creation of networks which they will be reluctant to give up (Pierson 2000). Political values (“liberalness”) Concerns about personal security and their acceptance into a new host society can be expected to be important considerations for potential migrants, in particular forced migrants who are leaving their country of origin because of concerns about their, or their family’s, safety. The reputation of a country in terms of its “liberal” credentials as well as its track record on issues such as adherence to human rights standards, international humanitarian crises, community relations, reception and integration of foreigners, naturalization policies, and the like, can be expected to play a role in a migrant’s consideration about the relative attractiveness of countries of destination. Geography Ease of access, in particular geographic proximity, between a country of origin and a country of destination can also be expected to be a pull

452

Immigration and the Transformation of Europe

factor in migration patterns. Despite technological developments which have made geographic distance less of an issue than it was in the past, most migrants’ resources are limited, and smaller distances will often mean lower costs of transport and hence easier access. In other words, geographic distance can often be regarded as a proxy for the costs of movement. Although other factors, such as length and relative openness of countries’ territorial borders, will also play a role as to how accessible a country of destination is for migrants, geographic distance can be expected to constitute an important consideration. We can also expect an interaction with the other pull factors that have already been discussed. Although geographic proximity does not guarantee the establishment of cross-border ties, geographic proximity can clearly facilitate the formation of such ties. Deterrence policy States often regard asylum burdens as a “zero sum” phenomenon, in which a reduction of one country’s burden will result in increasing burdens for the others. The assumption is that there are a certain number of migrants each year who intend to claim asylum and that the role of national asylum policy is to restrict the inflow into a particular country to an acceptable proportion. This means that policy-makers will try to use migration policy instruments to make sure that their country will not be seen as a “soft touch,” that is, an overly attractive destination country that will attract an unacceptable proportion of asylum seekers. Three sets of such instruments in particular are at their disposal: access control; the determination process; and migrant integration policy. Access control policy refers to the rules and procedures governing the admission of foreign nationals and its instruments include visa policy, regulations for carriers, safe third-country provisions, etc.14 Rules concerning determination procedures relate to entry into a country’s refugee recognition system, appeal rights, and rules concerning protection that is subsidiary to the rather narrowly defined Geneva Convention criteria for full refugee status.15 Finally, integration policy is concerned with rights and benefits given to asylum seekers inside a country of destination (e.g., work and 14

15

“Safe third country” provisions mean that asylum seekers are denied access to the refugee status determination procedure on the grounds that they could or should have requested and, if qualified, would actually have been granted asylum in another country. In practice this means that asylum seekers who have traveled through other countries before reaching their destination will not have their asylum application examined in the country of their choice, but will be returned to the other country (Hailbronner 1993; Kjaergaard 1994). See refugee definition in 12 above.

Government attempts to control migration

453

housing conditions, rules on freedom of movement, welfare provisions, educational opportunities, etc.). Policy-makers can introduce changes in the regulations in these three areas in an attempt to raise the deterrence effect of their policy, which in turn is expected to make their country less attractive to asylum seekers in relative terms. The various explanations of pull factors for migrants, and in particular asylum seekers, are obviously not mutually exclusive. On their own, as well as in combination, they can be expected to help explain why asylum seekers apply for asylum more in some OECD countries than in others. Individuals might engage in cost–benefit calculations that make them choose richer countries with more employment opportunities over poorer ones with fewer work opportunities; they might try to reduce risks and costs by using existing networks; they might prefer a more liberal country of destination over a less liberal one; they might be more likely to end up in a country that is relatively closer to their country of origin; and finally their decision might be affected by the relative asylum policy-mix of different potential countries of destination. The purpose of this chapter is not to examine necessarily competing theories, but to test the relative strength of hypotheses that can be drawn from the above discussion of possible pull factors. Methodology To do this, the chapter uses a time-series cross-section of aggregated data for twenty OECD countries for the period 1985–99 collected from UNHCR, OECD, and the US Committee for Refugees.16 This quantitative analysis allows the testing of hypotheses regarding the existence and relative strength of the different potential pull factors. In doing so, the chapter will control for particular country and time effects such as differences in reception capacity, and fluctuations of the absolute number of asylum applications over time. Dependent variable When looking at the issue of asylum from an international “burdensharing” perspective, it is more interesting to focus on the number of relative asylum applications (i.e. applications per capita) across time and place, than on the absolute numbers of applications which has tended to dominate the public debate on asylum. Relative figures are the crucial 16

Due to data incompatibilities, the US, Canada, and Australia could not yet be added to the dataset.

454

Immigration and the Transformation of Europe

reference figures if states want to check the success or failure of their attempt to achieve a more equitable distribution of burdens as a result of international cooperation. Also, it is hardly surprising (nor objectionable) that in absolute terms, larger countries will tend to attract more inward migration than smaller countries. This chapter, therefore, seeks to explain the number of asylum applications in each country and for each year of the data set, relative to the population of each of these countries while controlling for variations in the number of total applications and overall population growth across all the OECD countries included in the dataset.17 To calculate the dependent variable, I used the annual UNHCR statistics of asylum applications and OECD data on population developments. To arrive at the observations for my dependent variable for each country and each year, I divided the number of asylum applications by the country’s population size and put this figure in relation to the total number of asylum seekers divided by the total population of all countries under investigation.18 Explanatory variables The explanatory variables are constructed in such a way as to allow for the examination of the five above theories on key pull factors for asylum applications. First, to test for economic pull factors, the chapter analyzes OECD data on annual GDP growth (in percent) and the total number of registered unemployed. The expectation is that a country’s relative burden will be positively correlated with its economic performance and negatively with its numbers of unemployed. Second, to test the importance of geographic factors, I determine the average distance between the capital of a destination country and the capitals of the top five countries of origin in each year.19 The expectation 17

Asylum applications per population is the most commonly accepted way of analyzing relative burdens in this area. Controlling for GNP, instead of population size, leads to an almost identical ranking order in terms of relative burdens. As this analysis here is interested in explaining the distribution of relative asylum burdens over time, it does not seek to assess the role of push factors responsible for variations in absolute asylum applications.

18

Expressed formally: Bi,t =

19

ai,t pi,t At Pt

, whereby the term B represents the relative number of

asylum applications received in country i in year t; a stands for the absolute number of asylum applications received in country i in year t; p for the population of country i in year t; A for the sum total number of asylum applications received across all OECD countries in the dataset in year t and P represents the sum total population figure of all OECD countries in the dataset in year t. To do this I used a programme developed by John A. Byers which can be accessed at www.wcrl.ars.usda.gov/cec/moregen.htm.

Government attempts to control migration

455

Table 17.2 The main asylum seekers producing countries (top 5) Year/Rank 1.

2.

3.

4.

5.

1985 1986 1987 1988 1989 1990 1991 1992 1993

Sri Lanka Iran Turkey Poland Turkey Romania Yugoslavia Yugoslavia Yugoslavia

Iran Poland Iran Turkey Poland Turkey Romania Romania Romania

Poland Lebanon Sri Lanka Iran Sri Lanka Lebanon Albania Bulgaria Bulgaria

Ghana Ghana Yugoslavia Lebanon Iran Afghanistan Sri Lanka Congo Turkey

1994

Yugoslavia

Turkey

Romania

Iran

1995

Yugoslavia

Turkey

Turkey Turkey Poland Yugoslavia Yugoslavia Yugoslavia Turkey Turkey Bosnia & Herzegovina Bosnia & Herzegovina Iraq

Romania

1996 1997 1998 1999 2000

Yugoslavia Iraq Iraq Iraq Iraq

Iraq Turkey Turkey Afghanistan Afghanistan

2001

Turkey Yugoslavia Yugoslavia Yugoslavia Serbia & Montenegro Afghanistan

Bosnia & Herzegovina Sri Lanka Afghanistan Afghanistan Turkey Iran

Iraq

Turkey

2002

Iraq

2003

Russian Federation Russian Federation

Serbia & Turkey Montenegro Serbia & China Montenegro Serbia & China Montenegro

2004

Afghanistan Sri Lanka Sri Lanka Somalia Turkey

Serbia & Iran Montenegro Afghanistan Russian Federation Iraq Turkey Turkey

India

is that countries of destination that are geographically closer to the top five countries of origin in any particular year will attract relatively more asylum seekers. Table 17.2 lists the top five countries of origin for each year since 1985. Third, to test the role played by a country’s liberal reputation, I used overseas development aid (OECD data measured in millions of dollars) as a proxy variable for a country’s “liberalness.” I expect a more liberal country (i.e. one with a relatively high ODA/GDP ration) to attract a relatively high number of asylum seekers.20

20

Other possible indicators for the relative “liberalness” of a country, such as number of racial attacks or percentage of extreme right-wing votes were not used due to the lack of reliable comparative data.

456

Immigration and the Transformation of Europe

Fourth, to test network/historic ties theories, I add (at t−1) the stock of foreign population from the top five asylum countries (at time t). The expectation is that countries with an already relatively large stock of foreign nationals from the main countries from which asylum seekers are originating in a particular year will receive a relatively greater number of applications in relation to their size. Finally, to analyze the importance and the effectiveness of asylum/ deterrence policy measures, I use a deterrence index fluctuating between 0 (lowest deterrent effect) to 5 (highest deterrent effect). To calculate the index, I analyzed two sets of annual yearbooks, the OECD’s “Trends in International Migration” (SOPEMI) and the United States Committee for Refugees’ “World Refugee Survey” for the years 1984–99. Each describes and analyzes developments in national asylum policy measures for each country in the chapter’s data set. Five measures in particular stand out that have been widely regarded by policy-makers as having the potential to significantly influence an asylum seeker’s decision as to which country to apply to (United Kingdom Home Office 2002).21 First, in the area of access control, arguably the most important measure was the introduction of so called “safe third country” provisions, which mean that persons seeking asylum will be refused entry into a country if, on their way to this country, they traveled though another state which the first country regards as safe, and in which the asylum seeker could have applied for asylum. If an asylum seeker’s travel route only transpires in the course of the determination procedure, he or she can be sent back to the “safe third country.” The introduction of “safe third country provisions” meant that asylum seekers travelling “overland” to Europe were no longer able to legitimately claim asylum in the country of their destination, as the responsibility for their case was shifted onto a neighboring country through which they had traveled. To account for the introduction of safe third country provisions, I created a dummy variable which takes the value 1 for each year that safe third country provisions were applied in a country and the value 0 for all other years. Second, with respect to a country’s determination procedures, the most important potential pull factors that can be influenced by national policymakers are the rules concerning the granting of subsidiary protection status which allows an asylum seeker to remain in a country of destination even though their application for full refugee status under the Geneva 21

Other relevant indicators such as a countries’ detention and deportation rates, their visa requirements, and their readmission agreements with third countries have not yet been included in the index for lack of comparative data. However, these measure are expected to positively correlate with the other indicators used here and their omission is therefore not expected to significantly distort the results.

Government attempts to control migration

457

Convention is refused. Destination countries have complete discretion in defining the requirements that protection seekers have to fulfil to be awarded such subsidiary status, which means that within Europe, the percentage of asylum seekers allowed to stay in a country on the basis of the award of some protection status varies from single figures to over 70 percent (UNHCR 1999). Again, I have created a dummy variable which takes the value 1 if a country of destination is below the average with regard to the percentage of asylum seekers it allows to stay in its country in a particular year, and which takes the value 0 if the percentage of protection seekers allowed to stay is above the OECD average. Finally, much of the discussion of the past few years has focused on the potential pull effects entailed in a third category of asylum policy, namely that of integration measures for asylum seekers. Here three policy aspects are often regarded as being crucial: first, freedom of movement versus a compulsory dispersal policy; second, cash welfare payments versus a system of vouchers; and third, the right to work under certain conditions versus a general prohibition to take up employment as an asylum seeker. The first of these concerns is the right of asylum seekers to move freely within their country of destination until their asylum claim has been determined. While a federal state such as Germany has always had central reception centers from which asylum seekers are to be dispersed to the different L¨ander according to their relative population size, some unitary states – most notably the UK – have recently introduced similar measures. Although dispersal measures first and foremost are an attempt to alleviate pressures from particular (usually metropolitan) areas which are faced with a strong concentration of asylum seekers, such measures are also hoped to deter unfounded asylum claims. Second, the “cash” payment of welfare benefits rather than a payment “in kind” or through a voucher system has sometimes been regarded as a pull factor for asylum seekers.22 This has led a number of OECD countries to stop giving asylum seekers cash benefits and to replace cash payments by the direct provision of housing, food, and health care. In 1999, the UK and Ireland introduced a voucher system for asylum seekers, despite the fact that the two governments were advised that such a system would be more costly to administer than a cash-based system.23 However, governments have been attracted to vouchers due to the deterrent effect that has sometimes 22

23

The British government, for example, resisted pressure to abolish the UK’s voucher scheme. Government advisors warned that “re-introducing cash benefits would create a pull factor for thousands more asylum seekers” (“Details of Blunket’s asylum shake-up,” The Guardian, February 7, 2002). In light of strong protests by human rights NGOs and rising costs, the UK recently abandoned its voucher scheme and reintroduced the previous cash-based system.

458

Immigration and the Transformation of Europe

6

5

4

3

2

1

0 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Germany

EU

UK

France

Figure 17.2. Asylum policy deterrence index in the EU (1985–2000).

been ascribed to such non-cash schemes. Finally, allowing asylum seekers to work while their claim for asylum status is being assessed has also sometimes been regarded as a potential pull factor for asylum seekers. All countries of destination have work restrictions for asylum seekers in place. However, a number of countries have gone further and now prohibit asylum seekers from undertaking any work until their asylum claim has been accepted. To assess the potential deterrence effect of the above measures, three dummy variables were created which take the value 1 (for each year and country) for the existence of a dispersal scheme, a non-cash based system of benefits, and a law which prohibits asylum seekers from working until their claim has been accepted. Adding all the dummy variables for all five of the above potential deterrence measures for each country and each year results in a country’s deterrence index for a particular year.24 Figure 17.2 charts the resulting deterrence index for the three largest EU Member States and the EU as a whole. The expectation is that the higher the index for a particular country in a particular year, the lower that country’s relative attractiveness will 24

As a simplifying assumption, I take each of the five policy measures to have the same potential deterrence effect.

Government attempts to control migration

459

Table 17.3 Expected relationship between variables

Independent variables

Dependent variable: Relative number of asylum applications

Economic pull factors: ∗ number of registered unemployed (in t − 1) ∗ annual real GDP growth (in t − 1)

− +

∗ stock

Historical pull factors: of foreign nationals from top five countries of origin (in t − 1)

+

Political pull factors ∗ annual ODA payments as% of GDP (in t − 1)

+

Geographic pull factors: ∗ average geographic distance between capital of a destination country and capitals of the top five countries of origin

−

Policy related pull factors: deterrence index (in t − 1)

−

∗

be, and hence its relative burden stemming from asylum applications. Table 17.3 summarizes the expected relationships between the variables discussed above. Model estimation To estimate the relationship between these variables and relative burdens for individual countries, the chapter uses pooled time-series cross-section (TSCS) ordinary least square regressions (Stimson 1985) with panel corrected standard errors (PCSEs) (Beck and Katz 1995).25 Prais-Winston transformations are used to eliminate serial correlation of the errors and to take account of cross-section and panel specific auto-correlation. In running the regression of the above independent variables on the number of relative asylum seekers, I lagged GDP growth, unemployment, foreign population, and the deterrence index by one year as one might reasonably 25

“Pooled,” “panel” or “TSCS” analysis has become a popular tool for the empirical analysis of issues in comparative politics and international relations. It involves the analysis of N cross-sections (countries) and T time periods (years). It increases the number of observations available and allows for the analysis of dynamic factors in crossnational comparative research. It corrects for expected downward bias in standard errors and upward bias in t-statistics (Hicks 1994) by eliminating serial correlation of the errors applying Beck and Katz’s standard method of “panel corrected standard errors” (PCSE). For a recent review on “pooling” see Beck (2001).

9.6

9.6

9.6

9.6

0

0

0

0

1985

1999

1985

Spain

Japan

Germany

Belgium

1999

Sweden

Netherlands

Greece

Czech Rep.

1985

Figure 17.3. Relative asylum burdens, 1985–99.

Portugal

Italy

France

Austria

1999

1985

Switzerland

Norway

Hungary

Denmark

1999

1985

UK

Poland

Ireland

Finland

1999

Government attempts to control migration

461

expect that it was the performance of these indicators in t − 1, and not current performance, that constituted a pull factor for persons applying in the period t. Discussion of statistical results Figure 17.2 shows the distribution of relative burdens as a result of asylum applications per 1,000 of population over time in the twenty countries under investigation. In the majority of countries we can observe significant variations in relative burdens over time. We also observe that the relative burden of Germany over time has been comparable, and at times was considerably smaller, to that of some smaller countries such as Sweden and Switzerland. We also observe that many of the other bigger countries, in particular the UK, France, and Japan have attracted far fewer than average applications relative to their population size. Finally, one can observe that in a number of countries, most notably Ireland, Belgium, Hungary, and the UK, relative applications have increased significantly in more recent years. The results of the regression analysis are presented in Table 17.4. We find that some historic, economic, and political pull factors have a strong impact on the distribution of asylum applications. There is also evidence that the relative leniency or restrictiveness of a country’s asylum policy (expressed in the deterrence index) has a highly significant effect on the number of applications received. However, it will be shown below that the impact of the different policy measures that make up that index is highly varied. Economic pull factors As expected, high unemployment figures are negatively related to relative numbers of asylum applications and this relationship is highly significant. While some displaced persons will have little or no choice where they end up applying for asylum, as travel options might be limited (or predetermined by existing trafficking routes) and due to the fact that some forced migrants might be under great time pressure to leave their country, not leaving them sufficient time to weigh their options, other asylum seekers will have more time and the ability to choose where to apply for asylum. The data analyzed here suggests that economic factors do play a role when it comes to decisions about where to apply for asylum. When controlling for the other factors included in the model, one observes that countries which offer greater employment opportunities also tend to have higher relative numbers of asylum applications. This will of course

462

Immigration and the Transformation of Europe

be true for those economic migrants who use the asylum route in an attempt to circumvent the restrictive immigration regimes of developed countries. However, given the strength of the above correlation and the fact that almost 40 percent of all asylum seekers across OECD countries are awarded some protection status, we can reasonably interpret the above results in support of our expectation that labor market considerations do play a role in the considerations of forced migrants as well. Seeking physical security from persecution as well as economic opportunities in a country of destination can hardly be regarded as incompatible objectives for people forced to leave their country of origin. In contrast to the strong effect labor market factors, general economic growth appears to have no significant effect on the distribution of asylum seekers. Part of the reason for this might be that the OECD countries analyzed here are all likely to be perceived as rich, economically thriving, industrialised countries. For new arrivals to benefit from the economic situation of a host-country, employment opportunities will tend to be more important than a country’s performance with regard to short-term economic growth. Historic factors/networks The existence of historical ties and established networks also comes out as highly significant. The relative number of asylum applicants countries receive is strongly and positively correlated with the number of people from the main countries of origin already resident in these host countries. This supports the suggestion that the existence of interpersonal ties, be it with relatives, friends or people from “back home” already resident in a particular country of destination can act as a strong magnet for asylum seekers. This is not surprising as it is through such networks that information about the country of origin will be passed on to potential asylum seekers. Despite the fact that qualitative studies suggest that the amount of information passed from relatives and friends from the destination country to potential asylum seekers before they leave is quite limited (Robinson and Segrott 2002: 41),26 it appears that any kind of contact, no matter how fleeting, will constitute a pull factor for potential asylum seekers that can tip the balance towards the decision to claim asylum in a particular country. 26

Robinson and Segrott’s survey and interview data suggests a number of reasons for this limited degree of prior contacts – lack of time for those who had to flee at short notice, the dangers involved in risking that others might find out about their emigration plans, and loss of contact due to internal flight conditions prior to their decision to leave their country (2002: 41).

Government attempts to control migration

463

Liberalism As expected, favorable perceptions as to how liberal a potential hostcountry is, show a strong, and positive relationship with the relative numbers of applications that a country receives. Countries which show a high concern for people beyond their own border and engage disproportionately in efforts to alleviate underdevelopment in the third world (through ODA payments) attract larger relative numbers of asylum seekers than other countries (Thielemann 2003). Distance Given the difficulties and costs involved in long-distance travel, we expected a negative relationship between the average distance of a host country from the principal countries of origin and that host-country’s asylum burden in any particular year. However, in our dataset of twenty OECD countries, proximity as a pull factor does not produce any significant effect in the analysis. One principal reason for this will be due to the particular selection of host-countries analyzed here (mostly EU countries which are closely bunched together and which – some important exceptions notwithstanding – have tended to be quite far away from the principal countries of origin). If one included host countries in the developing world, one would expect the distance variable to matter a lot more. The results might, however, also partly be explained by the fact that an asylum seeker’s sense of security might increase the further away he or she settles from the country in which they suffered persecution. These findings suggest that one has to refrain from generalizing the results of single country case studies (like that by Holzer et al. 2000) which strongly emphasize the importance of geographic proximity. Clearly there are instances when geographic proximity does matter, especially when geographic pull factors interact with other pull factors such as existing historical ties (as was the case with refugees from former Yugoslavia fleeing to Germany or Switzerland in the 1990s). However, the broader analysis across time and space reveals that in today’s Europe, geographic factors are more limited in their effect than has sometimes been suggested. Deterrence The combined effects of deterrence measures, as shown in the deterrence index, comes out in the expected negative direction and does so at significant levels. The effect of policy-related factors, however, is not as significant as that of historic and economic factors. Moreover, we find

464

Immigration and the Transformation of Europe

Table 17.4 Determinants for the relative number of asylum applications

Number of unemployed (lagged) GDP growth (lagged) Stock of foreign nationals (lagged) Relative ODA payments Average distance Deterrence index

Expected sign

Coef

Z

− + + + − −

−.0007394 .0023206 .0012462 1270083 0.00000382 −.2266697

−7.49∗∗∗ 0.12 8.20∗∗∗ 2.88∗∗∗ 0.08 −2.56∗∗∗

N = 227; R-squared = 0.41; ∗∗∗ p < 0.01.

that the measures analyzed appear to be quite short term in their effect. When we lag the deterrence index by more than one year, it ceases to have any significance. Finally, if one disaggregates the measures included in the deterrence index (see Table 17.5), one finds that the index’s significance is due to the effect of only two of the five deterrence measures analyzed here: not allowing asylum seekers to work until their application has been successful or until they have been allowed to stay in the hostcountry more permanently on the basis of a subsidiary protection status; and granting protection status to a smaller percentage of asylum seekers (in relation to the total number of applications) than other host states. Each of these two measures on its own is significant at the 0.05 level. Their combined significance in the deterrence index is even stronger. No significant effect in reducing asylum applications, however, could be found for the other three measures that have dominated the public policy debate on asylum in recent years. These are: r measures that allow states to turn asylum seekers back at their borders and return them to so-called “safe third countries”; r measures that deny asylum seekers freedom of movement within a host country, that is, the introduction of dispersal schemes; and r measures that have meant the end of cash benefit payments to asylum seekers, that is, through the introduction of a voucher scheme. The following will provide some initial suggestions as to how to explain the variation in effectiveness of these different measures. Discussion At the most general level, it seems clear that information about status determination and employment opportunities do reach asylum seekers either directly or indirectly through their agents and traffickers, whereas knowledge about more detailed policy measures is either not available to

Government attempts to control migration

465

Table 17.5 Impact of individual deterrence measures Deterrence measure

Expected sign

Coef

Z

Below average recognition rate Prohibition to work Safe third country provisions No freedom of movement Non-cash benefit payments

− − − − −

−.2714122 −.4578465 −.1900568 −.0748292 −.1256857

−2.40∗∗ −2.31∗∗ −1.13 −0.30 −1.40

N = 227; R-squared = 0.41; ∗∗ p < 0.05.

asylum seekers, or is not considered to be important enough to determine decisions regarding the choice of country of destination. The likelihood of asylum seekers receiving some kind of status that allows them to remain in their host country should they wish to do so, unsurprisingly is of the utmost importance. We have known for some time that host-countries interpret their international obligations under the Geneva Refugee Convention in very different ways (ECRE 2000)27 and that recognition rates can vary greatly between host-countries at any particular time, even for asylum seekers from the same country of origin (Holzer and Schneider 2002: 43). Moreover, host-countries have also dealt very differently with discretionary granting of subsidiary protection status to those asylum seekers not qualifying for refugee status, but whom host-states feel cannot, or should not, be sent back to their country of origin. Even though asylum seekers will not of course, have access to comparative league tables on which to base their decisions, information on whether or not other asylum seekers were allowed to remain in a given host-country clearly appears to be transmitted back to agents, traffickers and other potential asylum seekers in the countries of origin. Given the high significance of employment opportunities as a pull factor for asylum seekers that we found earlier, it is not surprising that a policy of not allowing asylum seekers to work until their application has been decided upon (a process that in some countries can take several years) will act as an effective deterrent to those who are in a position to choose the country in which they want to lodge their application. Again this will not only be true for those applicants whose motives are primarily economic, but also for those forced migrants who have the choice between several safe host-countries. One common problem behind the other three measures which might at least in part explain their limited effectiveness, is clearly the issue of 27

The variation in countries’ treatment of “non-state agents of persecution” is a case in point.

466

Immigration and the Transformation of Europe

how much knowledge about specific policy measures asylum seekers can be expected to possess. In addition, there are a number of more specific problems with such measures that will also contribute to their limited effectiveness. In the case of safe third country provisions, problems arise as a state which wants to send a potential actual asylum seeker back to another safe third country can only do so if it can establish at least part of the migrant’s transit route. Often this proves difficult as asylum seekers are either unable or unwilling (following instructions from their traffickers) to provide such information. In particular with persons who apply for asylum only once they are already inside a country (rather than at the border), that country can only hope for the cooperation of transit countries who might have already registered these individuals. Judging from past experiences, this type of cooperation is often not forthcoming. Despite efforts by the EU to institutionalize such cooperation with the Schengen and Dublin Agreements and the joint EURODAC database, progress in this area has so far been limited (Noll 2000). Cooperation with countries outside the EU is even more difficult as here the application of safe third country provisions requires special bilateral or multilateral re-admission agreements. Therefore, the limited effects that safe third country provisions have had so far should not come as a surprise. Even in the case of the often quoted 71 percent drop in asylum applications in Germany from 1992 to 1994, which has often been attributed to the introduction of safe third country provisions with the 1993 changes to the German Basic Law, few observers appear to be aware of the fact that this drop happened against a 53 percent drop in overall applications across OECD countries and a 70 percent drop of applications from the former Yugoslavia, for which Germany had been the preferred country of destination in 1992. A similar explanation arguably lies behind the recent 42 percent drop in UK asylum applications (see Tables 17.6 and 17.7). These figures strongly support the argument that there are limitations when it comes to policy-based explanations for what have widely been regarded as “asylum policy success stories.” Regarding the remaining two measures – the denial of free movement and cash benefit payments – it is perhaps even less surprising that these measures have not deterred asylum seekers in any significant numbers. The prospect of personal safety from persecution and a green-card at the end of a successful determination procedure might even make the prospect of a few months on non-cash benefits in one of the less desirable regions of the host-country just about bearable. The effectiveness of dispersal regimes is further reduced by the fact that, short of a general policy of detainment, there appears to be little a host-state can do to prevent

Government attempts to control migration

467

Table 17.6 Questioning the effect of restrictive policy measures: The case of Germany

Asylum applications in Germany Asylum applications in Europe Asylum applications from former Yugoslavia in Europe Asylum applications from former Yugoslavia in Germany

1992

1994

Drop

438,191 (62% of total) 708,984 419,637

127,210 (38% of total) 335,811 124,149

71% 53% 70%

223,555 (52%)

55,034 (43%)

75%

Table 17.7 Questioning the effect of restrictive policy measures: The case of the UK

Asylum applications in UK Asylum applications in Europe Asylum applications from Iraq in industrialized world Asylum applications from Iraq in the UK

1992

1994

Drop

103,080 480,559 51,516

60,050 293,653 9,894

42% 39% 81%

14,570 (28%)

1,890 (19%)

87%

the movement of those determined to join relatives or friends in other parts of the country. Even the policy of withdrawing housing and welfare assistance in such cases, as practised by some states, has not always had a deterrent effect.28 Finally, the role of cash benefits payments as a pull factor for asylum seekers has without doubt been greatly exaggerated in the media and by policy-makers. Payments of benefits at a level that is often much less than the social assistance minimum will clearly be of limited attractiveness, in particular when many OECD countries have failed to effectively curtail illegal employment opportunities which promise vastly higher rewards. 28

Germany, for example, practises such a policy and has found it wanting in particular as asylum seekers assigned to some parts of Eastern Germany chose to forfeit assistance in the light of a disproportionately high incidence of racial violence in areas that until very recently had had very little experience with (non-white) foreigners. This suggests that dispersal schemes sometimes fail to achieve their goal of decreasing local residents’ adverse reactions towards asylum seekers, which is thought to result from the concentration of such groups in metropolitan areas.

468

Immigration and the Transformation of Europe

Conclusion This chapter has shown that some of the most prominent public policy measures aimed at regulating unwanted migration and at addressing the very unequal distribution of asylum seekers among OECD countries are less effective than has been assumed. In addition to the often mentioned institutional constraints that policy-makers face in this area (Soysal 1994; Joppke 1997), this chapter suggests four additional reasons. First, policy-making in this area sometimes appears to exaggerate the degree of choice and the level of information that asylum seekers and their agents are assumed to have. The evidence presented here suggests that asylum seekers who are in a position to choose between a number of alternative host-countries do so in a rational manner on the basis of some knowledge about the real or perceived differences between these states. However, we found little evidence for the claim that there is widespread and systematic “asylum shopping” to exploit differences in host-countries’ welfare provisions. Second, the empirical analysis has shown that the most powerful explanatory factors for an asylum seeker’s choice of host-country are not considerations of short-term welfare maximization, but legacies of migrant networks, relative employment opportunities and asylum seekers’ perceptions about the relative “liberalness” of particular hostcountries. This means that asylum destination choice is affected above all by “structural” factors that, at least in the short and medium term, are beyond the reach of asylum policy-makers. There are a number of plausible reasons why it might not be greatly surprising that individual deterrence measures will be overshadowed by these other pull factors. Most importantly, ties with friends or family are likely to prove very strong, even in the face of a country’s not-so-welcoming asylum regime. Moreover, path-dependent processes can be expected to play strong roles because of the sunk costs involved in the creation of forced migration networks. Restrictive immigration control policies create a profitable niche for those exploiting the black market of international migration. Organized trafficking gangs and individual “entrepreneurs” provide a range of services to migrants for which they are able to charge often extortionate fees.29 Networks between these traffickers, agents, potential migrants, and legal residents or citizens of destination countries are costly to build 29

According to IOM figures, fees for services such as the smuggling across borders, arranging forged documents and visas, organizing employment and lodging range from several hundred to over US $30,000 depending on which country of origin and which host country are involved. It is estimated that more than 70 percent of asylum seekers make use of such services.

Government attempts to control migration

469

and are unlikely to be given up lightly. Finally, a country’s reputation as a liberal, open, and fair society can be expected to evolve over decades, if not centuries, and is unlikely to be undermined overnight by the introduction of a particular set of restrictive asylum measures.30 Third, as states tend to copy deterrence measures introduced by other states, the desired impact of such attempts by one state to make its asylum policy more restrictive relative to other potential host-countries, is often limited to a very short-term first mover advantage. The rapid spread of “safe third country” provisions across Europe in the 1990s is perhaps the most prominent recent example of such processes of cross-country policy transfer that have become very common in this area. Finally, the effectiveness of unilateral policy measures can be expected to be further undermined by multilateral efforts of international policy harmonization (Thielemann 2004). Given the structural character of many of the pull factors identified above, we can expect attempts to harmonize asylum law across receiving countries to consolidate, rather than effectively address, existing disparities in the distribution of asylum burdens. By curtailing states’ abilities to use differentiated policy tools to counteract the effects of country-specific structural pull factors, policy harmonization initiatives, such as those currently developed by the EU (Thielemann 2003; 2005), further undermine the already limited capacity of states to effectively regulate asylum flows. While unilateral policy measures appear to be limited in their effectiveness, multilateral efforts of policy harmonization can be expected to prove counterproductive in addressing the problem of international burden-sharing in this area.  Beck, Nathaniel 2001 “Time-Series-Cross-Section Data: What Have We Learned in the Past Few Years?” Annual Review of Political Science 4(1)(June): 271–93. http://arjournals.annualreviews.org/doi/abs/10.1146/ annurev.polisci.4.1.271. Beck, Nathaniel and Jonathan N. Katz 1995 “What to Do (and Not to Do) with Time-Series-Cross-Section Data.” American Political Science Review 89(3) (September): 634–47. Bocker, ¨ Anita and Tetty Havinga 1998 “Asylum Migrations in the European Union: Patterns and Trends and the Effects of Policy Measures.” Journal of Refugee Studies 11(3) (September): 245–66. 30

In interviews, asylum seekers in the United Kingdom regularly mention its long-standing democratic tradition as one of the factors that attracted them to Britain (Robinson and Segrott 2002) – a reputation that the introduction of a voucher scheme is unlikely to challenge.

470

Immigration and the Transformation of Europe

Borjas, George J. 1990 Friends or Strangers: The Impact of Immigrants on the US Economy. New York: Basic Books. Cornelius, Wayne A., Philip L. Martin and James F. Hollifield 2004 Controlling Immigration: A Global Perspective. 2nd edn. Stanford, CA: Stanford University Press. Crisp, Jeff 1999 “‘Who Has Counted the Refugees?’ UNCHR and the Politics of Numbers.” New Issues in Refugee Research Working Paper No 12. Geneva: Office of the United Nations High Commissioner for Refugees, June. www.unhcr.ch/cgi-bin/texis/vtx/research/opendoc.pdf?tbl= RESEARCH&id=3ae6a0c22. European Council on Refugees and Exiles 2000 “Non-State Agents of Persecution and the Inability of the State to Protect: The German Interpretation.” European Legal Network on Asylum (ELENA) Research Paper. London: ECRE, September. www.ecre.org/research/nsagentsde.pdf. Fassmann, Heinz and Rainer Munz ¨ 1992 “Patterns and Trends of International Migration in Western Europe.” Population and Development Review 18(3) (September): 457–80. Freeman, Gary P. 1994 “Can Liberal States Control Unwanted Migration.” Annals of the American Academy of Political and Social Science 534 (July): 17–30. Guiraudon, Virginie and Gallya Lahav 2000 “A Reappraisal of the State Sovereignty Debate: The Case of Migration Control.” Comparative Political Studies 33(2) (March): 163–95. Hailbronner, Kay 1993 “The Concept of ‘Safe Country’ and Expeditious Asylum Procedures: A Western European Perspective.” International Journal of Refugee Law 5(1): 31–65. Harris, John R. and Michael P. Todaro 1970 “Migration, Unemployment and Development: A Two-Sector Analysis.” American Economic Review 60(1): 126–142. Hatton, Timothy J. 2004 “Seeking Asylum in Europe.” Economic Policy 38 (April): 5–62. Havinga, Tetty and Anita Bocker ¨ 1999 “Country of Asylum by Choice or by Chance: Asylum-Seekers in Belgium, the Netherlands and the UK.” Journal of Ethnic and Migration Studies 25(1) (January): 43–61. Hicks, Alexander M. 1994 “Introduction to Pooling,” in The Comparative Political Economy of the Welfare State, Thomas Janoski and Alexander M. Hicks (eds). New York: Cambridge University Press, 169–88. Holzer, Thomas and Gerald Schneider 2002 Asylpolitik auf Abwegen: nationalstaatliche und europ¨aische Reaktionen auf die Globalisierung der Fl¨uchtlingsstr¨ome. Opladen: Leske and Budrich. Holzer, Thomas, Gerald Schneider and Thomas Widmer 2000 “The Impact of Legislative Deterrence Measures on the Number of Asylum Applications in Switzerland (1986–1995).” International Migration Review 34(4) (Winter): 1182–1216. Hugo, Graeme J. 1981 “Village-Community Ties, Village Norms, and Ethnic and Social Networks,” in Migration Decision Making: Multidisciplinary Approaches to Microlevel Studies in Developed and Developing Countries, Gordon

Government attempts to control migration

471

F. De Jong and Robert W. Gardner (eds). New York: Pergamon Press, 186– 225. Jacobson, David 1996 Rights across Borders: Immigration and the Decline of Citizenship. Baltimore, MD: Johns Hopkins University Press. Joppke, Christian 1997 “Asylum and State Sovereignty: A Comparison of the United States, Germany, and Britain.” Comparative Political Studies 30 (June): 259–98. Joppke, Christian 1998 “Why Liberal States Accept Unwanted Immigration.” World Politics 50(2) (January): 266–93. Kjaergaard, Eva 1994 “The Concept of ‘Safe Third Country’ in Contemporary European Refugee Law.” International Journal of Refugee Law 6(4): 649–55. Kritz, Mary M., Lin Lean Lim and Hania Zlotnik (eds) 1992 International Migration Systems: A Global Approach. New York: Oxford University Press. Kunz, E. F. 1981 “Exile and Resettlement: Refugee Theory.” International Migration Review 15(142): 42–51. Massey, Douglas S., Joaquin Arango, Graeme Hugo, Ali Kouaouci, Adela Pellegrino, and J. Edward Taylor 1993 “Theories of International Migration: Review and Appraisal.” Population and Development Review 19(3): 431–66. Meyers, Eytan 2000 “Theories of International Immigration Policy: A Comparative Analysis.” International Migration Review 34(4): 1245–82. Neumayer, Eric 2004 “Asylum Destination Choice: What Makes some West European Countries more Attractive than Others.” European Union Politics 5(2): 155–80. Noll, Gregor 2000 Negotiating Asylum: The EU Acquis, Extrateritorial Protection, and the Common Market of Deflection. The Hague: Martinus Nijhoff Publishers. Pierson, Paul 2000 “Increasing Returns, Path Dependence, and the Study of Politics.” American Political Science Review 94(2) (June): 251–67. Ranis, Gustav and John C. H. Fei 1961 “A Theory of Economic Development.” American Economic Review 51(4) (September): 533–65. Robinson, Vaughan and Jeremy Segrott 2002 “Understanding the DecisionMaking of Asylum Seekers.” Home Office Research Study 243. London: Home Office Research, Development and Statistics Directorate, July. www.homeoffice.gov.uk/rds/pdfs2/hors243.pdf. Sassen, Saskia 1996 Losing Control? Sovereignty in an Age of Globalization. New York: Columbia University Press. Sjaastad, Larry A. 1962 “The Costs and Returns of Human Migration.” The Journal of Political Economy 70(5): 80–93. SOPEMI: Systeme d’Observation Permanente sur les Migrations. Various years. Trends in International Migration: SOPEMI. OECD Emerging Economies. Paris: OECD. Soysal, Yasemin Nuhoglu 1994 Limits of Citizenship: Migrants and Postnational Membership in Europe. Chicago, IL: University of Chicago Press. Stimson, James A. 1985 “Regression in Space and Time: A Statistical Essay.” American Journal of Political Science 29 (November): 914–47. Taylor, J. Edward 1986 “Differential Migration, Networks, Information and Risk,” in Migration, Human Capital and Development. Research in Human

472

Immigration and the Transformation of Europe

Capital and Development, Vol 4, Oded Stark (ed). Greenwich, CT: JAI Press, 147–71. Thielemann, Eiko R. 2003 “Between Interests and Norms: Explaining BurdenSharing in the European Union.” Journal of Refugee Studies 16(3): 253–73. Thielemann, Eiko R. 2004 “Why European Policy Harmonization Undermines Refugee Burden-Sharing.” European Journal of Migration and Law 6(1): 43– 61. Thielemann, Eiko R. 2005 “Symbolic Politics or Effective Burden-Sharing? The European Refugee Fund.” Journal of Common Market Studies 43(4): 807–24. Todaro, Michael P. 1989 Economic Development in the Third World. 4th edn. New York: Longman. US Committee for Refugees and Immigrants 1980–2002 World Refugee Survey. Washington, DC: USCRI. www.refugees.org/worldmap.aspx. United Kingdom Home Office 2002 “Reducing Asylum Applications.” Internal working paper in preparation for the Ministerial meeting of 15 May, on file with author. United Nations High Commissioner for Refugees 1999 Refugees, Asylum Seekers, and Others of Concern to UNCHR: Statistical Overview. Various years Geneva: UNHCR Population Data Unit. www.unhcr.ch/cgi-bin/texis/ vtx/home?page=statistics. Zolberg, Aristide R., Astri Suhrke and Sergio Aguayo 1989 Escape from Violence: Conflict and the Refugee Crisis in the Developing World. New York: Oxford University Press.

Index

African immigrants 6, 54, 55, 88, 104, 274, 275, 276, 277, 332 age dependency ratio (see also support ratio) 36–37 ageing 1, 2, 9, 11, 14, 15, 17, 36–37, 39, 41, 112, 135–36, 138, 141, 320, 327, 399 agenda 2010 (Germany) 407 agenda-setting 25, 364, 390, 437 agriculture 299, 306, 389, 413 Albania, Albanians 304, 411, 412 Algeria, Algerian 35, 260–61, 263, 265, 266–67, 301, 353 American Federation of Labor (AFL) 371, 386 Americanization 15, 396, 397 amnesty 282, 288, 289, 290, 294 Amsterdam Treaty 318, 319, 320, 423, 424, 428, 429 Annan, Kofi 30, 31, 33 apprenticeship, apprenticing 23, 241, 242, 245, 246 arena shopping Asia, Asians 6, 11, 35, 39, 41, 51, 54, 55, 66, 68, 85, 88, 90, 104, 105, 202, 209, 266, 275, 304, 332, 349, 400, 411 association agreements 430–31, 432 asylum, asylum seekers, asylum applications, asylum regimes 7, 19, 25–26, 68, 172, 174, 195, 285–86, 287–88, 289, 290, 292, 301, 304, 305, 316, 317–18, 319, 320, 321, 323, 324, 327, 328, 336–37, 338, 393, 394, 395, 397, 401, 402, 404, 405, 406–07, 408, 409, 410, 411, 412, 413, 414, 415, 424, 425, 430, 442–48, 449, 450, 452–69 asylum shopping 401, 424, 447, 468 Aussiedler 128 Australia 2, 3, 9, 51, 55, 85, 86, 149, 152, 153, 162, 163, 200, 202, 204, 206, 207, 208, 209, 210, 211, 213, 217, 218, 221, 222, 223, 301, 308

Austria 51, 55, 98, 100, 118, 128, 130, 131, 152, 156, 235, 236, 241, 242, 243, 244–45, 328, 329, 334, 371–72, 399, 435, 436 Belgium 55, 96–97, 98, 152, 235, 236, 241–42, 243, 244, 245, 252, 268, 300, 301, 302, 303, 306, 309, 311, 312, 322, 328, 431, 436, 461 Bismarckian welfare state 404–07 black economy 303 Blair, Tony 399 borders 1, 19, 23, 40, 258, 290, 324, 327, 328, 345, 352, 419, 425, 429, 452, 464 Borjas, George 43, 44, 46–47, 61, 68, 115, 119, 121, 147, 148, 149, 150, 151, 153, 162, 163, 174, 219, 443, 450 brain drain 135, 140 British Household Panel Survey 86, 126 Brussels 20, 25, 30, 262, 266, 302, 314, 419, 421, 425, 427 Buchanan, Patrick 384, 386 Bulgaria 58, 112 business, business interests 60, 66, 239, 289, 302, 344, 367, 370, 372, 396, 400, 415, 421 Canada 2, 3, 9, 51, 55, 85, 86, 98, 111, 149, 152, 153, 162, 163, 172, 173, 174, 177, 179, 180, 185, 186, 187, 188, 189, 192, 193, 194, 195, 200, 202, 206, 207, 208, 209, 210, 211, 213, 216, 217, 218, 221, 222, 223, 277–78, 280, 281, 286, 290, 291, 301, 308, 353 Cape Verdeans 276, 277 Central America 54, 55 Chandler, William E. 371–72 China 32, 39, 322, 411 Chinese Exclusion Act of 1882 371 Chirac, Jacques 18 Christianity, Christian 23, 151, 164, 166, 253, 255, 330

473

474

Index

citizenship 9, 90, 131, 173, 177, 178, 181, 195, 236, 253, 255, 257–58, 259, 264, 267, 292, 294, 308, 309, 321, 354, 370, 371, 386, 400, 405, 420, 426 Civil Rights Act of 1964 374 civil society 253, 256, 267, 268, 269 closed economy 114, 117–18, 121, 140 collective bargaining, collective wage setting 113 colonialism, colonial ties 60, 66, 259, 277, 281, 301, 302, 312, 401, 450 Common Market 111, 314, 420, 422 Commonwealth, British Empire 368–71, 450 Commonwealth Immigrants Act of 1962 369 commuting 275 Congo 302 Conservative Party, Tories 370, 380–81, 382, 389, 423 Consolidation Act of 1998 (Italy) 411 consultation (Italy) 254, 268, 269 Convention on the Future of Europe 424 corporatism 174, 179, 259 corporatist welfare state regimes 173–74 Council of Europe 319 Council of Ministers 314, 422 Council for the Muslim Religion (France; Conseil Français du Culte Musulman) 254, 267 country of origin effects, origin effects 22, 85, 147, 148, 149, 150, 153, 154, 156–65, 167, 353, 450–52, 453, 462 Cyprus 2 Czech Republic 49, 329, 335 de-commodification (of labor) 407 Democratic Party 371, 386–87 Denizenship 257 Denmark 15, 51, 91, 98, 99, 100, 102, 104, 128, 130–31, 152, 315, 319, 328, 329, 334, 409, 423, 424, 425, 428, 435 deportation 290, 293, 400 destination, destination country, destination effects 22, 44, 45, 46, 48–49, 58, 60–62, 64, 65, 66, 68, 85–86, 105, 113, 147, 148, 149, 150, 152, 153, 154, 156–65, 167, 274–94, 339, 353, 443, 447, 448, 450–52, 453, 454–55, 456, 457, 462, 465, 466, 468 deterrence, of asylum applications 406, 444, 446, 447–48, 452–53, 456, 458, 459, 461, 463–64, 468, 469 diaspora, delayed diasporization, African diaspora, Muslim diaspora 23, 251,

260, 261, 267, 274–75, 276, 277, 280, 281, 291, 293, 294 direct effect 428, 429 Directorate for Religious Affairs (Turkey, DITIB) 260, 261–62, 263, 264, 265 discrimination 22, 105, 148, 150, 166, 167, 187, 195, 241, 245, 252, 291, 312, 406, 416, 430, 432 dual labor market 398 Dublin agreement, Convention 318, 401, 466 earnings 6, 22, 45, 46, 85–105, 162, 163, 181, 186, 397, 404 East–West migration 114, 304 Eastern Europe 6, 49, 90, 99, 105, 128, 138, 141, 253, 275–76, 302, 304, 305, 329, 399, 404, 431 economic and monetary union, European monetary union, EMU 398 education, and national integration 23, 200–25, 235–46, 253, 406 educational achievement 23, 200–25, 397 educational attainment 91, 100, 142, 148, 149, 151–52, 155, 162, 164–65, 167, 168, 200–25, 237, 403 Egypt 35, 37, 134, 262, 264, 265, 411 elections, electoral politics 153, 253, 365, 368, 376, 377, 379, 380, 387, 388 elementary school, immigrant performance in 23, 205–11 Ellis Island 309 employer federations, business federations, business associations 113 employment rights Erasmus 294 Erbakan, Necmettin 262 Esping-Andersen, Gosta 173, 181, 394, 402, 407 Estonia 55 Ethiopians 274, 277, 411 ethnic capital 151, 167, 219 ethnicity, ethnic differences 9, 10, 40, 44, 62, 85, 86, 130, 150, 151, 176, 177, 188, 209, 212, 235, 236, 245, 246, 282, 285, 302, 308, 312, 330, 331, 337, 339, 340, 341, 342, 343, 344, 345, 348, 352, 353, 368, 373–74, 379, 387, 388, 397, 400, 401 eugenics 367, 371, 374 Euro-Arab dialogue 266 eurobarometer 331, 338, 339 European Commission 19, 25, 329, 390, 419, 420, 421, 423, 425, 426, 432–38

Index European Community Household Panel 22, 47, 86–88, 126 European constitutional treaty, European constitution 1, 19, 25, 258, 420, 425, 427, 438 European Convention on Human Rights 19, 258 European Council, Tampere European Council (1999), Justice and Home Affairs Council ( JHA Council) 315, 319, 389, 390, 426, 433, 435, 437 European Court of Justice 19, 25, 258, 315, 419, 420, 422, 423, 424, 425, 426–32, 434, 437–38 European Economic Area (EEA) 111–12 European Economic Community 313 European Free Trade Association (EFTA) 319 European integration 20, 236, 240–45, 252–53, 321, 419–36 European parliament 19, 20, 30, 40, 315, 419–36, 438 European social model 9, 15, 25, 26, 40 European Social Survey (ESS) 329, 331–32, 337, 339, 340–42, 350 European Union 1, 2–26, 30, 33, 54–58, 88, 90, 91, 96–100, 103–05, 111–12, 114–15, 126–31, 136, 140, 152, 176, 258, 274–75, 294, 300, 302, 310, 316, 317, 319, 320–24, 327, 329, 331, 336, 343, 345, 364, 390, 399, 400, 401, 409, 411, 419–36, 442, 443, 446, 458, 463, 466, 469 European Union enlargement, Eastern enlargement 19, 40–41, 117–18, 319 European Union Labour Force Survey (EULFS) 155 European Union Statistical Office (Eurostat) 86, 152, 274 euroskeptic, euroskepticism 280, 307, 315, 318, 324, 434, 435 expulsion 280, 435 fertility, fertility rates 11–17, 26, 32–36, 38, 39, 41, 138, 188, 301, 327, 399, 413, 436 Finland 49, 50, 55, 89, 90, 97, 100, 152, 329, 334, 436 fiscal policies, fiscal impact of immigration 43, 112, 118, 136–38, 394, 403–04, 413, 414 forced migration 442, 449, 468 foreign aid 19, 263, 280, 449 foreign direct investment (FDI) 134, 135, 141

475 former Soviet Union, former USSR 51, 58, 90, 98 Fortress Europe 293, 438 Fortuyn, Pim 328, 344 four freedoms 313, 422 framing, frames 23, 259, 364, 367, 368, 371, 372, 374, 375, 379, 384, 388, 389 France 6, 7, 8, 9, 14, 19, 24, 35, 37, 49, 50, 88, 89, 98, 100, 152, 205, 207, 208, 218, 219, 222, 223, 224, 241, 243, 252, 259, 261, 263, 269, 276, 301, 315, 316, 330, 335, 338, 339, 353, 362, 364, 366–71, 375, 376, 384, 387, 389, 400, 411, 423, 436, 442, 457, 461 Freedom House Index 61, 66, 70, 153 freedom of movement 312, 314, 315, 316, 430, 453, 457, 464 French Communist Party (PCF), Communists 376–78 Gastarbeiter, guestworkers 7, 259, 264, 311, 393–99 gender, differences in earnings 85–86, 89, 91, 96–103, 105, 165, 188 Geneva Convention on the Status of Refugees 444, 449, 452, 456 German Socio-Economic Panel 126, 132 Germany 9, 15, 17, 35, 37, 38, 47, 49, 55, 90, 91, 96, 97, 98–100, 102, 104, 105, 118, 128, 130–42, 152, 154, 159, 172, 173, 174, 179, 184, 185, 186, 187, 189, 190, 193, 194–95, 200, 202, 205, 206, 207, 208, 210, 211, 213, 218–23, 235, 236, 238, 241, 242, 243–45, 252, 254, 256, 259, 262, 263, 269, 276, 278, 280, 281, 285–90, 293, 300, 301, 303, 304, 310, 311, 312, 315, 316, 324, 327, 328, 329, 338, 339, 353, 362, 390, 394, 399, 400, 404–07, 408, 411, 414, 424, 435–37, 446, 447, 457, 461, 463, 466 Ghana, Ghanaian immigrants 23, 274–94 Gini index, Gini scores 61, 162 Global Commission for International Migration (United Nations) 30 globalization 15, 252, 269, 327, 354, 398, 400, 415 Greece 35, 49, 90, 96, 103, 152, 159, 304, 311, 329, 332, 338, 423, 436 Greens 17, 390, 428, 436 Hague Programme 390, 425, 428 Hammar, Thomas 257, 307 headscarf 252, 256, 258, 266 health care 353, 398, 401, 410, 457 House of Commons 365

476

Index

human capital 47, 116, 118, 135, 138, 147–50, 159, 162, 166 human rights 7, 9, 288, 299, 320, 324, 421, 457 human trafficking 299, 305, 323 Hungary 49, 51, 58, 329, 332, 335, 461 identity 9, 24, 257, 264, 267, 270, 299, 328, 330, 334, 340, 342–43, 352, 353, 363–64, 368, 370–71, 375, 380 illegal immigration 6, 7, 19, 59, 299, 303, 304, 305, 316, 317, 319, 320, 323, 411, 424, 425 Imams 252, 254, 258, 260, 261, 262, 264, 265 immigrant policy 307, 308 Immigration Law Practitioners’ Association (ILPA) 434 immigration policy 5, 17, 18, 19, 24–27, 39, 59, 60, 61, 85, 111, 149, 166, 172, 173, 196, 211, 224, 298–326, 327, 352, 362, 363, 364, 365, 366, 367, 369, 375, 388–90, 393, 397, 408, 409, 411, 412, 413, 415, 419–38, 468 immigration regimes 274, 275, 277, 285, 286, 288, 292, 293, 420, 421, 443, 462 Immigration Restriction League 372 India 32, 33, 39, 135, 177, 209, 283, 301, 411 industrial decline 302, 311 industrial districts 282, 283, 292, 294, 302 informal labor markets 275 intergovernmental, intergovernmentalism 19, 251, 313, 314, 316, 317, 318, 319, 420, 423, 424, 425, 426, 427, 428–29, 432, 435, 437 interior ministries 20, 254, 259, 262, 265, 267, 268, 269, 327, 393, 411 International File of Immigration Surveys 152 International Labour Organization (ILO) 48, 184 International Monetary Fund (IMF) 48, 70, 134, 140 International Socio-Economic Index (ISEI) 153, 154, 159, 221 involuntary immobility 276 Iraq 266, 269, 328, 411 Ireland 2, 5, 11, 88, 91, 97, 98, 100, 104, 128, 130, 131, 152, 316, 319, 329, 338, 362, 399, 401, 423, 424, 435, 457, 461 Islam, Islamic fundamentalism, Islamicization, transnational Islam, political Islam 7, 23, 151, 251, 299, 328, 341, 354

Islamic Commission (Spain, Comision ´ Isl´amica) 254 Islamic Community (Bavaria) 254 Islamische Gemeinschaft Milli Gor ¨ us ¨ (IGMG) 262, 267 Israel 265, 269, 329 Italian immigration 275 Italy, southern Italy, northern Italy, Veneto region 5, 275, 282 Japan, as destination country 2, 287, 291, 461 Jewish community 252 Jewish immigration 6, 304, 404 Jordan Commission on Immigration Reform 375, 384 Jospin, Lionel 18, 399 Judaism, Jews 23, 254, 256, 330 jus sanguinis 9, 17 jus solis 9 Kennedy, John F. 373, 374 Kok, Wim 399 Krasner, Stephen 252 labor market participation [Labour] 43, 62, 275, 277, 281, 283, 286, 291, 292, 293, 462 labor migrants 236, 237, 260, 267, 270, 274–75, 276, 280, 281, 282, 283, 284, 286, 290, 291, 293, 294, 400, 409, 412 Labour Party 380, 381, 382, 383, 388, 399 laissez-faire 263, 265, 270 language, linguistic ties 21, 44, 55, 60, 66, 68, 70, 103, 105, 150, 163, 281, 341, 401, 415, 450 Latin America 66, 68, 105, 301, 302, 411 Le Pen, Jean-Marie 8, 18 legalization 282, 288, 416 liberal, liberal welfare state regimes 173, 394, 396, 401–04 Lisbon Agenda (European Union) 400 literacy 61, 65, 69, 200, 207, 371, 372 Lodge, Henry Cabot 371, 372 long term residents directive 427, 432–36 Luxembourg 11, 35, 49, 51, 86, 88, 89, 90, 91, 96, 97, 98, 99, 104, 127, 128, 130, 152, 423, 428, 436 Luxembourg Employment Study (LES) 175, 184 Luxembourg Income Study (LIS) 175–76, 177, 178, 184, 191

Index Maastricht Treaty 25, 317, 318, 319, 420, 422, 423, 427 Madrid, 2004 bombings 7 Maghreb 367, 411, 413 magnet effect, welfare magnet effect 401 marital status 89, 155 May 1968, crisis of 367, 369 means-tested benefits 183, 412 mediatization 18, 298, 305 Mediterranean countries 304, 306, 411 Mediterranean welfare state 393, 394, 396 Mexico 47, 55, 90, 310, 375, 384, 388 Migration Board (Sweden) 409 migration, circulatory migration 275, 294 migratory flow 1, 2, 5, 7, 20, 33, 37, 38, 39, 43, 44, 46, 47, 48, 49, 51, 58, 60, 61, 62, 64, 65, 66, 68, 274, 298, 299, 301, 305, 306, 307, 309, 316, 321, 324, 353, 415, 416, 448, 450, 451 mobility, shuttle mobility, sequential mobility, migration mobility, mobility and gender, mobility rights, mobility enhancing programmes 43, 65, 148, 267, 274–94, 300, 307, 314, 316, 397 Morocco 156, 263, 264, 265, 311, 411, 412, 431 multicultural, multiculturalism 152, 242, 257, 328, 353, 373, 374 Muslim Brotherhood 152, 262, 263 Muslim Council of Britain 268 Muslim Executive (Belgium, Ex´ecutif Musulman) 254, 268 Muslim integration 251, 255 Muslim World League 260, 263 myths, of political identity 2, 9, 254, 276 National Front (FN, France) 349, 368, 376, 377, 378–80, 389 national integration model 240, 242 nationalism 327, 350, 353, 354, 371, 436 naturalization 9, 49, 177, 263, 328, 353, 390 neoclassical theory 45, 58, 450 neoliberal, neoliberal discourse 2, 6, 173, 181, 257, 396, 399, 401, 410, 414, 463 Netherlands 1, 6, 8, 9, 15, 35, 38, 49, 88, 90, 100, 103, 105, 152, 200, 205, 206, 207, 208, 210, 211, 213, 216, 218, 219, 222, 223, 224, 235, 236, 238, 240, 241, 242, 243, 244, 245, 277, 278, 280, 281, 286, 289, 291, 293, 315, 321, 328, 329, 330, 335, 338, 339, 400, 423, 436 network effects, migration networks, immigration networks 21, 24, 44, 45, 47, 59, 62, 64, 65, 66, 68, 105, 255,

477 256, 262, 266, 270, 275, 293, 294, 322, 348, 352, 365, 368, 394, 401, 447, 450–51, 453, 456, 462, 468 network externalities 45–46 New York 309, 383, 384, 386, 388 New Zealand 49, 55, 200, 202, 205, 206, 207, 208, 211, 215, 216, 218, 221, 222, 223 NGOs 260, 262, 268, 270, 424, 432, 457 Nigerians 274, 281, 290 North American Free Trade Area (NAFTA) 111 North–South migration 322 Norway 49, 172, 173, 174, 177–78, 179, 180, 184, 185, 187, 188, 189, 190, 193, 194, 195, 316, 329, 409 occupational status 22, 147–59, 346 oil crisis 111, 266, 302, 303, 310–13, 366, 394, 411 open economy 117–18, 121 Organisation for Economic Co-operation and Development (OECD) 21, 23, 200, 319, 453, 454, 456 Organization for Security and Co-operation in Europe (OSCE) 319 organized crime 299, 305 Palma document 315, 318 Paris 263, 266, 314, 379, 384 parties, political parties 17, 18, 19, 24, 151, 153, 154, 156, 164, 166, 259, 264, 328, 330, 352, 363, 364, 370, 377, 378, 380, 381–83, 386, 387, 389, 396, 399, 421 party systems 17, 18, 19, 20, 253, 379 pensions, pension reform 2, 14, 15, 126, 136, 320, 393, 395, 398, 403, 404, 409, 410, 412, 413 Philippines 31, 411 pieds noirs 301 pluralism, pluralist 253, 257, 258, 348 point system 149, 153, 156, 162, 166, 310 Poland, Poles 6, 35, 37, 49, 51, 58, 324, 334 police cooperation 298, 315, 316 political opportunity structures 256, 389, 432 population growth, population increase 2–5, 11, 116 population projections 1, 2, 21, 31, 32, 33, 34, 41, 200, 328, 454 Portugal 49, 88, 91, 96, 97, 98, 99, 100, 102, 104, 127, 152, 276, 332, 335, 339, 362, 399, 436, 442

478

Index

postindustrial society 18, 398 postmodern values 18, 398 poverty, poverty rates 135, 141, 172–96, 301, 303, 312, 398 Prodi, Romano, Prodi reforms of 1997 412 productivity 116, 135, 159 Programme of International Reading Literacy Study (PIRLS) 202–22 Programme of International Student Assessment (PISA) 202–19 pro-immigrant groups 20, 421, 424, 432, 434 proposition 187 (California) 375, 384, 386 public opinion 9, 20, 24, 298, 328, 330, 334, 338, 340, 341, 342, 346, 352, 353, 363, 364, 365, 366, 369, 370, 374, 375, 376, 377, 378, 383, 384, 386, 388, 390, 421, 444, 449 pull factors, for migration 58, 60, 61, 64, 65, 415, 443, 449–53, 454–57 race, racial differences 86, 151, 331–44, 449 race to the bottom 15, 398, 443 race policies, in Britain, in France 308, 315, 369, 374 Race Relations Act of 1965 370 racism, anti-racism 267, 302, 367, 369, 371, 374, 396, 406 real wages 43 refugee 43, 66, 68, 174, 237, 286, 290, 298, 300, 302, 303–04, 317, 318, 321, 327, 329, 337, 343, 397, 401, 402, 403, 408, 409, 410, 412, 413, 414, 442, 447, 448, 449, 451, 452, 456, 463, 465 religious communities 23, 251–59, 260, 266, 268, 270 religious integration 251, 252–53, 255, 259, 263, 264, 267, 269, 270 religious practices 23, 260, 261, 264, 268, 269 remittances 112, 113, 126–42, 397, 410 Republican Party 371, 372, 384, 386, 387 reputation, reputational factors in migration choices 26, 443, 451, 455, 469 retirement 2, 13, 14, 15, 327, 397, 412 Rhodes group 317 Right, radical Right, extreme Right, far Right 17, 18, 264, 299, 367, 375–81, 387, 389, 396, 398, 409, 421 right of residence 312, 431, 437

rights, post-national rights 257, 258 Romania 51, 58, 112, 411, 412 Rome Treaty, Treaty of Rome 111, 313 Roosevelt, Theodore 372, 373 Rotterdam 293, 328 Saudi Arabia, Saudi 35, 37, 260, 263, 264, 265, 266, 268 scams, migration 288, 289, 290, 291, 293 Scandinavian welfare state 393, 394, 395, 396, 401, 407–10, 414 Schengen agreement, Schengen Treaty, Schengen accord 19, 316–18, 319, 423, 424, 466 Schengen Information System (SIS) 316 Schroder, ¨ Gerhard 399, 407 second generation immigrants 201, 205, 207, 213, 218, 222, 223, 237, 241 secondary school, immigrant performance in 23, 91, 203, 206, 212, 213, 240, 241, 289, 406 secularism, secularization 135, 251, 253, 255, 257, 269, 341 security, insecurity, crime, criminality 1, 8, 9, 10, 13, 20, 24, 26, 265, 270, 290, 298, 299, 305, 313, 316, 317, 319, 320, 321, 323, 328, 332, 334, 363, 416, 419, 428, 429, 433, 451, 462, 463 segregation, school segregation 202–03, 212, 219–25, 409 Senegalese 261, 265, 274, 276, 277, 282, 283, 306 September 11, 2001, 7, 269, 321, 328, 363 single European Act 315, 422, 423, 427 skills, skill differentials 15, 21, 22, 43, 46, 47, 62, 105, 113–14, 118, 121, 127, 130, 138, 140, 142, 148–50, 155, 159–63, 167, 182, 191, 202, 204, 205, 209, 212, 213, 219, 222, 224, 238, 245, 280, 281, 292, 293, 294, 302, 306, 310, 321, 341, 390, 395–96, 398, 400, 405, 406–07, 413, 415 Slovak Republic 49, 51, 58 small and medium sized enterprises (SMEs) 283 social Darwinism 373 social democratic welfare state regimes 173, 194, 195 social Europe 431 social trust 344, 348, 349, 352 Socialist Party (PS, France) 18, 368, 376, 380, 387 socialization 348 Socrates 294

Index Somalis 274, 277, 283, 285, 328, 411 Soysal, Yasmin 257, 258 Spain 2, 5, 6, 49, 88, 96, 97, 98, 100, 102, 104, 127, 131, 152, 268, 300, 301, 304, 311, 318, 329, 334, 353, 362, 436 spillover 427–28, 429–32 Sri Lankans 135, 274, 278, 411 state-building 253, 256, 310 State–Church relations, State–Church institutions 259, 264, 267, 269–70 statewatch 434 Stoiber, Edmund 436 subnational regions 17 support ratio 36–37 supranational, supranationalism 111, 258, 298, 308, 313, 319, 343, 419–36 Sweden 49, 89, 90, 91, 96, 98, 100, 103, 104, 136, 172, 173, 174, 178, 179, 180, 184, 185, 187, 188, 189, 190, 193, 194, 200, 205, 206, 207, 208, 213, 216, 218, 219, 222, 223, 224, 238, 240, 278, 329, 332, 335, 337, 394, 400, 401, 407–10, 414, 415, 422, 436, 442, 446, 461 Switzerland 49, 55, 200, 202, 204, 205, 207, 208, 210, 211, 213, 216, 218, 219, 221, 222, 223, 224, 225, 290, 311, 329, 332, 335, 337, 339, 362, 442, 446, 448, 461, 463 taxes 15, 44, 61, 62, 65, 66, 68, 69, 126, 130, 136, 137, 255, 320, 327, 330, 334, 352, 394, 404, 414, 416 terrorism, anti-terrorism, terrorist 1, 7, 10, 153, 316, 328 Thatcher, Margaret, Thatcherism 399 third country nationals (TCNs) 420, 421, 422, 425, 426, 427–38 third generation immigrants 9, 252, 267, 393, 397, 405, 406, 412 trade unions 275, 310, 320, 371 Trends in International Math and Science Study (TIMSS) 201–09, 212, 213–19, 222 Trevi group 9 Turkey, Turkish 23, 35, 37, 51, 112, 135, 156–59, 235, 258, 260, 261–66, 301–02, 310, 311, 353, 406, 407, 431 Ukraine 19 Ummah 267 undocumented immigrants, migrants 277, 281, 282, 289, 293, 383, 394, 412, 413, 416 unemployment 1, 8, 10, 15, 22, 24, 25, 40, 45, 46, 61, 64, 65–68, 69, 85, 89,

479 112, 113–21, 130, 133, 137–41, 147, 152, 155, 164, 172–88, 238, 240, 241, 242, 245, 252, 267, 292, 301, 303, 311, 330, 337–38, 342, 346, 349, 378, 384, 403, 404, 405, 406, 407, 410, 412, 414, 431, 433, 434, 454, 459, 461 unemployment insurance 137, 180, 181, 183 UNESCO 260 Union des Organisations Islamiques de France 263 United Kingdom, UK, Britain, Great Britain 7, 35, 47, 49, 55, 88, 90, 97, 98–100, 103, 104, 152, 159, 172, 173, 177, 179, 185, 186, 188–89, 190, 191, 193, 194, 195, 200, 206–11, 213, 218, 222, 252, 276, 280, 281, 284, 285, 286, 287, 288, 290, 292, 293, 318, 335, 339, 362, 369, 393, 394, 399, 401–04, 414, 415, 423, 424, 430, 434, 435, 436, 437, 444, 447, 448, 450, 456, 457, 461, 466, 469 United Nations 30, 31, 33, 35, 39, 48, 111, 136, 307 United Nations High Commissioner for Refugees (UNHCR) 443, 453, 454 United States 1, 2, 3, 6, 15, 19, 24, 46, 47, 49, 55, 69, 85, 98, 105, 111, 121, 135, 152, 153, 159–63, 172–74, 177, 179, 185, 186, 189, 190, 191–95, 205, 207, 208, 209, 211, 213, 216, 223, 266, 277, 280, 287, 289, 290, 291, 298, 300, 306, 308–09, 353, 354, 362–92, 395, 396, 400, 407, 409, 456 unskilled migrants 113, 114, 121, 293, 320, 330, 397, 398, 406 Van Gogh, Theo 328 visa, visa applications, visa overstaying 149, 264, 275, 286, 288, 289, 293, 305, 310, 311, 312, 314, 315–16, 324, 413, 428, 429, 447, 452, 468 vocational training, vocational schools 23, 245, 433–34 Voting Rights Act of 1965 374 vouchers, for asylum seekers 406, 457–58, 464 wage dispersion 46 wage rigidities 113, 114, 116, 140 Wahhabi 260 welfare capitalism 394 welfare drain 8, 393, 395, 400, 403, 404

480

Index

welfare shopping 401, 436 welfare states, welfare benefits 1, 2, 9, 25, 44, 46, 47, 68, 89, 114, 115, 140, 192–96, 299, 375, 394, 397, 402, 403, 407, 409, 410, 412, 413, 448, 452, 457 welfare tourism 401 Wilson, Pete 384 work permits 303, 311, 314, 390, 402 workfare, workfarism 395, 407

Working Party on Migration and Expulsion 11–17, 436 working poor 342 World Bank 48, 154 Yemen 35 Yugoslavia 51, 58, 135, 284, 311, 411, 463, 466 zero immigration 38, 137, 310, 313, 320, 321, 323, 324