Pioneers of European Integration: Citizenship and Mobility in the EU 1848446594, 9781848446595

Table of contents :
Copyright......Page 5
Contents......Page 6
Contributors......Page 7
Acknowledgements......Page 8
1. Pioneers of European integration:an introduction......Page 10
2. The demographics of movers and stayers in the European Union......Page 35
3. Deciding to move: migration projects in an integrating Europe......Page 61
4. The social mobility of mobile Europeans......Page 81
5. Living across cultures in a transnational Europe......Page 107
6. More mobile, more European? Free movement and EU identity......Page 129
7. EU movers and politics: towards a fully-fledged European citizenship?......Page 165
8. A common information space? The media use of EU movers......Page 188
9. Internal and external movers: East–West migration and the impact of EU enlargement......Page 214
References......Page 233
Appendix A: Methodological notes......Page 250
Appendix B: EIMSS questionnaire (English version)......Page 264
Appendix C: External movers’ experiences of migration and integration into the EU15: interview guideline......Page 298
Index......Page 306

Citation preview

Pioneers of European Integration

A Cosimo e Giovanni, veri pionieri

Pioneers of European Integration Citizenship and Mobility in the EU

Edited by

Ettore Recchi University of Florence, Italy

Adrian Favell Aarhus University, Denmark

Edward Elgar Cheltenham, UK • Northampton, MA, USA

© Ettore Recchi and Adrian Favell 2009 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited The Lypiatts 15 Lansdown Road Cheltenham Glos GL50 2JA UK Edward Elgar Publishing, Inc. William Pratt House 9 Dewey Court Northampton Massachusetts 01060 USA

A catalogue record for this book is available from the British Library Library of Congress Control Number: 2009930859

ISBN 978 1 84844 659 5 Printed and bound by MPG Books Group, UK

Contents List of contributors Acknowledgements

vi vii

1 Pioneers of European integration: an introduction Adrian Favell and Ettore Recchi 2 The demographics of movers and stayers in the European Union Michael Braun and Camelia Arsene 3 Deciding to move: migration projects in an integrating Europe Oscar Santacreu, Emiliana Baldoni and María Carmen Albert 4 The social mobility of mobile Europeans Ettore Recchi 5 Living across cultures in a transnational Europe Antonio Alaminos and Oscar Santacreu 6 More mobile, more European? Free movement and EU identity Nina Rother and Tina M. Nebe 7 EU movers and politics: towards a fully-fledged European citizenship? Anne Muxel 8 A common information space? The media use of EU movers Damian Tambini and Nina Rother 9 Internal and external movers: East–West migration and the impact of EU enlargement Adrian Favell and Tina M. Nebe References Appendix A: Methodological notes Michael Braun and Oscar Santacreu Appendix B: EIMSS questionnaire Appendix C: External movers’ experiences of migration and integration into the EU15: interview guideline Index

v

1

26 52 72 98

120

156 179

205

224 241 255 289 297

Contributors Antonio Alaminos, Universidad de Alicante María Carmen Albert, Universidad de Alicante Camelia Arsene, University of Warwick Emiliana Baldoni, University of Florence Michael Braun, Zentrum für Umfragen und Methoden Analysis (ZUMA), Mannheim Adrian Favell, Aarhus University Anne Muxel, CEVIPOF (Centre de Recherches Politiques de Sciences Po), Paris Tina M. Nebe, UNFPA, Geneva and Rabat Ettore Recchi, University of Florence Nina Rother, Bundesamt für Migration und Flüchtlinge (BAMF), Nürnberg Oscar Santacreu, Universidad de Alicante Damian Tambini, London School of Economics and Political Science (LSE)

vi

Acknowledgements When we are older and talk to our grandchildren, how will we describe the age in which we lived? Perhaps we shall say that we saw the collapse of the Berlin wall and the almost simultaneous – and perhaps not unrelated – disappearance of century-old borders dividing nation states in Europe. EU citizenship and EU enlargements – offering the inherent possibility of unrestricted geographical mobility within an expanding Union – have transformed Europe forever. The notion of a frontierless continent would have been a utopia only a generation before us. Still, we have had the good fortune to experience it. The free movement of EU citizens is the immediate and most visible sociological consequence of this political and peaceful (r)evolution – arguably the noblest, bravest and most popular achievement of the process of European integration so far. The human dimension of this phenomenon is the focus of this book and the project it presents. We started thinking and talking about the book in the gloomy Christmas of 2001, over a telephone line between Madrid (Ettore) and Los Angeles (Adrian). Ettore was about to submit a comparative research proposal to the European Commission to carry out a quantitative-oriented study of moving Europeans, while Adrian was in the middle of his own ethnographic investigation of the same groups of people in Amsterdam, London and Brussels. We became aware of each other’s converging interests through Damian Tambini, another former classmate of ours from a place that was the seedbed of this common focus – the Department of Social and Political Sciences of the European University Institute, Florence – in the early 1990s. Eventually, the proposal was generously funded by the European Commission within the scope of its Fifth Framework Programme of research, and became the PIONEUR project: Pioneers of European Integration ‘From Below’: Mobility and the Emergence of European Identity Among National and Foreign Citizens in the EU (contract HPSE-CT-2002-00128). Ettore served as coordinator of the project at the Centro Interuniversitario di Sociologia Politica (CIUSPO, University of Florence); Antonio Alaminos (OBETS, Universidad de Alicante), Michael Braun (ZUMA, Mannheim), Anne Muxel (CEVIPOF, Paris) and Damian Tambini (CSLS, Oxford University) directed the four other partner institutions. In one way or another, all contributors to this volume were intensively involved in the different stages of the project. The vii

viii

Pioneers of European integration

successful implementation of PIONEUR was possible also thanks to the following persons: Giulia Amaducci in DG Research of the European Commission; Emanuele Lepri, Matteo Abbate and Nicoletta Baistrocchi at the University of Florence; Maxime Vanhoenacker, Cécile Riou and Viviane Le Hay at CEVIPOF in Paris; Joshua Hatton at Oxford University; Kristin Surak at UCLA; Miriam Baumgärtner and Lena Voecklinghaus at ZUMA in Mannheim, as well as Frank Faulbaum and his team at SUZ (University of Duisburg). PIONEUR is the primary, but not the sole, basis of the evidence and ideas contained in this volume. As said, Adrian brought the insights gained in his fieldwork, published in 2008 as Eurostars and Eurocities: Free Movement and Mobility in an Integrating Europe (Favell 2008a). Cooperation with DG Employment, Social Affairs and Equal Opportunities during the ‘European Year of Workers’ Mobility’ offered the chance to collect additional information used in the first chapter of this book. In DG Employment, we are particularly grateful to Christoph Maier, while Letizia Mencarini and Francesca Francavilla helped in the analysis of Eurostat data. Over the years, we had many occasions to discuss the concepts, methods and findings of our study with numerous colleagues and friends, whose comments spurred us on to improve the work. In particular, we would like to mention Martin Kohli, Yossi Shavit, Gloria Pirzio Ammassari, Massimo Livi Bacci, Max Haller, Moshe Semyonov, Hiroshi Ishida, Vittorio Cotesta, Louise Ackers, Riccardo Scartezzini, Juan Díez Medrano, Neil Fligstein, Simon Hix, Sophie Duchesne, Anja Weiss, John Salt, Phil Rees, Russell King and Claire Wallace. Gianfranco Bettin encouraged this line of research from the beginning with his unique spirit and thoughtfulness. In the editorial stage, Letizia La Barba helped us assemble the bibliography. Finally, we are grateful to Catherine and Edward Elgar for their support, and to Jenny Wilcox and David Fairclough for assisting us in the publication process. Ettore Recchi and Adrian Favell Firenze and Aarhus, January 2009

1.

Pioneers of European integration: an introduction Adrian Favell and Ettore Recchi

INTEGRATING EUROPE ‘FROM BELOW’: THE ROLE OF FREE MOVEMENT The European Union stands as a unique economic, political, legal and social experiment in transnational regional integration. The world we live in may still be one primarily organized by and for territorial nation states, but if one empirical example is to be sought of how a post-national or cosmopolitan polity and society might be built, the EU is the only actually existing institutional example. Built on a regional territorial logic, its complex structures are also the best guide to the way a progressive and governable political order might be constructed from the economic free-for-all of globalisation. In no other part of the world have sovereign nation states bonded together to voluntarily relinquish large aspects of their sovereign control of economy and polity to a set of common supranational institutions. And in no other part of the world have such institutions created a form of post-national citizenship within a transnational regional political order. Arguably the most fundamental part of the traditional nation state’s claim to sovereignty is its claim of territorial jurisdiction over the member citizens that live within its borders (Torpey 2000). Yet at the heart of the European Union lies the principle of free movement: of the capital, goods and services that oil the wheels of international trade and business, but also of persons who, within its realm, now have the right to move, travel, study-work, settle and retire anywhere within its member states. EU citizens can move and demand equal treatment to that of all nationals of the same territory in every dimension of work and public life. This revolutionary principle has existed within European law since the Treaty of Rome in 1957; with the expansion of the EU to 25 members in 2004 and 2007, it now extends the notion of European citizenship – and potentially of a single European society – all the way from the Atlantic to the Urals and the borders of the Black Sea (European Commission 2002). 1

2

Pioneers of European integration

Curiously, though, scholars have been slow to explore the social effects of European integration. EU studies are dominated by policy and legal studies, the sweep of intergovernmental and functionalist theories, macroeconomic models of monetary union and the detail of diplomatic history. There is much less work on the consequences – or, indeed, the sources – of the integration process, in the lives and activities of the almost 500 million citizens who now make up this extraordinary experiment (Favell 2006a). Freedom of movement, in fact, is the core right of European citizenship: it is the right most often cited by Europeans as the single most valuable benefit of that membership (European Commission 2006).1 As a social phenomenon, then, cross-border movement within Europe – the spatial mobility of European citizens – ought to be seen as one of the key dimensions of European integration. Indeed, when thought of more broadly, it is a key indicator of the very possibility of post-national global or regional integration at the individual, human level. Many European citizens are using these rights to move to other countries of the EU. Large numbers of southern workers used newly minted freedom of movement accords in the 1950s and 1960s to migrate north to the expanding industrial economies of the north. This worker-based migration still continues in modified form. Now, though, many other categories of mover are also visible, particularly among the citizens of the EU15 member states, who have enjoyed these rights for decades. From 1987 to 2005, 1.4 million European students benefited from Erasmus and Socrates programmes to spend up to two years at a foreign university. Likewise, 12 000 advanced researchers and scientists have received support to move and work abroad.2 Major cities like London, Amsterdam, Paris, Milan, Barcelona and Munich have become dynamic hubs of the intra-European mobility of young workers, who choose to look for work on an individual basis. These possibilities have particularly impacted the life and career choices of provincial and peripheral citizens of the EU frustrated with their national opportunities, and of young women or those with unconventional lifestyles, such as homosexuals, who have used international mobility as an avenue to self-liberation. Furthermore, an ever-increasing number of retirees and near-to-retirement citizens from the north of Europe are deciding to pack up and buy a house in the south. There is, as Russell King calls it (2002), a ‘new map of European migration’. All of these potential avenues to European freedom are now, or will be at some point, accessed by new European citizens from new member states since the enlargements of 2004 and 2007 (Favell 2008b). These are, simply put, extraordinary movements. For Europe is a continent better known for low levels of cross-national, even cross-regional movement, and very high expectations of sedentary regional and national identification.

Pioneers of European integration: an introduction

3

Perplexingly, these on-the-ground social shifts are not always seen in the statistics. These sometimes frame the story in a different way. Debate about freedom of movement of workers focuses on the surprisingly low levels of intra-state resettlement, particularly when compared to the dynamic crossstate and regional movements of Americans (European Foundation for the Improvement of Living and Working Conditions 2006; see also Recchi 2005 and 2006). The USA remains a potent federal model of how a ‘United States of Europe’ might look, and its economic dynamism in these terms is the inspiration for the EU’s own adherence to notions of mobility and free movement. In fact, when looking at statistics, fewer than one in 50 Europeans lives outside their nation of origin. Around 4 per cent meanwhile have some experience of living and working outside their country of origin. These are figures that suggest the population of ‘EU movers’, as we call them, is a marginal, if not statistically insignificant counterpoint to the vast majority of ‘EU stayers’, those people who stay living and working in their own country. However, the symbolic importance of the moving EU population is not lessened by the numbers. Movement and mobility have huge effects on those involved, both those who move and those who encounter movers. In each and every one of these lives, the hopes and aspirations of the architects of the European integration process are inscribed. EU movers are the prototypical ‘Highly Europeanized Citizens’. They are the human face of European integration, from whom we might learn what it means to be a European. Their lives and experiences are the best guide to finding out how easy it is to shift one’s identity or horizon to a post-national or cosmopolitan level, and of the practical benefits, insights, barriers and failings of a life lived outside the place where you historically belong (see also Favell 2008a). Pioneers of European Integration sets out to explore the new Europe being built by these individuals. It represents the first systematic, quantitative attempt to study the impact and consequences of European freedom of movement of persons. Drawing on a multi-headed international research project, with a large scale original survey – the European Internal Movers Social Survey (EIMSS) – at its heart, it presents findings and evidence on all the big questions posed by the movement of European citizens within a relatively static continent. The project, funded by the European Commission (Fifth Framework Programme of research, 2003–06) was titled Pioneers of European Integration ‘From Below’: Mobility and the Emergence of European Identity Among National and Foreign Citizens in the EU, and was directed by Ettore Recchi. It was best known by its acronym PIONEUR.3 EIMSS surveyed movers from the five most populous EU15 member states (Germans, French, British, Italians and Spanish) to the five most

4

Pioneers of European integration

populous EU15 member states (Germany, France, Britain, Italy and Spain). It was conducted in 2004. One of the largest-ever comparative migrant surveys, EIMSS surveyed a total of 5000 European citizens who settled abroad in EU member states after 1973.4 The questions were designed to allow for the highest possible comparability with the European Social Survey (ESS) and Eurobarometer (EB), in order to highlight the specificities of EU movers versus EU stayers. For more technical information, we include a methodological appendix at the end of the book (Appendix A). EIMSS permits a vast range of questions to be asked. Who are the EU movers? Where do they come from regionally and socially, and where do they go? How well do they acclimatize to their new settings, and how do they self-select? Why did they move? How do their values, orientations and identifications compare to the EU stayers? What are the consequences of their political participation and media consumption as cross-national European citizens? Are these migrants somehow building a new European identity – for themselves, if not for the continent? And how do movers from the pre-2004 EU15 member states compare to the new movers that we will see taking up these rights from the A12 member states who joined the EU in 2004 and 2007? If nothing else, the backgrounds, experiences, social trajectories, careers and (perhaps) transformations of these prototypical EU movers will contrast sharply with the more stable and familiar patterns of class, values and participation of the majority European population of EU stayers, who remain spatially attached to their nations, regions and localities. EU movers’ behaviour and experiences are also a good indicator of new forms of mobility that deviate from the norms of international and regional migration. The subject may also offer the key to new thinking in social mobility research, if spatial mobility in Europe is linked – as can be hypothesized – to upward social mobility. Researchers have struggled to transcend the methodological nationalism that is inherent in social stratification tied to comparison of social mobility within national societies (Breen and Rottman 1998). Cross-national spatial mobility has been recognized as a key area for testing new avenues of social mobility potentially linked to global and regional economic processes (Breen 2004), and EIMSS allows us to do this. Pioneers of European Integration answers these questions with systematic evidence and analysis. In the remainder of this introduction, we provide some background to the project. First, we sketch the history of free movement rights in terms of European law and its evolution. Second, we consider briefly the underlying economic and political rationale for their expansion. Then, in the third section, we offer an overview of existing

Pioneers of European integration: an introduction

5

statistics on migration within the EU, drawing on various international sources. Finally, there is an introduction to the chapter-by-chapter empirical analyses that follow in the book.

A SHORT HISTORY OF FREE MOVEMENT RIGHTS The origins of free movement rights in the European Union date back to the early 1950s. Free movement was in fact first introduced by the pioneer supranational organisation of shared economic interests, the European Coal and Steel Community (ECSC) formed in 1951, to facilitate specialised workforce recruitment across national borders. Among the six founding states of the Community (Germany, France, Belgium, the Netherlands, Luxembourg and Italy), Italy was particularly keen to support this goal, in order to lower domestic unemployment and underemployment, as well as to improve the living conditions of nationals who had already migrated abroad (Romero 1993; on the development of free movement rights, see also Wiener 1998; Maas 2007). The ECSC Treaty limited free movement to ‘workers who are nationals of member states and have recognized qualifications in a coalmining or steelmaking occupation’ (article 69), but the right of free movement was generalized in the founding Treaty of the European Economic Community signed in Rome in 1957. Article 48 of this Treaty affirmed the right to accept offers of employment made in another member state and to move freely within the Community, as well as to reside and remain in another member state after having been employed. In its original version, however, the Treaty limited the right to move to ‘workers’ rather than ‘citizens’. At this stage, Community law openly treated migrants as production factors rather than persons tout court, in line with the market-oriented view of European integration. While bilateral agreements between national governments within Western Europe had been established soon after World War II, and extended later to more peripheral countries such as Portugal and Yugoslavia, the EEC treaty built a more solid legal framework for an intra-European mass migration system of its age. In short, free movement met the interests of both would-be foreign workers in Italy and potential employers in Germany, Benelux and France. However, the legal enactment of free movement proceeded at a slow and discontinuous pace (for more detail on the EU’s legal evolution and jurisprudence, see especially O’Leary 1996; Guild 1999; White 2004; De Bruycker 2006). For more than a decade, in spite of the Rome treaty, citizens of EEC member states who intended to work in a different member state continued to be subject to national immigration laws. They had

6

Pioneers of European integration

to apply for work and residence permits, which could be discretionarily denied – just like any other immigrants. According to article 49 of the Treaty, free movement was a fully intergovernmental policy left to decisions taken in the Council of Ministers. Central and Northern European EEC member states resisted the interference of supranational regulations on their sovereign power to control aliens. More specifically, in these countries it was often argued that free movement would have given some competitive advantage to Italians over other southern European migrants, making them less disposable as guest workers. The real implementation of free movement was thus postponed to 1968, when Council Regulation 1612/68 and Council Directive 68/360 abolished movement and residence restrictions on member state workers and their families in the entire EEC territory. On the one hand, Regulation 1612/68 made illegal all nationality-based discrimination between workers of member states, in terms of work conditions, salary and unemployment benefits. Furthermore, it established the foreign workers’ right to the same social and tax benefits as national workers, including access to training in vocational schools and housing benefits (where existing). Family members of foreign workers were entitled to reside with them and to be allowed access to any kind of employment in the host country. On the other hand, Directive 68/360 reduced the bureaucratic formalities of moving within the EU considerably, recognising the workers’ and their families’ rights to enter a different member state by simply showing an identity card or valid passport, without being forced to obtain a visa. Migrants within the EU were also entitled to a residence permit, with a validity of at least five years and an automatic renewal, by presenting an employment certificate. In addition, permanent residency thereafter in the host country was also guaranteed, as the residence permit could not be withdrawn in the case of involuntary unemployment. Neither could being unemployed justify expulsion. The 1968 provisions thus represented a turning point. They ended the transitional regime set by article 49 of the Treaty, and created the conditions for a full exercise of free movement rights. In the following decades, admission, residence and equal treatment of foreign residents from other member states were dealt with by a vast secondary legislation. Community law and the European Court of Justice (ECJ) have increasingly widened the matter and scope of the right to free movement originally contained in the Treaty. In particular, since the 1970s, the ECJ has played a key role in widening the scope of free movement. It has systematically shifted its focus from the free movement of workers to the free movement of persons – that is, movement independent of an individual’s capacity as an economic actor. Following high-profile cases brought by citizens, the

Pioneers of European integration: an introduction

7

Court was able to broaden interpretation of the Treaty, emphasising the citizenship dimension of free movement. According to the ECJ, the right to equal treatment implies a fully-fledged integration, not only in the job market, but in the whole of society, including social, cultural and educational aspects of workers’ and families’ lives. With the decisive contribution of the ECJ jurisprudence, laws on free movement were extended in the 1970s to foreign self-employed workers, and in the 1980s to foreign EU nationals who take up a paid apprenticeship, those who enter university in a member state different from their own after having taken up a job activity, and to seasonal workers. It can also be noted, though, that the controversial position of posted workers – workers who are temporarily sent to another member state to perform services there and who return to their country of origin after completion of their work – was clarified only 20 years later, with Council Directive 96/71. The Single European Act of 1986 aimed at creating an ‘area without internal frontiers in which the free movement of goods, persons, services and capital is ensured’. A logical consequence of this was the step taken in 1990, when freedom of movement and residence was now explicitly extended to non-economically active categories (as well as their families): students, pensioners and the unemployed. The legal bases of these changes are to be found in Directives 90/364, 90/365 and 90/366 (this last was then replaced by Directive 93/96). Such arrangements are still subject to two conditions from which workers are exempted: students, pensioners and the unemployed must have sickness insurance, and they must have sufficient resources to avoid becoming a burden on the national health systems or social assistance of the host member state. Another collateral advance was the adoption of the Schengen system, which took place progressively in the 1980s and 1990s, to eliminate passport controls between EU national borders. Symbolically, however, the most spectacular step in the process took place with the Treaty on the European Union, signed in Maastricht in 1992, and which entered into force on 1 November 1993. The Maastricht Treaty introduced citizenship of the European Union to ‘reinforce the protection of the rights and interests of the nationals of its member states’. Concretely, EU citizenship consists of a set of rights allowing all European nationals to vote and stand as candidates in elections of the European Parliament in the member state they reside in, regardless of their nationality; to submit petitions to the European Parliament and appeal to the EU Ombudsman; to be protected by the consular authorities of another member state in third countries that lack diplomatic representation of one’s state; and to move and reside freely in the territory of any of the EU member states.

8

Pioneers of European integration

The first three provisions are small in scope compared to the last. For the ordinary EU citizen, access to diplomatic protection in third countries is an extremely unusual event and petitions to the European Parliament or the Ombudsman look like quite remote options. Meanwhile, to vote as a foreign resident for the European Parliament has proven of little political relevance, with high abstention rates. Local voting rights are more widely used. But it is the rights to free movement and settlement in the entire EU territory that form the most potent cornerstone of EU citizenship. In more recent years, Commission Directive 2004/38 consolidated the residence rights of EU movers throughout the Union territory. This Directive not only unifies pre-existing scattered legislation on free movement, but also incorporates some key tenets of the jurisprudence of the ECJ. In particular, it acknowledges the right to permanent residence of EU citizens after five years of stay in any other member state without additional formalities (such as a carte de séjour), as well as the host state’s responsibility for the social welfare of movers. Overall, this Directive defines three categories of mobile EU citizens – short-term movers (less than three months), long-term (between three months and five years), and permanent (over five years) – whose social welfare rights in the host country are correspondingly graduated. Apparently the only remaining limits to the universalisation of free movement rights in the EU are the ability to vote in the general elections of the host country – which remains the ultimate seal of acquiring national citizenship; the unconditional access to state-based social benefits – inactive movers still need to have their own sickness insurance; and (more often than not) the non-cumulativeness of pension benefits gained in different member states. These are significant but not necessarily unsurpassable barriers to the achievement of some form of post-national citizenship in the EU. And inasmuch as free movement rights are associated with welfare rights that are insensitive to nationality, they foster the transnationalisation of social solidarity in Europe – the potential construction of a fully harmonised if not unitary welfare system. While the process is controversial and contested, it clearly goes a long way from a conception of the EU as a mere network of intergovernmental relationships. If only for this, free movement undermines the intergovernmentalist description of the European Union. The potential of free movement of persons for the deepening of the process of European integration can thus hardly be underestimated. One has only to be reminded that the full right to travel was not established as a constitutional citizenship right in the US before the 1940s, and that when it was, it was taken to be a fundamental step towards the creation of a fully-fledged federal state.

Pioneers of European integration: an introduction

9

The EU enlargements of 2004 and 2007 again put citizenship mobility at the top of the political agenda in Europe. Heated media debates – for example on the possible invasion of ‘Polish plumbers’ into France – drew big headlines. As in previous enlargements – that is, when Greece (1981), Spain (1986) and Portugal (1986) joined the European Union – the movement rights of the new European citizens were subject to temporary restrictions. Notably, though, Britain, Ireland and Sweden all opened their doors to the new citizens, and saw substantial in-migration. In 2006 and 2007, the mobility restrictions for citizens from the so-called A10 new EU member states (those who joined in 2004) were mostly lifted in all other member states, with the exception of Germany and Austria. These countries, which have had some of the largest numbers of informal migrants from Central and Eastern Europe, announced that they would maintain national measures until April 2009. For workers of the two new member states joining the EU in 2007 – Romania and Bulgaria – only Finland and Sweden among EU15 states granted full mobility rights from the start. But progressive openings took place elsewhere quite rapidly on a piecemeal basis, with eased conditions of access to job permits for workers from these countries.5 To sum up, the free movement of persons across national borders within the EU – a key aim of the overall European integration process – has, step by step, become an everyday feature of European economy and society. The two major legal steps in this process were the abolition of restrictions on movement and residence for workers of member states and their families in 1968, and the introduction of EU citizenship in 1992, alongside the less visible but decisive actions of the European Court of Justice. The Court has stood as a bulwark against all attempts to maintain privileges rooted in pre-existing or re-emerging nationality-based pieces of legislation. Migrant ‘workers’ – as they were called in the ECSC treaty of 1951 – have become EU ‘movers’, as they are now usually referred to in EU documents.

ECONOMIC AND POLITICAL RATIONALES FOR FREEDOM OF MOVEMENT As we have seen, free movement represents a constant and quintessential concern of the European Union. But why has the EU (and its earlier incarnations) always insisted on promoting free movement? Both economic and political rationales have been advanced. In this section, we review them briefly (see also Recchi 2008). Generally speaking, in the collective imagery of advocates of European integration, cross-border migration within Europe – whatever its shifting

10

Pioneers of European integration

political boundaries – is framed and sustained as the sociodemographic counterpart to international trade in an unrestricted single market. Together, a single market and a free movement area would be the triggers of an ‘ever closer Union’. In this respect, the project of European integration anticipates the globalisation of the last decades of the 20th century. Indeed, it might be said that the EU constitutes a world-regional variant of globalisation (Castells 1998; Favell 2003). In somewhat more technical terms, economists conceive of intra-EU migration as the labour market complement to monetary union. Workers’ mobility is required in an optimally functioning single currency area. Typically, it is said to absorb local economic shocks to employment which might follow from the collapse of production in any particular local or national context. This argument won Robert Mundell (1961) the Nobel prize for economics in 1999, and stands as the major theory-based justification for the continuing support for free movement of market-oriented pundits and policy makers (see also Belassa 1961; Mattli 1999). Another economic argument has come to the fore more recently: mobility as an ‘autarchic’ response to demand at the higher end of the labour market. Mobility is particularly productive when it equates with brain circulation. While the free movement doctrine was born of the need to facilitate working-class migration from the south in the 1950s, in the last decade or so it has been re-focused on encouraging the knowledge economy through a more efficient use of the educated and highly skilled workforce. The ambition is to promote an increased use of European human capital within Europe as a means to ‘make the EU the world’s most dynamic and competitive economy’, as stated in the EU’s Lisbon Agenda of 2000. This would especially require the expansion of supranational research and development structures in both the public and private sectors, as part of the global struggle for competitiveness (see Sapir et al. 2004). From a political viewpoint, on the other hand, free movement is frequently perceived as an instrument to deepen European integration at the societal level. The Action Plan for Skills and Mobility (European Commission 2002) makes it clear that individuals who have tasted free movement rights are expected to better appreciate European citizenship and endorse European unification more wholeheartedly than the rest of the population. This point is taken up again by the most recent legislation consolidating and expanding free movement rights: Enjoyment of permanent residence by Union citizens who have chosen to settle long term in the host member state would strengthen the feeling of Union citizenship and is a key element in promoting social cohesion, which is one of the fundamental objectives of the Union. (Directive 2004/38)

Pioneers of European integration: an introduction

11

In other words, movers are expected to contribute to overcoming national differences, fulfilling the vision of Euro-enthusiast intellectuals like Václav Havel. It evokes also certain cosmopolitan antecedents in Europe, as Havel’s commentary makes clear: If regulations on the movements of citizens disappear, we will see the sort of blending produced in the Austro-Hungarian Monarchy before the First World War. Subjects came and went, married all over the place, tried their luck at many things, without any of the preceding impeding development of national cultures. (Havel 1998, 119)

In this scenario, the fusion of European societies would be greatly facilitated by burgeoning mobility. Such an emphasis on mobility does not go uncontested. In particular, scholars in the Marxist tradition contend that all efforts to create a ‘Europe of flows’ (Hajer 2000) are functional to the expansion of capitalism – that is, to the rise of exploitative social relations, the disruption of community life, and negative psychological consequences (for these general arguments, see also Deleuze and Guattari 1980; Harvey 1982; Sennett 1998). The European Year of Workers’ Mobility of 2006 thus increasingly sought to focus its efforts on also managing the downsides and social dysfunctions linked to the promotion of mobility. Nonetheless, even critics acknowledge that European integration has become organized around a principle of spatial governance within a frame of ‘seamless mobility’, with mobility the ‘defining feature of contemporary Europe’ (Jensen and Richardson 2004, 3).

MOBILE EUROPEANS IN EU STATISTICS The next set of questions concerns numbers. How many EU movers are there? Who are they? How are they distributed spatially in terms of origins and destinations? Are their numbers growing? Unfortunately, differences in national systems of registration of residents and freedom of movement itself make it hard to count EU movers in a consistent and systematic way (on these measurement issues, see Poulain et al. 2006). Nevertheless, in this section, we describe our population of reference on the basis of the Eurostat Dissemination Database (EDD), as well as data from the EU Labour Force Survey. The bulk of figures in this section refer to population stocks in EU15 between 1987 and 2004.6 Flows have not varied dramatically from the mid-1990s on, but do show a rising trend: from 0.6 million in 1997 to 1.1 million persons in 2003

12

Pioneers of European integration

(excluding inflows into Greece, France and Ireland). Interestingly, 64 per cent of EU15 movers are returning migrants. Germany, Britain and Spain receive the largest numbers of intra-EU movers yearly. About four out of ten of these movers are in the 25–39 age bracket. British, German, French and Italian citizens (in that order) form the largest nationality groups resettling abroad in the EU. Regarding stocks, during the period 1987 to 2004, while the overall number of foreign residents grew from 14.4 to 21.4 million (148.7 per cent), the stock of EU movers rose at a slower pace – from about 5.2 to 6.3 million people (121.1 per cent). The number of EU15 non-national residents increased in all member states – particularly in Britain (1443 000), Germany (1207 000) and Spain (1181 000) – with the exception of Sweden, France and Italy. In fact, the relative size of EU non-national citizens among foreigners varies considerably from country to country. Intra-EU movers are a majority among foreigners in Luxembourg, Belgium and Ireland. In all other EU member states, there are more nonEuropean immigrants than foreign European residents. There are often three to four times more immigrants than EU movers, and Italy has the lowest proportion of EU movers among its alien residents: only 9.9 per cent. Since the mid-1980s, Germany, France and Britain have continued to be the countries hosting the largest numbers of EU citizens living abroad. On the other hand, Italians and Portuguese continue to be the largest communities of Europeans residing in another EU member state. However, the size of the Italian and Portuguese communities has oscillated only slightly over these two decades. Interestingly, the fastest growing communities of Europeans abroad are not formed by southern Europeans – as they were traditionally in the 1950s–1970s – but rather by central and northern Europeans. Between 1987 and 2002, the number of Germans living outside their native country in the EU nearly doubled (196 per cent), while the growth of expatriate French (159 per cent), Belgians (151 per cent), British (142 per cent) and Dutch (132 per cent) is also remarkable. Only the Spanish and Finnish communities abroad have shrunk in this period, albeit modestly. In 1987, the single largest community of EU non-nationals in a EU member state was formed by the Portuguese in France (649 000 persons), where they made up almost half the EU mover population. Their number declined, however, to 555 000 persons in 2002. Therefore, in the early 2000s the single largest community of EU movers is formed by Italians in Germany: 644 000 persons, representing 36.6 per cent of EU non-national residents in that country. Their number has increased modestly but constantly from the mid-1980s onwards. This could be due to both continuing

Pioneers of European integration: an introduction

13

net migration flows from Italy to Germany and the expansion of secondgeneration Italian migrants living in Germany. These same reasons can equally account for the growth in the number of Portuguese citizens living in Luxembourg, which doubled between 1987 and 2002. The importance of geographical and cultural proximity for intra-EU mobility is attested by the high proportion of British citizens in Ireland and of the Irish in Britain, of Finns in Sweden and Swedes in Finland, and Germans in Austria. The number of Austrians in Germany, even though representing 77 per cent of Austrians abroad in the EU, is overshadowed by the larger presence of southern European migrants in Germany. Overall, including citizens of the accession countries already residing in another EU member state, European citizens living abroad in the EU amount to (only) 1.6 per cent of the EU25 population in 2004 (Table 1.1). Although data on gender are incomplete, apparently there is a slight prevalence of men, with the notable exception of four EU15 member states where women exceed men among EU non-national residents: Italy (65.6 per cent), Greece (59.9 per cent), Ireland (51.6 per cent) and Portugal (51.5 per cent). In 2002, the 25–34 age group was the biggest in six out of the 12 member states for which this information is available. People aged 25 to 44 formed from one-third to half the total of EU nonnationals in each country (ranging from 34 per cent in France to 48 per cent in Italy). The educational level of intra-EU movers varies considerably from country to country. The proportion of EU15 non-national residents with a post-secondary degree ranges from 11.8 per cent in Spain to 61 per cent in Lithuania (where absolute numbers are very low, though). Generally speaking, traditional receiving countries of intra-European migration, like Germany, Sweden and France, host a less-educated workforce (EU movers with a tertiary-level degree are 14.4 per cent, 22.4 per cent and 25.2 per cent respectively). In this regard, the difference between Germany and Britain is striking, as in the latter country one-third of EU non-nationals have tertiary-level credentials. These individuals are probably taking jobs in the high-skilled sectors, whereas Germany tends to offer employment opportunities to intra-EU movers in mainly low-end occupations. Gender differences also deserve some attention. In Germany, Italy, Spain and Portugal, as well as in all Eastern European countries, among EU15 nonnationals it is men who have higher educational qualifications. The reverse is true in France and Sweden. As is well known, education is expanding all over the world. But, as migrants represent a select population, this general trend is not necessarily reflected in their educational level. In particular, when labour markets demand low-skilled workers, immigrants tend to be negatively selected

14

8 032 926 10 239 085 58 395 594 689 565 10 203 269 5 397 640 1 370 052 5 219 732 58 513 700 82 440 309 10 934 097 10 116 742 3 978 880 56 995 744 2 319 203 3 445 857 451 600 16 258 032

Total population

753 528 853 369 2 760 031 – 179 154 271 211 – 107 003 3 263 186 7 334 765 – 130 109 215 473 1 334 889 33 344 – 174 200 702 185

Total foreigners* 9.4 8.3 4.7 – 1.8 5.0 – 2.0 5.6 8.9 – 1.3 5.4 2.3 1.4 – 38.6 4.3

% of total population 106 173 562 975 1 268 616 32 214 9 832 54 310 1 410 16 656 1 194 135 1 767 302 46 897 10 718 133 436 132 067 703 534 139 691 201 557

EU15 movers 14.1 66.0 46.0 – 5.5 20.0 – 15.6 36.6 24.1 – 8.2 61.9 9.9 2.1 – 80.2 28.7

% of total foreigners

Stocks of foreign population in EU25 member states (2004)

Austria Belgium Britain Cyprus Czech Rep. Denmark Estonia Finland France Germany Greece Hungary Ireland Italy Latvia Lithuania Luxembourg Netherlands

Table 1.1

57 031 11 476 241 101 479 38 409 8 811 2 603 12 836 41 991 250 563 32 110 6 690 8 455 40 948 2 060 1 223 1 096 10 206

EU10 movers 7.6 1.3 8.7 – 21.4 3.2 – 12.0 1.3 3.4 – 5.1 3.9 3.1 6.2 – 0.6 1.5

% of total foreigners 590 324 278 918 1 250 314 – 130 913 208 090 – 77 511 2 027 060 5 316 900 – 112 701 73 582 1 161 874 30 581 – 33 413 490 422

TCN

78.3 32.7 45.3 – 73.1 76.7 – 72.4 62.1 72.5 – 86.6 34.1 87.0 91.7 – 19.2 69.8

% of total foreigners

15

38 230 080 10 356 117 5 379 455 1 996 433 42 197 865 8 975 670 452 137 647

700 329 238 746 29 855 45 294 2 772 200 476 076 22 492 417

1.8 2.3 0.6 2.3 6.6 5.3 5.0

12 350 49 343 1 282 1 218 360 192 180 191 6 283 802

1.8 20.7 4.3 2.7 13.0 37.8 27.9

2 041 855 10 491 201 25 240 24 039 830 955

0.3 0.4 35.1 0.4 0.9 5.0 3.7

685 938 188 548 18 082 43 875 2 386 768 271 846 15 377 660

97.9 79.0 60.6 96.9 86.1 57.1 68.4

Source:

Eurostat Dissemination Database (domain: Population and Social Conditions)

Notes: 1. *Total foreigners: Czech Republic, Ireland, Portugal, Slovakia, Britain (2003); Poland, Italy (2002); Belgium, Austria (2000); France (1999). TCN5 Third-country nationals. 2. Data for Malta is missing.

Poland Portugal Slovakia Slovenia Spain Sweden Total

16

Pioneers of European integration

on education (Borjas 1989). Therefore, a diachronic analysis of intra-EU movers’ education is informative of the kind of occupational incorporation to which this particular fraction of the migrant population is channelled. Table 1.2 shows that the education of EU non-nationals has improved constantly and markedly from 1995 to 2005. If in the mid-1990s only 14.3 per cent of intra-EU movers had a tertiary degree, by 2005 this is the case for almost a quarter of them. But the most significant finding is the following: while in 1995 the proportion of national residents with a tertiary-level education was higher than that of EU movers, the situation was reversed in 2005. In other words, the upgrading of the educational level of EU movers has exceeded that of the general population. EU movers are now a positively selected population in terms of education. The education differentials between EU non-nationals and nationals are at their peak in Denmark, Ireland, Luxembourg, Austria, Spain and Portugal. In all these countries, the share of tertiary-educated EU movers is about twice as large as that of nationals. There can be few doubts that geographical mobility within the EU spreads human capital across the labour market, although the size of the population at stake varies considerably from one country to another. Over time, intra-EU movers are also redefining their position in the workforce (Table 1.3). In terms of sectors of employment, EU citizens living abroad are less likely than nationals and also third-country nationals to have a job in agriculture. More significantly, their involvement in the industrial sector is declining at a faster pace than that of nationals. Indeed, in 1995 the proportion of industrial workers among EU movers was still well above that of nationals (55.8 per cent versus 50.9 per cent); after 2000, the two are about the same. Finally, we can distinguish two separate categories in the broad service sector: a low-level sector (mainly including personal service activities) and a mid-high level sector (in which non-manual activities are the rule). In the low-level service sector, EU15 non-nationals are more frequently employed than nationals, but less than third-country nationals. The growth of this sector of employment from the mid-1990s onwards has been almost entirely fed by immigrants from outside the EU. As far as mid-high service jobs are concerned, EU nonnationals are following the path of nationals. Although they still lag behind national workers in relative terms, they are employed in increasingly larger numbers in this more privileged occupational sector.

A GUIDE TO THE VOLUME We can now turn to the organisation of the book itself. The volume draws together a number of different analyses by members of the PIONEUR

17

National EU25 Non-nat. Total National EU25 Non-nat. Total National EU25 Non-nat. Total National EU25 Non-nat. Total National EU25 Non-nat. Total National EU25 Non-nat. Total

AT

ES

EL

DK

DE

BE

Nationality

57.5 55.5 55.9 29.9 29.6 57.1 40.6 55.6 45.2 44.9 26.8 53.6 27.0 15.5 22.6 15.5

52.6 24.6 45.3 26.6 35.8

35.9 62.9 17.6 62.6 71.8 47.1 71.6

Med.

36.6 19.7 37.9 51.9

Low

1995

19.1 10.3 28.8 10.4 12.7 30.3 12.9

17.8 18.3 14.1 17.8 18.0

5.9 24.8 6.1 18.2

High

31.4 16.9 31.3 56.2 20.7 55.9 65.9 39.2 65.5

32.0 18.3 33.1 49.2 56.3 50.0

Low

20.0 31.8 20.1 11.9 25.8 12.0 17.6 32.8 17.8

11.4 20.8 17.3 20.5

55.5 30.0 26.4 29.5

48.6 51.3 48.6 31.9 53.5 32.1 16.5 28.0 16.7

11.4

High

56.6

Med.

2000

27.9 14.0 29.0 44.1 49.8 44.8 25.6 40.2 27.7 29.7 20.0 29.9 50.0 18.0 49.9 60.2 29.5 59.0

Low

58.0 59.3 56.9 32.6 26.3 31.9 54.7 42.3 53.0 45.2 29.0 44.6 34.8 61.4 35.1 18.0 29.0 19.0

Med.

2005

14.1 26.7 14.1 23.3 23.9 23.3 19.7 17.5 19.3 25.1 51.0 25.5 15.2 20.6 15.0 21.8 41.5 22.0

High

Nationals and EU non-nationals by educational level, 1995–2005 (row %, individuals aged 15 and over)

COR

Table 1.2

18

National EU25 Non-nat. Total National EU25 Non-nat. Total National EU25 Non-nat. Total National EU25 Non-nat. Total National EU25 Non-nat. Total National EU25 Non-nat. Total National EU25 Non-nat. Total

FI

PT

NL

LU

IE

GB*

FR

Nationality

(continued)

COR

Table 1.2

11.9 12.1

80.1

79.8

29.2 19.8 26.2

39.3 37.0 19.6 36.0 32.3 19.6 31.7

44.6 48.1 70.5 49.4 48.2 64.1 48.9

59.3 65.4 61.3

39.3

Med.

44.5

Low

1995

8.1

8.0

11.5 14.8 12.5

16.1 14.9 9.9 14.6 19.5 16.3 19.4

16.2

High

39.0 50.1 43.3 41.0 36.7 41.6 82.4 60.6 82.2

32.5 37.0 32.6 46.3 67.7 47.6

Low

48.0 31.7 41.9 39.7 35.3 39.2 11.1 19.3 11.2

40.5 42.4 40.4 36.1 18.8 35.1

Med.

2000

13.0 18.2 14.8 19.3 28.0 19.2 6.5 20.1 6.6

27.0 20.6 27.0 17.6 13.5 17.3

High 34.5 38.0 34.7 42.6 58.3 43.6 14.8 16.7 15.1 43.7 24.3 42.4 33.0 39.2 35.1 37.1 17.7 37.2 77.5 46.5 77.1

Low

39.8 42.1 39.8 37.5 22.8 36.6 58.7 58.2 58.7 34.3 37.3 34.4 50.5 32.9 44.1 38.3 50.1 38.4 13.3 29.0 13.5

Med.

2005

25.7 19.9 25.5 19.9 18.9 19.8 26.5 25.1 26.3 22.0 38.4 23.2 16.5 27.9 20.8 24.6 32.2 24.4 9.2 24.5 9.4

High

19

National EU25 Non-nat. Total National EU25 Non-nat. Total

43.0 42.9 37.8 30.0 37.4

35.1

35.1 46.2 55.7 46.8

22.0 16.0 14.3 15.8

21.9

29.2 32.2 29.4 39.7 47.2 40.4

45.5 40.0 45.1 41.2 32.2 40.6

25.3 27.8 25.5 19.1 20.6 19.0

23.7 22.3 23.9 36.6 37.1 37.2

Source:

Labour Force Survey, spring data

Notes: 1. Educational levels defined according to ISCED classification. * Figures with low reliability. Row totals for each year 100%. 2. Data for Italy is missing. 3. Low5levels 0–2; Medium5levels 3–4; High5levels 5–6 of ISCED. On ISCED, see http://www.uis.unesco.org/TEMPLATE/pdf/isced/ISCED_A.pdf

EU15

SE

51.7 45.7 51.1 42.4 39.3 41.9

24.6 32.0 25.0 21.0 23.6 20.9

20

Pioneers of European integration

Table 1.3

1995

2000

2005

Employed nationals, EU15 non-nationals and third-country nationals by sector of employment in EU15, 1995–2005 (row %, individuals aged 15 and over)

Nationals EU non-nationals TCN Total Nationals EU non-nationals TCN Total Nationals EU non-nationals TCN Total

Agriculture

Industry

Service (low level)

Service (mid-high level)

Total

5.04 1.61 1.73 4.87 4.31 1.43 2.45 4.19 3.71 1.85 3.35 3.65

50.90 55.83 60.39 51.29 49.17 49.03 54.40 49.33 46.65 46.62 48.48 46.73

9.16 17.11 14.88 9.49 9.51 18.38 18.00 9.97 9.83 18.01 23.53 10.62

34.90 25.44 23.00 34.36 37.01 31.16 25.15 36.51 39.82 33.51 24.65 39.01

100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00

Notes: NACE activities are grouped as follows: Agriculture 5 A. Agriculture, hunting and forestry; B. Fishing. Industry 5 C. Mining and quarrying; D. Manufacturing; E. Electricity, gas and water supply; F. Construction; G. Wholesale and retail trade; repair of motor vehicles, motorcycles and personal and household goods; I. Transport, storage and communication. Low level services 5 H. Hotels and restaurants; O. Other community, social and personal service activities; P. Private households with employed persons. Mid-high level services 5 J. Financial intermediation; K. Real estate, renting and business activities; L. Public administration and defence; compulsory social security; M. Education; N. Health and social work; Q. Extra-territorial organizations and bodies. TCN5Third country nationals. Source:

Labour Force Survey, spring data

group, each using the basic EIMSS dataset. Some pose questions in relation to existing European survey sources, such as Eurobarometer (EB) or the European Social Survey (ESS), others derive questions in relation to established studies of migration or immigration. In total, the chapters add up to a comprehensive and unique portrait of the social effects of the EU’s extraordinary free movement regime. Chapter 2 by Michael Braun and Camelia Arsene offers an overview of the mobile population of EU movers surveyed by the EIMSS. Relatively little is known about the population of intra-EU movers, aside from crude figures on stocks and flows in different European countries. First, it details

Pioneers of European integration: an introduction

21

the social background and age of migrants, the duration of migration, geographic origins, and the cross-national marriage patterns of movers. Then, breaking down the population in terms of periods of migration and age at migration, it derives a preliminary typology of EU movers. Four clusters emerge: ‘late traditional migrants’, continuing the classical labour migration from south to north; ‘pure retirement migrants’, who are moving from north to south to settle after ending work; ‘pre-retirement migrants’, who are middle aged, affluent movers who move as part of a lifestyle decision; and finally ‘Eurostars’, a new highly mobile class of younger professionals and others using European mobility opportunities. Classic migration theory focuses on the motivation, timing and patterns of migration, and on the social organisation of migration systems. Intra-EU migration can be viewed in these terms, but it also suggests the emergence of new forms of mobility within the evolving transnational space of the European Union. In Chapter 3, Oscar Santacreu, Emiliana Baldoni and María-Carmen Albert analyse the complex motivations of intra-EU movers, and the way such migration often builds on successive experiences of mobility within and outside Europe. They focus on the subjective motivations, personal networks and family support that structure these movements, both as separate factors and in combination. It turns out that intra-EU migration is not just driven by economic factors. In fact, the lure of cross-national romance, leisure and lifestyle opportunities (that is, ‘quality of life’), and adventure against the new European background outweigh purely economic motivations. Significant gender differences also emerge, especially within the category of ‘romance migration’, as cross-national partnerships take on specific cross-cultural forms. Sexuality also features in European migration, given higher than average mobility among gays and lesbians. Together with the previous chapter, the analysis begins to draw a new map of migration within Western Europe. In Chapter 4, Ettore Recchi then tackles the crucial issue of social mobility as it might be related to spatial mobility in the EU. Traditional migrants typically move spatially in order to be mobile socially, particularly across generations. Opportunities for work, education, career and the bettering of lifestyles have, in highly fluid societies, led spatial mobility to be a kind of escalator up the social ladder, often associated with moves from periphery regions to core cities. This has always held for internal migration within European nations, but European integration poses the question of whether spatial mobility within and around the new European space may offer new possibilities to ambitious social climbers on an international scale. EIMSS permits a systematic comparison of social mobility patterns of movers with stayers, using ESS as a reference. Recchi offers a detailed analysis that checks for a number of potential hypotheses of

22

Pioneers of European integration

inter-generational and intra-generational social mobility. Contrary to some of the hopes associated with European integration, he finds that there is not a substantial structural mobility effect. EU movement opportunities tend to be more used by middle- and upper-class Europeans, and the bulk of movers stay fixed in the same occupational class they belonged to before moving abroad. A large part of traditional immigration studies focuses on processes of acculturation, assimilation or integration (depending on the model) of immigrants into national host societies, according to a series of indexes regarding language use, cultural adaptation, family life and social practices. EIMSS was set up to enable the evaluation of acculturation processes of foreign resident Europeans in their destination countries, and does so for different countries and different nationalities, across cohorts, gender and social classes. In Chapter 5, Antonio Alaminos and Oscar Santacreu highlight the discrepancies between the social and cultural integration of EU movers, showing that these distinctly transnational migrants in Europe differ substantially from typical immigrants in how they combine the pull of their home culture with the challenges of the new country of residency. Lack of language proficiency proves to be no great barrier to effective integration of migrants in the host society, even though it limits networks that take them out of their own national milieus. Chapter 6 by Nina Rother and Tina Nebe pushes this line of inquiry deep into the question of European and national identity. Much of the European Commission’s justificatory commentary on freedom of movement stresses the positive impact of mobility on the emergence of European identity. Movement is meant to establish a more solid foundation for the EU, through positive, affective identification with the European project. Referring to the large body of data and existing research based on regular Eurobarometer surveys measuring support for European integration, territorial attachment and the hierarchisation of national, local and transnational identity, EIMSS repeats these questions for the EU movers. These individuals might indeed be considered the prototypical ‘Highly Europeanized Citizens’ (HECs), benefiting the most from the European integration project. Nebe and Rother create a typology of movers on the basis of distinct patterns of identification with country of origin, country of residence and the EU. Elaborating the social psychology literature, they find that one potential for European identity is that it reduces the ‘cognitive dissonance’ felt by movers caught between strong national identifications. A European identity in this way is not incompatible with existing national identities. In Chapter 7, by Anne Muxel, the big political questions about European citizenship are asked. Much of the literature on European

Pioneers of European integration: an introduction

23

citizenship has focused on the opportunities European nationals have to participate as EU citizens abroad in local and European elections. Our survey took place at the time of the 2004 European elections, and so was well placed to gauge this participation amongst a representative sample of such citizens. As expected, movers do participate more than stayers, but rates are not impressive. It is necessary, therefore, to go beyond voting and compare the importance of this to other forms of political involvement. The chapter also analyses data for the first time available on the left/right political orientation of EU movers and their attitudes on liberalisation and state intervention. These prototypical beneficiaries of a more liberal, open European market, paradoxically tend at the same time to be more supportive of state intervention. A closer look shows that the social class background and education of EU movers matter more to participation than national differences. A further essential area of inquiry on the question of an emergent European transnational space is, of course, the media. A large body of literature on migration and media studies has emerged in recent years discussing the possibilities of transnational media in the context of European integration and beyond, and of new diaspora media and new channels of communication associated with innovative technologies. Europe however is also characterized by a very weak public sphere beyond nation-state contexts, given the modest success of attempts to create print or visual media for a European market. Chapter 8 by Damian Tambini and Nina Rother shows that EU movers are consumers of transnational and European media where available, and they show a strikingly low use of television media compared to print media and the internet. However, a more accurate classification would see their media consumption in terms of combinatory approaches that also often involve appreciation and use of local (host country) media alongside international sources. They also underline that emerging transnational patterns are linked to the availability and quality of media in different countries, findings that are relevant to attempts to construct a more effective European media space. Finally, in Chapter 9, Adrian Favell and Tina Nebe push the questioning of PIONEUR out into the effects of EU enlargement. As well as offering a systematic survey of West European intra-EU migration, PIONEUR took up the challenge of comparing findings on intra-EU migration with what might be expected from new member states which joined the EU in May 2004 and which were about to join in 2007. A series of 40 qualitative in-depth interviews were thus conducted with ‘higher’ and ‘lower’ end labour migrants from both Poland and Romania (the two largest accession countries) in the five selected Western member states. Similar questions to EIMSS were posed, bearing in mind that the features of intra-EU

24

Pioneers of European integration

migration – borderless mobility, no technical barriers, minimal prejudice – are certainly not a given for movers from any East-Central European country. The life stories of these new pioneers in fact speak more to a traditional immigrant experience. They face discrimination and barriers not apparent to Western EU movers, while also seeing Western Europe as a place to advance career aspirations or opportunities for their children. The chapter points towards new research on the future of the regional migration system in Europe, as East–West migration comes to more resemble the intra-EU migration within Western Europe surveyed by EIMSS.

CONCLUDING REMARKS The study of European integration has been slow to incorporate sociological analyses. Pioneers of European Integration represents a key empirical contribution to an emerging sociology and human geography of the European Union (for an overview of existing contributions see Bettin Lattes and Recchi 2005; Favell and Guiraudon 2010). Migration and mobility have long been recognized as important dimensions of the European project, but systematic empirical studies have been scarce. Some studies have tackled dimensions of this new mobility (King et al. 2000; Favell 2008a), and there has been much reflection on the significance of European citizenship (Magnette 1999; Bellamy et al. 2006) and European identity (Hermann et al. 2004; Checkel and Katzenstein 2009; Fligstein 2008). It is the first time, however, that systematic, representative data on spatial mobility and its social effects within and across Western Europe have been presented and analysed in this form. Europe is unfinished but the freedom of movement of persons has wrought dramatic social change. Much of this is not yet fully visible. Some of the most dramatic examples can be found in the lives and experiences of those who took up most enthusiastically their new rights of European citizenship, and decided to give it a go living and working in another member state. Successful movement and resettlement is far from a given. But, as we show in this book, Europe will never be the same again: neither for the EU movers, nor for those stayers now getting used to their foreign neighbours as colleagues, friends and partners. The pioneers of European integration are also, perhaps, the pioneers of a better, brighter Europe for all.

Pioneers of European integration: an introduction

25

NOTES 1. The claim is based on data from Eurobarometer 64.1 of 2005. The question is: ‘Which of the following statements best describe(s) what the European Union means to you personally?’ It was also asked in 1997, 1998 and 2001. 2. Figures for mobile students and researchers are taken from European Commission reports, available via: http://ec.europa.eu/education/index_en.htm. 3. See the PIONEUR website: http://www.obets.ua.es/pioneur/. 4. In most original random large-scale surveys – such as the Labour Force Surveys – the number of migrants surveyed is usually a very small percentage of the overall population. EIMSS differed by devising a names-based telephone methodology for identifying migrants – who are a very small part of the overall population – in sufficient numbers. Our survey is comparable to ones made in Israel and Australia, among the few countries where different immigrant groups can be identified in sufficient numbers to survey comparatively (see Appendix A for more details). 5. An overview of the issue and the changing national measures can be found on the EC website: http://ec.europa.eu/social/main.jsp?catid5466&langid5en?/free_movement/enlargement_en.htm. 6. Data and tables presented in this section are taken from Recchi et al. (2006), which also contains more fine-tuned analyses.

2.

The demographics of movers and stayers in the European Union Michael Braun and Camelia Arsene

INTRODUCTION Who are EU movers? Do they have some common demographic and background features? Are they different from the general population of the country they leave? And how do they compare to the population of the country they have settled in? These basic questions are addressed in this chapter. In it we describe the sociodemographic characteristics of the EU movers sampled by the PIONEUR project in 2004. First, we offer a brief summary of the methodological strengths and weaknesses of the quantitative EIMSS survey, outlining some challenges we faced in our analysis. Then, starting with an overview of the social and geographical background of movers, the development of migration over time, and the age at migration, we use the last two variables to arrive at a classification of migrants into four groups according to their duration of sojourn in the country of residence and their age at migration. We then go on to portray the four migrant groups with regard to a variety of other variables.1 The EIMSS survey defined its population of interest as all migrants from Britain, France, Germany, Italy and Spain – the five largest EU15 member states – who had moved to one or other of these same countries from 1974 through 2003, were adults at the time of migration, and have been living in their respective country of residence in 2004 for at least one year. From each country of origin/country of residence combination we sampled some 250 individuals, for an overall survey size of 5000 cases. A survey of this kind has never been done in Europe before. In fact, it stands as a unique contribution to migration research anywhere in the world, which is often fatally undercut by the fact that migrant populations are normally too small in general population surveys to be identified and analysed as a distinct population. Moreover, once a means for identifying the EU mover population has been set up, it is then possible to systematically compare it to what we know from other surveys about the average European population – the non-moving ‘EU stayers’ in our terminology. 26

The demographics of movers and stayers

27

EIMSS is thus set up to plug in directly to existing European Social Survey and Eurobarometer data. Most of the data gathering was done by traditional telephone survey techniques. Much has been written in recent years about the limitations of this approach, yet in most locations it proved a viable way of locating respondents. Where the survey represented a bold innovation was in the development of a new names-based telephone directory sampling technique, which identified foreign populations in each country through probability calculation on first and family names. This worked remarkably well. One typical problem in telephone research as set up – the underrepresentation of women married to nationals of the country of residence – was rectified by the use of a small network component in our survey. In one case – Britain – where mobile phone use had perhaps already rendered landline telephone sampling unreliable by 2004, we also supplemented the telephone survey with snowball sampling of respondents (see Appendix A for more details). A number of caveats and limitations about our survey, however, must also be pointed out. One limitation is that we cannot analyse return migration with a cross-sectional sample, nor account for the potential selectivity in our sample due to early remigration of migrants who might have special characteristics. However, an act of return migration or an intention to return to one’s home country should not be assumed to be negative selfselection. In the end, this really depends on the original intentions of the migrant. If migration was intended to be temporary to begin with, return migration might even demonstrate success. It should also be noted that, with a cross-sectional design, the duration of the sojourn in the country of residence and the period of immigration cannot vary independently, as all migrants with a given duration of stay have by design necessarily arrived in the same year. Furthermore, as with most migrant surveys, when considering how results can be interpreted in the different domains covered, the differing self-selectivity of the national groups surveyed has to be taken into account. A key question is whether and to what extent the various groups sampled – that is, the country of origin (COO) and country of residence (COR) combinations – are comparable. Individuals do not move at random from the different countries of origin to the different countries of residence. Rather there is substantial self-selection by different personal characteristics. For instance, if we want to find out whether Italians in France fare better than in other countries of residence, we have to take into account whether Italian migrants vary according to the different destinations that they select to begin with. Initial characteristics might differ according to important variables such as age, social background

28

Pioneers of European integration

and educational qualifications, depending on whether an Italian migrant has moved to France, Germany or Britain. These must all be allowed for. In addition to these differences which are measured by our survey questions, there might also be unobserved heterogeneity which was not measured or is in principle not measurable. Countries of residence differ with regard to which people they attract the most: labour, study or retirement migrants. Traditional labour migrants are one group we might find in different places, but not all. They are unlikely to be equally represented across destinations. A second obvious group might be retirement migrants. Again, it is not hard to imagine that these will have special features according to where they are coming from and where they are going – mainly of a geographical kind. Another self-selected group are those ‘free movers’ (Favell 2008a) that we identify as an emergent part of EU internal migration. Again, different variables will come into play for this group. In contrast to traditional migration by unskilled labour, free movers might be assumed to migrate because of individual choices (that is, the importance of chain migration and formal recruitment arrangements is reduced), and differences regarding the economic level between countries of origin and countries of residence should be of less importance. Methodologically, the problem is that if both the traditional migrant and the free-mover types are not equally represented in the different groups across countries, then this has to be taken into account when considering the evidence found by the various chapters that follow. For example, although our data at first might suggest this, it might not be possible to conclude that, say, Britain is more open for migrant integration than, say Germans, simply on the basis of a straightforward comparison of Italians in Britain with Italians in Germany. This is because the Italians in each country might, for the self-selective processes mentioned above, differ substantially in terms of their social characteristics to begin with. Italians might be mostly ‘free movers’ in Britain, and ‘traditional guest workers’ in Germany: this will impact all their other measurable behaviour. Throughout the volume then we carefully allow for the self-selective nature of the groups assessed to avoid coming to such naïve conclusions, and rather put the accent on looking to see how and why Italian migrants (for example) differ in different locations. These considerations apply in principle to all international comparative migration research, as long as fixed-number samples of migrants from similar countries of origin or countries of residence are selected for investigation. However, given the basic diversity of the movers included in the EIMSS survey, we are especially alert to this issue. Besides illustrating the basic sociodemographic traits of respondents to our survey, this chapter

The demographics of movers and stayers

29

will thus give an evidence-based idea of the way self-selectivity has worked out in the sample and why EU movers are so diverse. Moreover, since we sampled an equal number of migrants from each country of origin/country of residence combination, our results have to allow for the fact that there are quite different totals of migrants by nationality in each of these combinations in reality. That is, our sample of Italians in Germany is a much smaller relative proportion of the total of Italians in Germany than, say, our sample of Italians in Spain. This too demands special care in the analysis: the size of the overall migrant population by nationality may have important effects on the characteristics and dynamics of that population. For this reason, our next step should be to indicate the official figures of migration for our country of origin/country of residence combination, in order to give a sense of the proportion of the migration we studied relative to the overall picture of present-day migration in Europe.

INTRA-EU MIGRANTS IN THE PIONEUR COUNTRIES At the time of the EIMSS survey in 2004, Germany had the highest number of (legal) resident foreigners with 7.3 million, followed by France (3.3 million), Britain (2.9 million), Spain (2.8 million) and Italy (1.3 million). Since the early 1990s, there had been sizeable differences across the five countries in the development of the numbers of foreign nationals both from third countries (that is, non-EU members) and those from other EU15 countries (Table 1). With regard to all foreigners irrespective Table 2.1

Stocks of all foreigners and EU15 non-nationals, 1990 and 2004 (in thousands) 1990

Germany France Britain Italy Spain

2004

Foreigners

EU

% EU

Foreigners

EU

% EU

4 845.9 3 596.6 2 416.0 490.4 398.1

1 516.8 1 321.5 910.0 140.8 245.8

31.3 36.7 37.7 28.7 61.7

7 334.8 3 263.2 2 760.0 1 334.9 2 772.2

1 767.3 1 195.5 1 268.6 132.1 360.2

24.1 36.6 46.0 9.9 13.0

Source: Eurostat New Cronos Database and Eurostat Dissemination Database (see Recchi et al. 2006); data for France from 1999, for Britain from 2003, for Italy from 2002

30

Pioneers of European integration

Table 2.2

Germany France Britain Italy Spain Source:

Stocks of movers from EIMSS countries, 2002 Germans

French

British

Italians

Spanish

– 76 882 266 136 35 091 78 020

86 696 – 96 281 29 313 46 894

99 587 75 546 – 19 957 94 862

644 955 200 632 107 244 – 36 815

115 818 160 194 54 482 12 327 –

Eurostat Dissemination Database (see Recchi et al. 2006)

of geographic origin, for France, in the 1990s, there was a slight decline, probably linked to the relatively high number of naturalizations during this period, whereas in Britain the numbers rose slightly (by 14 per cent) and in Germany substantially (by 51 per cent). The numbers in Italy and Spain, however, grew dramatically during the same period. In Italy, the number of foreign nationals nearly tripled, and for Spain it increased nearly sevenfold from 1990 (see also Münz and Fassmann 2004; Salt 1997 and 2005; and the annual SOPEMI reports). EU15 foreigners are a minority among foreigners in all five countries in 2004, ranging from 1.8 million in Germany to 132 000 in Italy. Moreover, the number of EU15 nonnationals since the early 1990s declined in France and Italy and increased modestly in Germany (16.5 per cent) with larger increases in Britain (39.4 per cent) and Spain (46.5 per cent). When we look at country of origin/country of residence combinations in 2002, the single largest group are Italians in Germany with more than 600 000 people, followed by Germans in Britain with more than 260 000, and by Italians and Spanish in France with some 200 000 and 160 000 respectively (Table 2.2). The smallest groups are found in Italy, where no national group of the four surpasses around 35 000 (Germans), and the Spanish immigrants constitute the smallest with some 12 000. In 2004, no national group in Germany had less than 100 000 persons (Eurostat 2006). These stocks reflect both the legacy of traditional guest-worker migration (movement exclusively from South to North) as well as, in the case of growing numbers in Spain, more recent trends (movement increasingly also from North to South). In line with this last observation, the sharpest increases can be found for Germans and British in Spain and France, while at the same time the number of Spanish in Germany and France and of Italians in France has decreased (Recchi et al. 2003). While the two decades before 1974 were characterized by guest-worker migration from the South to the North, there was hardly any migration from the North to the South

The demographics of movers and stayers

31

in those years. From 1974 on, the year when guest-worker recruitment stopped, family reunification became the main type of migration (Bauer and Zimmermann 1996; Zimmermann 1995). In sharp contrast to the other traditional guest-worker populations, Italians were not affected by the end of active recruitment of foreign labour: Italy was a member of the European Community, and Italians could continue to move. However, labour migration from Italy to the North was largely discouraged due to the paucity of available jobs. On the other hand, migration to Britain has continued to increase for all our Western European nationalities from the 1970s onwards. These figures are compatible with a general decline, observable across Europe, of low-skilled labour migration and family reunification, and increases in highly-skilled labour and retirement migration (King 2002). Britain, in fact, emerges as the target of highly-skilled labour migration among our five countries, a result in line with past research (Salt 1992), while Spain becomes the target of retirement migration.

BASIC SOCIODEMOGRAPHIC CHARACTERISTICS OF EU MOVERS In this section we illustrate the social background, education and employment status of movers, their age and duration of migration, geographical aspects of mobility, marriage status and the national origin of partners. Then we will focus on the migration process over time, and in more detail on age at migration and its change over the different time periods. Having identified these two key variables, we will then use multilevel analysis to show how age at the time of migration varies in the different periods considered and how this variable is linked to other individual characteristics of the EU movers. We will also introduce one higher-level variable which seeks to capture the essence of the country of origin/country of residence combinations. Finally, we use age at migration and duration of stay to classify groups of migrants, which we will then describe using additional variables. The goal is to move towards identifying specific clusters or types of migrants within the overall group of movers. When appropriate, we will also compare movers to the corresponding stayer populations, as is done in the following chapters of this book. For this, we use principally the first wave of the European Social Survey (ESS) as a reference survey. Social Background A first finding is that, when measured by parental education, migrants from Southern Europe stem from a lower social background than migrants from

32

Pioneers of European integration

Table 2.3

Country of residence France Germany Britain Italy Spain Notes: Source:

Respondents’ education (lower-secondary qualification or less) by country of origin and country of residence (in %) Country of origin French

Germans

British

Italians

Spanish

56.9 18.2 10.6 12.5 23.8

38.2 70.0 15.6 33.2 60.8

18.5 31.5 56.5 20.1 43.4

33.2 72.6 23.9 56.1 39.4

28.4 50.8 25.6 26.0 66.7

Stayer population is indicated in the main diagonal (shaded). ESS 2004 and EIMSS, N5 13 839

the other countries. This is particularly the case for migration to Germany. However, it is also the case for migration to Spain from nearly all countries (with the partial exception of the French); this is partly but not entirely due to the higher age of the migrants. With very few exceptions, movers come from a higher social background than stayers, of whom more than 75 per cent have fathers with a lower level of education. The exceptions are Italians and Spanish in Germany and Germans in Spain; these groups have a lower social background than the respective stayer populations. For Germans in Spain this is due to their higher age; however, this is not the case for Italians and Spanish in Germany, who have come at lower ages. The majority of EU movers are highly educated. However, in Britain and especially in Germany, migrants from Spain and Italy have a lower educational background than migrants from the other countries (Table 2.3). This also applies to German and British migrants to Spain, when compared to Germans and British in the other countries. Movers differ also on this dimension from stayers, 55–70 per cent of whom have a lower secondary education or less. In fact, almost all groups of movers have a higher education than the corresponding stayer populations. This even applies for the British and Germans in Spain, though the differences from the stayer populations are not as pronounced as for other mover groups. There is only one clear exception: Italian movers settled in Germany have a noticeably lower education than the Italian stayer population overall. Regarding their present occupation, British and German nationals living in Spain and Italy are more likely to be self-employed – if they are still working at all. For respondents working full- or part-time, the percentages are similar for most of the migrant groups, with the exception

33

1920 1930 1940 1950 1960 1970 1980

Year of birth

The demographics of movers and stayers

D

GB I E

France

F GB I E

Germany

D F

I E

Britain

D F GB

Italy

E

D F GB I

Spain

Note: Countries of origin (first row below graphic): D 5 Germans, F 5 French, GB 5 British, I 5 Italians, E 5 Spanish. Source:

EIMSS, N54901

Figure 2.1

Distribution of year of birth dependent on country of origin and country of residence

of Spain as country of residence, where the biggest percentages exist for retired Germans and British. The percentage of retired people among the stayer populations ranges from slightly below one fifth in Italy and Spain, to around one fourth in Germany and Britain and 28 per cent in France. Thus, leaving aside pre-1974 guest-worker migration or possible return migration since, intra-EU migration since 1974 has in fact increased the active population in Germany and Britain particularly, but increased the inactive population in Spain. Age of Migrants and Duration of Migration A second key dimension concerns the age of migrants and the duration of their stay. Germans and British living in Spain are the oldest migrants in the sample, which reveals a dominant pattern of retirement migration to that country (Figure 2.1). This can be seen even more clearly in their

Pioneers of European integration

60 20

40

Age at migration

80

100

34

D

GB I E

France

F GB I E

Germany

D F

I E

Britain

D F GB

Italy

E

D F GB I

Spain

Note: Countries of origin (first row below graphic): D 5 Germans, F 5 French, GB 5 British, I 5 Italians, E 5 Spanish. Source:

EIMSS, N54 901

Figure 2.2

Distribution of age at migration dependent on country of origin and country of residence

age at the time of settlement (Figure 2.2): British citizens were on average 51 years old when they came to Spain, while Italians and Spanish who moved to Germany (in order to work or to join their partners) were on average in their mid-20s. The mean year of birth for stayers ranges from 1953 in Britain and 1954 in Spain to 1956 in France, Germany and Italy. Averaged over all countries of destination, German and British movers are older than the respective stayers, and Spanish movers are younger, while there is no age difference between French and Italian movers and stayers. However, there is some variance for the single country of origin/country of residence combinations. On average, intra-EU migrants have lived in their destination countries for nearly 15 years at the time of the interview. Overall, movers from Italy have stayed somewhat longer and those from Britain somewhat less time than average. The extreme cases are Italians and Spanish in Germany on the one hand, and British in France and Spain on the other. Between the

The demographics of movers and stayers

Table 2.4 Country of residence France Germany Britain Italy Spain

35

Respondents coming from a big city or the suburbs (%) Country of origin French

Germans

British

Italians

Spanish

35.9 43.1 51.9 53.6 57.4

44.2 33.1 41.5 53.4 66.4

37.2 44.4 29.4 50.4 44.0

46.3 19.3 41.8 17.6 52.8

42.3 35.5 36.7 52.6 29.9

Notes: ‘Big city’ and ‘suburbs of big city’ are self-assessed by respondents. Stayer population is on the main diagonal (shaded) Source:

ESS 2004 and EIMSS, N513 985

former and the latter, there is a difference of almost nine years, showing the role of Germany as a traditional immigration magnet for these two Southern European countries and the emergence of new patterns of shorter-term migration in other cases. Geographic Mobility Table 2.4 shows the proportion of migrants coming from big cities and the suburbs, with the main diagonal showing the corresponding figures for the stayer populations. As the percentage of stayers who live in a big city or a suburb ranges from less than one-fifth in Italy to roughly one-third in the other four countries, one can conclude that international movers come especially from urbanized areas. This is particularly the case for those who move to Italy and Spain. There is, however, one notable exception to this rule: again, Italian migrants in Germany. The proportion here from urban areas corresponds to the Italian stayer population as a whole. At the same time, movers seem to be attracted by the metropolitan regions in the countries of residence, and are more frequently found there than the respective stayer populations of these countries. This is especially the case for migration to Britain, especially due to a ‘London effect’ (on the appeal of Britain’s capital for EU movers, see Favell 2006b). Spain and France are partial exceptions here as destination countries. This can be explained by the fact that these countries are prime targets for retirement migrants. Though there are many retired people living in Italy as well, EU movers living in Italy have to a greater degree moved there for other reasons, be it work or love, before they eventually retired (see King

36

Pioneers of European integration

and Patterson 1998). However, there is one characteristic that intra-EU migration to Italy and France seems to have in common: a trend to counter-urbanization, which consists in resettling deserted rural areas by refurbishing existing housing stock (see also Buller and Hoggart 1994; Hoggart and Buller 1995). National Origin of Partners A fourth key question is cross-national intermarriage. The two Southern European countries are positioned at opposite ends with regard to their EU movers having a partner from the country of origin: In Italy, relatively few have co-nationals as partners, whereas in Spain the vast majority does. For intra-EU migrants to Britain, having a partner from their own national community is not very common. The very low proportion of Italians in Germany having a German partner might be due to the large number of Italians in this country from which these people can choose. Our result is in sharp contrast to results for binational marriages in Germany of first-generation migrants (Schroedter 2004), which show that Italians and Spanish are the most inter-married of the traditional guestworker populations in this respect. However, these other analyses do not include immigrants from France and Britain and put a special emphasis on the low integration of Turks in Germany. In addition, they are based on a different definition of first-generation migrants, which includes all those who migrated to Germany after the age of six. Given the fact that the Italian migrants in our sample came to Germany at an average age of 26, it is obvious that they had a high probability of marrying or at least getting a partner from their home country before they arrived.

MIGRATION OVER TIME We will now look in greater detail at how intra-EU migration has developed over time. Three periods are distinguished here: from 1974 to 1983, 1984 to 1993 and 1994 to 2003. The first period, from 1974 to 1983, followed immediately the end of the active recruitment of foreign labour in the Northern industrial countries in the aftermath of the mid-1970s oil crisis. The formal restrictions to free movement which were established at that time affected only Spanish movers. The second period, from 1984 to 1993, was somewhat transitional, with the noticeable fact that Spanish citizens were granted free movement rights in 1993. The last period is characterized by decisive improvements in the mobility potential of intraEU migrants resulting from the Maastricht Treaty, while at the same time

The demographics of movers and stayers

37

economic incentives for emigration from Southern Europe were much weaker than in earlier periods. Figure 2.3 shows the number of migrants for the different periods of the groups defined by country of origin (represented by the rows) and country of residence.2 It can be seen that some of the groups have moved fairly recently, such as Germans to France, French to Germany, and British to France and Spain. In contrast, other groups moved in larger numbers relatively early, for example Italians and Spanish to Germany. We have broken down the number of migrants arriving during each of the three migration periods by country of residence and whether they have ever worked in their present country of residence (see Figure 2.4, that specifies the case for women). Each of the columns corresponds to one country of residence. The first row includes respondents who have never worked in the country of residence, and the second represents the corresponding information for those who have worked in the country. The numbers of migrants to France, Italy and Spain who have never worked in these countries sharply increases with recency of migration – in Spain, it is high for all periods of migration – reflecting the growing importance of these countries as targets for retirement migration. There are two additional possible reasons why the number of movers who have never worked in the country of residence might increase with the recency of migration. Migrants who have come for study reasons may not have entered the labour market (yet), and recent movers might not yet have been successful in finding a job, something most likely to have affected particularly family members of labour migrants. The differences between migration periods are less pronounced for those who have indeed worked in the country of residence. Gender differences are relatively small, although there is a consistently higher number of women who have never worked in the country of residence for all periods of migration. There are women accompanying or following at a later time their labour-migrant husband, without starting a work career abroad on their own. However, among those who came in the earliest period under investigation, there is not an especially high number of women who have never worked in the country of residence. This is presumably so because these women had a longer time-frame for getting a job where they moved to. Of particular interest are the changes in the formal educational level of the different groups of migrants over time (Figure 2.5). Corresponding to trends in all countries in recent decades, most migrant groups include proportionally smaller numbers of less-educated individuals in the more recent periods. Two developments, however, deserve attention. The first concerns labour migrants. While Spanish and Italians in Germany start

100 200

0

100 200

74–83 84–93 94–03

Spanish, in France

74–83 84–93 94–03

Italians, in France

74–83 84–93 94–03

British, in France

74–83 84–93 94–03

French, in Germany

74–83 84–93 94–03

Germans, in France

Germans, in Italy

74–83 84–93 94–03

Spanish, in Britain

74–83 84–93 94–03

Italians, in Britain

74–83 84–93 94–03

British, in Italy

74–83 84–93 94–03

French, in Italy

74–83 84–93 94–03

Migration period

74–83 84–93 94–03

Spanish, in Germany

74–83 84–93 94–03

Italians, in Germany

74–83 84–93 94–03

British, in Germany

74–83 84–93 94–03

French, in Britain

74–83 84–93 94–03

Germans, in Britain

Period of migration dependent on country of origin and country of residence

EIMSS, N 5 4 902

Figure 2.3

Source:

Frequency

0

100 200

0

100 200

0

100 200

0

100 200 0 100 200 0 100 200 0 100 200 0 100 200 0

100 200 0 100 200 0 100 200 0 100 200 0 100 200 0

100 200 0 100 200 0 100 200 0 100 200 0 100 200 0

38

74–83 84–93 94–03

Spanish, in Italy

74–83 84–93 94–03

Italians, in Spain

74–83 84–93 94–03

British, in Spain

74–83 84–93 94–03

French, in Spain

74–83 84–93 94–03

Germans, in Spain

74

3 3 3 –8 4–9 4–0 8 9

work, in F

3 3 3 –8 4–9 4–0 8 9

74

no work, in F

3 3 3 –8 4–9 4–0 8 9

74

work, in D

3 3 3 –8 4–9 4–0 8 9

74

no work, in D

3 3 3 –8 4–9 4–0 8 9

74

work, in GB

3 3 3 –8 4–9 4–0 8 9

74

no work, in GB

3 3 3 –8 –9 –0 74 84 94

work, in I

3 3 3 –8 –9 –0 74 84 94

no work, in I

Migration period

3 3 3 –8 –9 –0 74 84 94

work, in E

3 3 3 –8 –9 –0 74 84 94

no work, in E

Period of migration dependent on country of residence and whether they ever worked there (women)

EIMSS, N 5 4 876

Figure 2.4

Source:

400

200

0

400

200

0

Frequency

400 200 0 200 0

400

400 200 0 200 0

400

400 200 0 200 0

400

400 200 0 400 200 0

39

40

0

10

20

30

40

50

60

70

80

90

Germany

Britain

Italy

74–80 81–90 91–03 74–80 81–90 91–03

Spain

74–80 81–90 91–03

Spanish

Italians

British

French

Germans

Migrants with lower secondary education or less dependent on period of migration, country of origin and country of residence (%)

France

74–80 81–90 91–03 74–80 81–90 91–03

EIMSS, N 5 4 850

Figure 2.5

Source:

Percent

The demographics of movers and stayers

41

with a high level of less-educated migrants in the first period (80 per cent), the percentage is down to 30 per cent for the Spanish in the third one. For the Italians, on the other hand, it still hovers above 65 per cent. Germany thus continues to attract poorly educated Italians, perhaps to cater for the thriving sector of Italian restaurants, as well as other ethnic-based activities for the large and long-established Italian community there. The second concerns retirement migration. The proportion of less-educated Germans in Spain is 70 per cent in the first period and reduces by only 10 percentage points, yet the proportion of their British counterparts actually increases by nearly 20 percentage points to almost 50 per cent. A similar but not as dramatic tendency can be seen for both migrant groups in France, while for movers living in Italy (pre-)retirement migration comes closer to the idea of elite migration – if we take education as a proxy of social status.

AGE AT MIGRATION Age at migration is a crucial variable in migration research. It implicitly tells us a lot about the motives of migrants and it is also paramount in the discussion of different aspects of integration. We will now look at the age distributions of movers from both the perspective of the countries of origin and the countries of residence. Then we will explore the changes of age at migration over time. Finally we will analyse the determinants of age at migration by using a series of multi-level models. Descriptive Analysis Figure 2.6 shows that the French, Italians and Spanish predominantly leave their country of origin at younger ages, while the Germans and especially the British leave later. For movers from Britain, and to a lesser degree also from Germany, the age distribution has actually become bimodal. This hints at two distinct migration types which are determined by different processes and for which different consequences are likely to be revealed. It has to be noted that migration when older is not just pure retirement migration, but in fact includes a high proportion of ‘preretirement migration’, which might also entail labour-force participation in the country of residence, whether or not this was the main driving force behind the move. Also characteristic is the dip observable for people in their late 40s and early 50s. This might be related to the presence of children in the household, which discourages both explicit labour- and pre-retirement migration.

42

Pioneers of European integration

40 20 0

0

20

40

60

French

60

Germans

20

40

60

80

20

80

60 40 0

20

40 20 0

Frequency

60

Spanish

60

Italians

40

20

40

60

80

20

40

60

80

0

20

40

60

British

20

40

60

80 Age at migration

Source:

EIMSS, N 5 4 901

Figure 2.6

Age at migration dependent on country of origin

Figure 2.7 shows the frequency of different ages at migration for immigration into the five countries of residence. For Germany, Britain and Italy, it is predominantly the younger people who come, while for Spain it is the older age groups. France is in-between. But how do the ages at

The demographics of movers and stayers

100 50

20

40

60

80

20

40

60

80

50

100

Spain

0

50

100

Italy

0

Frequency

Germany

0

0

50

100

France

43

20

40

60

80

20

40

60

80

0

50

100

Britain

20

40

60

80 Age at migration

Source:

EIMSS, N 5 4 901

Figure 2.7

Age at migration dependent on country of residence

migration to the different countries of residence vary across the three periods we study? Most telling are the figures for Spain, showing that older immigrants become increasingly dominant. France is on the way towards a bimodal distribution of age at migration, with growing numbers of older

44

Pioneers of European integration

people coming to this country – in conjunction with a net increase of migrants, particularly in the latest period. In Italy, the age distribution of the newly arrived has not changed so much over time. In Germany and Britain, the traditional labour-migration pattern remains intact with mostly young people moving in. Nevertheless, comparing age at migration between the last and the first period in Germany, there is no doubt that the number of very young migrants has dropped. Correspondingly, migration from Britain develops a clear bimodal distribution of age at migration over the three periods, and in the last period Britains emigrating when relatively older have actually become the majority. The same tendency is visible for Germans, but it is less pronounced. The French, Italians and Spanish, however, continue to migrate at relatively young ages. Multivariate Analysis In this section we estimate a series of multi-level models that help us explain statistically a key aspect of mobility choices – that is, the age at which migrants leave their country of origin, our dependent variable. Multi-level or hierarchical linear models (Raudenbush and Bryk 2002) are appropriate when variables pertain to two levels, an individual level (in our case, the individual migrants) and a group level (in our case, the country of origin/country of residence combinations). Usually, a series of models is estimated beginning with a so-called ‘empty’ or variancecomponent model. This model will show how much of the variance in the dependent variable (age at migration) is located within groups (that is, individual variance around the group means) and how much between groups (that is, the variance of the group means around the grand mean for all groups). A second model typically includes the individual-level variables, as in a normal OLS regression. A comparison of the variance components of both models shows how much variance these individual-level variables are able to explain at both levels. The explanatory power of the individuallevel variables with regard to the group-level variance is related to a composition effect: that is, the part of the differences between groups that can be explained by taking into account the individual-level variables alone. A third model usually estimated is one which, in addition to the individuallevel variables, also includes one or more variables at the group level. As explanatory (independent) variables here we use period of migration, gender, education, motives for migration, previous work experience in the country of origin and previous sojourns of three months or more in the country of residence or a third country. Period of migration is entered as two dummy variables, with the first period as the baseline: that is, the effects are estimated for the 1984–93 and the 1994–2003 periods. Gender

The demographics of movers and stayers

Table 2.5

Random intercept models for age at migration

Fixed effects Constant Migration 1984–1993 Migration 1994–2003 Women Secondary-level education Tertiary-level education Work as motive Family/love as motive Quality of life as motive Study as motive Work experience in COO Sojourn in COR Sojourn in third country North–South migration Random effects Level 1: residual variance Level 2: random intercept N Note: Source:

45

Model 1

Model 2

Model 3

Empty model

Individual-level variables

1 group-level information

34.9

34.5 3.1 7.9 21.8 23.0 23.2 25.9 24.8 (20.2) 28.7 2.7 1.0 2.1

32.2 3.1 7.9 21.8 23.0 23.2 25.8 24.8 (20.1) 28.7 2.7 1.0 2.1 7.5

123.4 52.4 4 807

101.0 26.8 4 807

101.0 14.9 4 807

Insignificant values in parentheses. EIMSS, N 5 4 807

is entered as a dummy variable with male respondents as the baseline: that is, the effect pertains to women. Education is transformed into two dummy variables with migrants with less than university-entrance qualification as the baseline, so the effects refer to respondents with a universityentrance qualification and those with a university degree respectively. Four dummies for migration motives are entered with miscellaneous motives as the baseline category; thus, the effects pertain to the pure motives of work, family/love, quality of life and study. Previous work experience in the country of origin and previous sojourn of three months or more in the country of residence or in a third country are entered as three dummy variables with respondents lacking such experiences forming the baseline category. Table 2.5 shows the three models, a variance-component model without any explanatory variables, a model which includes the individuallevel variables and a model which additionally includes information on

46

Pioneers of European integration

whether the migrant belongs to one of the six COO/COR groups moving from the North (Britain and Germany) to the South (France, Italy and Spain). Model 1 shows that the grand mean of age at migration for all migrants is nearly 35 years. Of the variance of age at migration 29.8 per cent is related to national COO/COR group membership and 70.2 per cent is attributable to within-groups individual differences. The group-level variance is extremely high, indicating that we can predict nearly one-third of the variation of the age of migration alone by group membership, without taking any individual characteristics of the migrants into account. Model 2 introduces effects at the individual level, results which can be interpreted as averages for all migrant groups. Most marked are the differences between the three periods. The migrants of the second period are three years older than the migrants in the first period and those of the third period nearly eight years older, controlling for all the other independent variables. Women moved on average nearly two years earlier than men, and both the intermediate and the highly educated migrants came three years earlier in their lives than the less-qualified. Age at migration also varies considerably by migration motives. While the quality-of-life migrants do not differ from those who came for miscellaneous motives, if work was the only motive, migrants came about six years earlier, if study was the only motive about nine years earlier, and those for family or love reasons five years earlier. Migrants with previous work experience in their country of origin, that is, those who were sojourners abroad before they finally migrated to their present country of residence, moved on average three, one and two years later in their lives, respectively. Note that the different effects are independent of each other. Together these sociodemographic and behavioural variables explain 18.2 per cent of the individual-level (residual) variance of age at migration. They have an even higher impact at the group level, where they explain more than 50 per cent of the variance between groups. This comparatively high value is due to the large compositional differences between the COO/ COR combinations. Controlling for individual-level characteristics of the migrants, we can thus reduce these compositional effects. Though half of the variation of the random intercept is explained by individual-level variables, the variance component is still highly significant. This indicates that the means of age at migration in the different COO/COR combinations are very different from each other, even after compositional differences between these combinations have been taken into account. Finally, in model 3 a dummy variable for North–South migration is entered referring to British and German migrants to the other three

The demographics of movers and stayers

47

countries. The dummy variable shows that the North–South migrants are seven and a half years older than all the other migrants, over and above what is already explained by individual-level variables. The individuallevel effects remain unaffected by the inclusion of this group-level variable, with the only exception being the intercept, which changes in interpretation: now it refers to all migrants with baseline-category values on all individual-level variables who are not members of a North–South migrating group. There is again a marked change in the size of the random intercept, as this declines by 44 per cent (compared to model 2). This decline in unexplained variance at the group level is entirely due to the explanatory power of the North–South origin dichotomy we have constructed. We can thus expect that older migrants from Northern Europe can be singled out as a type of movers with highly distinct characteristics.

A SIMPLE CLASSIFICATION OF MIGRANTS In this final section, we will go on to present the results of a simple cluster analysis, building on the multivariate analysis. We use only age at migration and the duration of the stay in the country of residence as clustering variables and require four clusters to be computed by the KMEANS4 procedure. The idea behind this analysis is to get something similar to a crossclassification of age at migration and the length of the stay in the country of residence. The resulting clusters are then described by additional variables which were not used in the definition of the groups (Table 2.6). Cluster 1 (1539 movers) consists of migrants who moved on average when they were 26 years old and have already stayed in their countries of residence for an average of 24 years. We label them late traditional migrants. Three-quarters of these migrants came in the first period and none in the last. This group is not too unevenly distributed with respect to the country of origin (only 16 per cent of this group are Germans, versus 25 per cent Italians), but it is with regard to the country of residence (only 10 per cent of this group are in Spain, but 30 per cent in Germany). Only one-third of this group has a university degree. Nine out of ten have worked in the country of residence. The dominant migration motive was love, but work was also important. Half of them has a country of residence partner. Cluster 2 (916 movers) has an average age at migration of 44 years and 13 years of stay in the country of residence. We label them pre-retirement movers. These migrants have moved mainly during the last two periods. There is a particularly strong representation of Germans and a weak one of Spanish. France and especially Spain are overrepresented as countries

48

Pioneers of European integration

Table 2.6

Description of the four clusters Cluster 1 Cluster 2 Cluster 3 Cluster 4 Eurostars Pure Late Preretirement traditionals retirement movers movers

Age at migration Years of stay Migration from 1974–83 (%) Migration from 1984–93 (%) Migration from 1994–2003 (%) COO: Germans (%) COO: French (%) COO: British (%) COO: Italians (%) COO: Spanish (%) COR: Germany (%) COR: France (%) COR: Britain (%) COR: Italy (%) COR: Spain (%) Gender (%), female (%) University entrance qualification (%) University diploma (%) Ever worked in COR (%) Migration motive: Work (%) Migration motive: Love (%) Migration motive: Quality of life (%) Migration motive: Study (%) Partner from COO (%) Partner from COR (%) Partner from third country (%) Number of migrants Source:

26 24 75 25 0 16 20 18 25 21 30 14 21 25 10 51 26

44 13 17 37 46 30 19 23 19 10 11 26 14 19 31 50 21

28 8 0 31 69 17 23 13 21 27 25 21 25 19 11 55 27

60 8 3 19 77 29 15 40 12 4 3 26 4 17 51 42 22

34 89 27 42 18

43 54 23 28 38

55 81 31 32 16

36 18 6 14 69

10 27 50 6 1539

4 55 19 8 916

17 35 37 11 1721

4 60 9 9 725

EIMSS, N 5 4 901

of residence. Four out of ten migrants have a university degree. Only half of them have ever worked in the country of residence. The dominant migration motive is quality of life, though half of them have also moved for work or love reasons. Only two out of ten have a partner from the country of residence. Cluster 3 (1721 movers) has moved on average at the age of 28 (close to

The demographics of movers and stayers

49

the ‘late traditionals’), but have stayed for an average of eight years only. We call them Eurostars, as they seem to embody the ideal type of Favell’s (2008a) study of European free-moving professionals. Two-thirds of them have arrived in the last period, and the remaining one third in the intermediary period. British (and German) nationals are underrepresented in this group and the Spanish are overrepresented. As for the destinations, the picture is reversed: Britain and Germany are overrepresented and Spain underrepresented. More than half hold a university degree, a proportion higher than in any other cluster. Eighty per cent of them have worked in their country of residence. In addition to love and work, study motives were important for nearly one out of five, more than in any other group. More than one-third have a partner from the country of residence. Individuals in cluster 4 (725 movers) migrated at an average age of 60 and stayed for eight years. We call them pure retirement movers. Nearly eight out of ten came in the last period. There is a particularly strong representation of British and Germans, and a particularly weak one of Spanish. France and especially Spain are overrepresented as countries of residence (actually, half of the cluster is located in the latter country), while only very few in this group have gone to Germany or Britain. While the gender distribution for the other groups is balanced, only 40 per cent are women in this group. The proportion of migrants holding a university degree is one-third – that is, as low as among late traditionals. Seven out of ten name ‘quality of life’ as their migration motive. Less than 10 per cent have a partner from the country of residence. A caution should be added here in particular with regard to the proportions of country of origin migrants in the four groups. For reasons mentioned earlier, the figures refer to the EIMSS sample and do not reflect the relative numbers of different nationalities we would find if we were to sample proportionately to the number of foreign nationals in the country. That is, the results should not lead us to conclude definitively that there are far more Spanish Eurostars out there than German and British ones. This is because the huge numbers of German and British pre-retirement and retirement migrants we found in Spain and France technically set an upper limit to the number of Eurostars we could sample there.

CONCLUSIONS This chapter has offered a preliminary description of the sample of EU movers we surveyed with EIMSS, as well as a preliminary multivariate and cluster analysis. It shows how this migration has changed over time, becoming a better educated and more diversified movement. With very

50

Pioneers of European integration

few exceptions, which can be related to the remnants of the typical ‘guestworker’ migration, geographically mobile Western Europeans now tend to come from higher social backgrounds and have a higher level of education themselves than the respective stayer populations. Economic and labour market differences between the five countries under investigation have largely vanished in recent decades. In addition, the supply of low-qualified labour from these five countries has declined with the expansion of education and with the rising immigration of third-country nationals filling up the bottom of the labour market. Thus, the migration of Western unqualified labour has become dependent on the existence of ethnic niches which provide a relatively protected labour market – as is especially the case for Italians in Germany. One major finding of these analyses is that geographical mobility within the EU embraces a variety of social types. In addition to the remnants of the typical ‘guest-worker’ migration and the ensuing family reunification (which are represented best by the ‘late traditionals’ type of our classification based on age at migration and duration of stay), we find increasingly retirement migration and study migration but also a ‘free mover’ kind of migration, what we call the ‘Eurostars’. The latter is characterized by individualistic motives and migration strategies, beyond formal recruitment and chain migration. This type also encompasses growing student migration, which has its own dynamics. There is also a trend towards movement at older ages. This is driven by the increase of retirement (the ‘pure retiree’ type) and part-time work (or ‘pre-retirement’) migration, but also by highly qualified labour migrants crossing borders more frequently. This heterogeneity of intra-European movers has to be taken into account when making descriptive statements or testing hypotheses in migration research in Europe. What can we learn from this about possible future trends of intraEuropean migration? We surmise that for migration inside EU15 countries, the trends observed here will persist in the future. There is likely to be a further reduction in the number of unqualified labour migrants and their families (with the only exception being ethnic niches such as the Italians in Germany), and a rise in retirement, study and highly skilled migration. Age at migration will thus increase, unless study migration gets additional momentum. Such an increase in study migration would, as a consequence, also lead to an influx of young highly qualified workers, further slanting migration in Western Europe towards the higher social end. The situation with regard to the new accession countries’ citizens, who are beginning to get unrestricted access to EU15 member states (with variation as to which countries open their doors and when), will nevertheless, at least for the next decade, remain different. Push and pull factors will

The demographics of movers and stayers

51

remain intact for A12 nationals, as long as the economic development of the bulk of these countries still lags behind that in the West. Thus, a considerable supply of unqualified labour can still be expected from Eastern Europe. Given the stronger economic growth in this area compared to the EU15, however, the time frame might be very short. The patterns of migration which we observe for the EU15 now may well eventually be generalizable to the entire EU27.

NOTES 1. For further details on sampling and other features of the survey, please refer to Appendix A. 2. This and the following figures use frequencies instead of percentages. This should increase readability of those figures in which the information is broken down by more than one independent variable. 3. This is computed as the variance component at the group level divided by the total variation, that is, 52.4/(52.41123.4). 4. KMEANS is the standard algorithm for calculating clusters.

3.

Deciding to move: migration projects in an integrating Europe Oscar Santacreu, Emiliana Baldoni and María Carmen Albert1

INTRA-EUROPEAN MIGRATION IN THE CONTEXT OF OTHER INTERNATIONAL MIGRATION The aim of this chapter is to describe the motivations, social organisation and previous mobility experiences that characterize intra-EU migration within the EU15. We thus focus on the subjective motivations, the history of migration, the personal networks and the family support structures of mobile citizens from Spain, Italy, France, Britain and Germany. As other migration scholars have pointed out (most notably King 2002), current migration patterns can no longer be captured within a traditional labour migration paradigm. Rather, they reflect emerging characteristics of European societies more generally: privileging flexibility in time and space, alternative modes of consumption and leisure, and a search for healthier environments, self-fulfilment, new lifestyles and a better quality of life. We provide answers to the following questions. Why do EU15 citizens migrate? How are their motivations distributed among the five countries under scrutiny? Do men and women share the same motivations? What is the role played by human capital? How do family, relationships and personal networks influence the migration decision-making process? Are migrants within the EU a ‘supermobile’ population? What is the part played by previous mobility experiences? Do movers retain the same motivations in subsequent mobility experiences? Is mobility within the EU ‘international’ or ‘internal’? The first part of this chapter looks at migration within the EU15 in the context of other studies on international migration, particularly the more widely available research on the EU10 new member states admitted in 2004, as well as other recent attempts to capture the diversity of migration in Europe today. The second part is a descriptive analysis of the socioeconomic reasons for migration (that is, work) in relation to other motivations such as study, family/love, or quality of life. The final section focuses 52

Deciding to move

53

on previous mobility experiences of mobile European citizens and their connection with current experiences. Despite the potential interest of new forms of mobility associated with the EU as a unique experiment in regional integration (Favell 2008a), studies of international migration within and among the EU15 member states that constituted the EU until 2004 have been rare. International migration research, rather, typically focuses on classic forms of immigration into clearly delineated national political units; it considers only standard forms of migration motivations such as economic push/pull factors or asylum seeking; and it draws a firm line between international migration and forms of ‘internal’ migration within nation states (Massey et al. 1998; Portes 1999; Castles and Miller 2005). All of these staple distinctions in migration research are challenged by the emergent new forms found within the EU. Distinctions with classic migration can immediately be found in terms of motivations. Here, a certain number of previous studies do exist. These include investigations carried out by Eurostat (Eurostat and European Commission 2002) and the European Foundation for the Improvement of Living and Working Conditions (2006) based on the analysis of data from the 2001 and 2005 Eurobarometers regarding the mobility of EU citizens. According to Eurostat, which only takes into account short-distance mobility, the main reasons for migration are family/personal reasons (54 per cent), followed by housing (18 per cent), and work-related reasons (15 per cent) (Eurostat and European Commission 2002, 16). The 2005 Eurobarometer (EB 64.1) also considers long-distance mobility, that is, moving from one region to another, or to another country within the EU. Here, among the most frequent reasons for moving cited are firstly those associated with the labour market (‘new job or job transfer’, 34 per cent), and secondly those involving family or partners (‘change of marital situation’, 18 per cent), or following the partner to a different location (14 per cent) (European Foundation for the Improvement of Living and Working Conditions 2006, 18–19). However, these studies are limited because the Eurobarometer did not pay attention to the category ‘quality of life’, or offer the finer distinctions between motivations that were built into our study. A far broader resource for considering these questions is provided by the more numerous studies on migration that have focused on analysing East–West mobility intentions and flows, before and after the EU enlargements of 2004 and 2007 (Bauer and Zimmermann 1999a; Brücker 2003; Gorny and Ruspini 2004; Kelo and Wächter 2004; Krieger 2004; Krieger and Fernández 2006; Favell 2008a). Many of these studies suggest that intentions to move outnumber the actual number of movers, and

54

Pioneers of European integration

the best guide to future East–West migration from new member states remains the limited and often circular trends established by the previous Mediterranean enlargements of the EU involving Spain, Portugal and Greece (Kupiszewski 2002). However, the latest data from Eurobarometer 2005 (Krieger and Fernández 2006) have suggested that there is an upward trend in mobility intentions in some of the new member states (especially Poland, Latvia, Estonia and Lithuania), particularly linked to the increasing reliance on Eastern workers to fill low-level or skilled service-sector jobs in the West (Favell and Hansen 2002). In a study for the International Organisation for Migration, Wallace (1998, 27–28) points out the influence of several economic factors of East–West migration that can still be classified according to classic pushpull theories. Among the pull factors are better living conditions, higher salaries, other people’s (positive) migration experiences, better job prospects and more personal freedom. On the other hand, only two salient push factors are identified: ethnic problems and the economic conditions in the country of origin. Following the push-pull approach, Kelo and Wächter (2004, 71) show that East–West movements are indeed more ‘pulled’ than ‘pushed’. The traditional push factors, such as famine, war, persecution, ethnic and racial discrimination, and racism among others, are hardly present at all within Europe, even at its Eastern extremes. There are however strong pull factors in terms of job prospects, higher income or family connection. Mobility intentions are based mainly on the search for job opportunities, higher earnings and, as the third main reason, a change in marital status. Krieger (2004, 36–7) concludes that, despite the importance of economic reasons, there are significant differences between countries, as well as different sets of motivations and expectations among the citizens of the new member states. In this sense, economic and financial motivations are more important in Hungary, Lithuania, Slovakia, Turkey, Bulgaria and Romania, whereas family reasons appear to be more significant in Malta and Cyprus. Moreover, there seems to be a pattern that combines economic, financial and family reasons in the Czech Republic, Latvia and Slovenia. Despite some variations, then, economic motivations do still prevail within East–West migration, something which further underlines the potential distinctiveness of intra-EU mobility within the ‘old Europe’. The starting point for our analysis is work by human geographers such as King (2002) and Lazaridis and Williams (2002), who have begun to chart a ‘new map’ of migration in the integrating space of the new Europe that was already emerging before the enlargements of 2004 and 2007. They seek to identify the new motivations for migration within Europe, drawing a distinction between traditional migration and the emerging types of spatial mobility: travel, tourism, circulation and commuting. These have

Deciding to move

55

been much more marked among the older member states of the West. Clues to these characteristics are provided by various forms of migration that are typically marginal to mainstream migration studies: that is, highskilled migration, student migration and retirement migration. All play an important part in the new intra-EU migrations. Global and regional economic integration have made high-skilled migration an ever-more relevant category of spatial mobility (Salt 1992; Iredale 2001; Mahroum 2001; Smith and Favell 2006). Within Europe, certain destinations have certainly become privileged in terms of their attractiveness as hubs in the global economy. For example, research funded by the British Department of Trade and Industry shows that among the main reasons why high-skilled migrants choose Britain as a destination are knowledge of the language and an assumed familiarity with the country’s culture – salaries are not necessarily the main factor in their decision. On the other hand, the reasons deterring people from moving to Britain include climate, distance from family and friends, the standard of public services and, finally, difficulties in obtaining a work permit. The fact that among high-skilled migrants from developing countries prospects for economic improvement are significant, but not a dominant factor, indicates that such migrants surveyed in this context can be considered ‘knowledge migrants’ rather than ‘economic migrants’ per se (Pearson and Morell 2002, 12). Mahroum (2001, 29) in turn puts forward the differences between employment sectors, and also considers personal aspirations and scientific curiosity to be more significant than salary conditions. Favell’s study (2008a) on the mobility of young professionals living in large European cities points out that the motivations of these EU citizens are basically related to ambitions that include adventure and personal self-fulfilment more than just income as such. Drawing on previous research on the mobility of highly skilled workers in different employment sectors, Ackers (1998 and 2005) analyses the mobility of scientists within the EU from an economic and personal perspective. Ackers highlights above all gender differences in their motivations. For women, the decision to migrate is a complex issue that is not based on a single factor. Women’s intra-EU mobility does not fit in with the traditional ‘male breadwinner’ family model. Instead, in the first stage of their migration, mobile women are mostly single and their motivations are associated with the search for career opportunities. However, in what is called ‘post-migration’, the traditional male breadwinner model is suitable, since women pair off with men from the destination countries, and this leads them to have mobility motivations linked with their partner (that is, family/personal reasons). Within the new map of European migration, students are also

56

Pioneers of European integration

increasingly important actors of mobility. Despite the difficulty of finding reliable data (Poulain et al. 2006), there is no doubt that student exchange and mobility, both internal and international, is ever more significant. As King reminds us (2002, 98–9), this is quite an important form of European mobility, promoted by the European Commission through the Erasmus and Socrates programmes, as well as the traditions of some EU countries that encourage students to leave their homes when they go to college. Its relevance as a form of mobility is established by a study of the characteristics of young students of the University of Sussex who took part in an international mobility programme (King and Ruiz-Gelices 2003). According to the authors of the study, the motivations of the students in exchange programmes relate to three aspects: linguistic self-improvement, the cultural experience of living in another country, and general personal development. All these reasons have, in the opinion of the students, allowed them to improve their employment and career opportunities, but they clearly cannot be reduced only to economic motivations. As King argues, it is possible to view student migration as a subset of youth migration, motivated by a mix of education goals and the combination of experience and pleasure-seeking facilitated by travel and forms of casual or temporary work (King et al. 2004, 99). Finally, research on the migration of European retirees has grown significantly over the last few years. The main thesis of these studies is that mobile retirees go in search of healthier environments, where weather and the quality of life are essential. Among these retirees we also find what we could call ‘part-time retirees’, that is, workers who decide to spend part of the year in southern European countries, who give up part of their income in exchange for a better quality of life. Many of these studies of the migration of retirees have examined the Spanish case, particularly the settlements on the Mediterranean coast (Rodríguez et al. 1998; O’Reilly 2000; Casado-Díaz 2005; Huber 2004; Huber and O’Reilly 2004; Kaiser and Friedrich 2004). Other studies have explored this phenomenon by comparing Mediterranean countries (Italy, Spain and Portugal) (King et al. 1998 and 2000; Casado-Díaz et al. 2004). According to Casado-Díaz et al. (2004), although the search for a better quality of life (through the climate) remains the main motivation to move, other reasons are increasingly relevant: economic advantages (lower cost of living), the Mediterranean lifestyle, social life and, finally, easy and cheap communication with the migrants’ country of origin thanks to low-cost flights. Again, pure economic motivations are hard to isolate. Nevertheless, there are differences when it comes to choosing a destination. Climate is more important on the Costa del Sol (southern Spain), whereas the environment and the landscape are the main aspects in Tuscany (Italy). The fact that international

Deciding to move

57

retirees living in Tuscany enjoy a higher socioeconomic status means that the cost of living is not an important motivation in their decision to move, whereas it is a noteworthy reason in Spain, Malta and Portugal (King et al. 1998, 100–101).

REASONS FOR INTRA-EU MOBILITY Migration Motivations In this section we can now pose a series of key questions that will enable us to better characterise intra-EU mobility based on the migration projects of respondents to the EIMSS survey. This draws on the findings of existing studies about the motivations and various types of migration within the new Europe, but allows a much more detailed and representative breakdown of these factors. We ask whether and how since 1973 the mobility of workers has lost its predominant role, how this varies by country of origin or destination, and whether other factors identified in the general migration literature such as the feminization of migration or the growing role of human capital are also shaping these mobility choices within Europe. A first overview of the reasons for migration uncovered by the EIMSS survey reveals above all the importance of the affective sphere (see Table 3.1). 29.2 per cent of the answers given refer to reasons of a sentimental nature (‘to live with partner/spouse/children’), with a further 6.2 per cent indicating reasons linked to the family of origin (‘to live together with members of the family’). People in Europe, first of all, move because of love and family. This is particularly striking in terms of mobility to Italy (44.2 per cent), dwarfing economic, educational and quality of life reasonings there. Obviously, more needs to be said about migration ‘for love’. We might indeed speculate about the cultural impact of mixed couples, considering their strategic role as laboratories of re-elaboration of bi-national identity and the construction of new meanings, languages and styles of social interaction. Examining the nationality of partners, it becomes evident that 36.2 per cent of the respondents who have mentioned ‘love’ as their reason to move have a partner of the same nationality, while as many as 61.6 per cent are married or live with a citizen of the country of destination. A new characterization of the EU-mover for love – who leaves the country of origin mainly to form a new family with a citizen of the country of destination – would seem therefore to prevail over the traditional migrant story of moving for love, where a spouse follows a partner of his/her own nationality, either leaving their country of origin together or in order to reunite a family. It is interesting to observe that the two

58

Pioneers of European integration

Table 3.1

Main reasons for migration to the country of residence (COR) of EU movers (%)

Reasons

France Germany Britain Italy Spain Total

Work To look for a job To accept a job offer To start a business Because of my occupation Education To study in secondary school To participate in a study exchange programme To study at university level (undergraduate) To study at graduate/ postgraduate/specialisation level To do an internship To learn the language Quality of life To gain new experiences To live in a better natural environment To live in better/healthier weather, enjoy climate Love/family To live together with members of family To live with partner/spouse/ children Source:

7.3 13.4 1.2 4.2

25.5 19.7 0.4 8.2

14.2 22.4 1.1 5.1

5.1 14.6 0.5 6.5

2.1 6.1 3.2 1.7

10.9 15.2 1.3 5.2

0.2 0.4

0.6 1.2

0.9 0.9

0.4 2.3

0.7 0.5

0.6 1.1

2.2

2.7

3.3

2.0

0.1

2.0

1.8

1.4

2.8

1.4

0.8

1.6

1.1 4.6

1.1 3.6

0.6 12.2

0.8 1.9

1.0 6.2

0.9 5.6

10.1 21.9

7.0 1.8

11.1 4.8

6.6 12.3

12.6 37.4

9.4 15.7

9.0

1.0

0.3

8.9

49.3

13.7

4.2

7.4

4.2

4.5

10.5

6.2

29.6

31.7

30.9

39.7

13.8

29.2

EIMSS, N 5 4 878

different types present distinctive specific traits. Same-nationality couples are mentioned mainly by interviewed women and among those of Italian nationality. In contrast, mobility to live in the country of the partner is mentioned especially by male respondents (82.4 per cent versus 49 per cent of the women), and is not especially related to nationality. The second reason for moving most often mentioned is ‘to live in a better natural environment’ (15.7 per cent), closely followed by ‘to live in better/ healthier weather, enjoy climate’ (13.7 per cent). Underlining the centrality

Deciding to move

59

60

50

40 COR France COR Germany COR Britain COR Italy COR Spain

30

20

10

0 Work

Source:

Study

Family/Love

Quality of life Misc. reasons

EIMSS, N 5 4 706

Figure 3.1

Reasons for movement to the country of residence of EU movers (index) (%)

of forms of mobility free from purely material considerations, both these motivations are considerably prevalent in France and Spain. It is only at this point that economic factors begin to have weight. The third ranked reason for mobility is ‘to accept a job offer’(15.2 per cent), followed by ‘to look for a job’ (10.9 per cent). The migration of workers with a job offer is most common in Britain (22.4 per cent), whilst the riskier option of job search is more frequent in Germany (25.5 per cent). A minority of people interviewed (5.2 per cent), moreover, stated that they had moved ‘because of my occupation’, in this case, most likely as a result of intra-firm mobility within multinational companies. These are directed mostly to Germany and in a smaller degree to Italy. Finally, learning the language of the country as a reason to move there is mentioned by only 5.6 per cent of respondents, while ‘to participate in a study exchange programme’ is mentioned only by 1.1 per cent of respondents. It is worth noting that Erasmus grants, in spite of their overall success in other terms, may thus have been of fairly marginal importance with regard to definitive or long-term migration. In Figure 3.1, all these detailed reasons for migration have been collapsed into an index composed of four ‘pure’ categories, which counts those who have indicated only one motive for migration or two motives which fall within the same category)2, as well as an additional category including multiple motives which refer to different categories (‘miscellaneous reasons’).

60

Pioneers of European integration

The reduction in categories helps clarify the findings. The category ‘love/ family’ (29.8 per cent) prevails over reasons regarding jobs (25.3 per cent) and ‘quality of life’ (23.9 per cent), while reasoning related to study (‘study reasons’) applies to only 7.1 per cent of the sample. These findings suggest three points. First of all, no truly dominant reason for moving emerges. This suggests the need for finer-grained distinctions among countries. As we see here, ‘work’ reasons prevail in Germany, ‘family/love’ in Italy and ‘quality of life’ in Spain. However, in terms of the overriding economic focus of most forms of international migration studied, it is clear that intra-EU migration shows a much greater tendency towards affective factors, which by no means always stem from the classic pattern of reuniting families. Equally, far more significant are choices about living in a better environment, meant as either a natural environment (a healthy climate, beautiful countryside and so on) or in seeing migration as a social space in which to explore or achieve personal lifestyle ambitions. Thirdly, the presence of a relevant number of respondents who indicate two fundamentally different reasons for mobility (13.9 per cent) proves that migration projects are often complex and multi-dimensional actions. Country, Period, Gender and Human Capital Effects Although intra-EU mobility does not appear to be guided by a single predominant reason, the analysis by country of residence and country of origin clearly identifies different types of migration. Work-driven movements are still the prevailing form of migration from southern Europe (in our survey, Italy and Spain) towards central and northern Europe (Germany, France and Great Britain) (Figure 3.2). Of particular note are the 64.8 per cent of Italians in Germany who resettled there to improve their employment situation. This type of migration has deep historical roots: in the period between 1946 and 1970, Italy was the most common origin and West Germany the most common destination of all the flows towards Central-Northern European industrial areas (Pugliese 2002, 21–2). Italians in our sample have also mostly moved for work reasons in Britain (37.9 per cent) and in France (35.1 per cent). More generally, as already shown in Figure 3.1, mobility for work represents the main form of migration into Germany (42.3 per cent) and Britain (31.7 per cent). This finding confirms the persistence of the classical South–North route of intra-European mobility. However, there is also a noticeable proportion of work-related French (31.8 per cent) and British migrants (29.7 per cent) in Italy, British in Germany (43.7 per cent) and Germans in Britain (38.2 per cent).

Deciding to move

61

70 COR France COR Germany COR Britain COR Italy COR Spain

60 50 40 30 20 10 0 France

Germany

Britain

Italy

Spain

Country of residence (COR) Source:

EIMSS, N 5 1 186

Figure 3.2

Labour mobility by country of residence (COR) and country of origin (COO) (%)

In Spain, on the other hand, mobility for ‘quality of life’ reasons is clearly predominant (56.5 per cent). The climate, the weather and the natural beauty of Spain have a great attraction for respondents of all nationalities, but above all for British (70.3 per cent) and Germans (69.4 per cent). ‘Quality of life’ constitutes the main motive for migration also for the German and British respondents in France (40.3 per cent and 61.4 per cent respectively) and for Germans in Italy (39.9 per cent). The love/ family dimension has priority for Spanish respondents in Italy (56.4 per cent), France (47.9 per cent), Germany (41.5 per cent) and Britain (32.1 per cent). Love/family reasons, meanwhile, were mentioned by 45.7 per cent of French migrants in Italy and 40.8 per cent of Italians in France. Love/family reasons play a fairly important role also in Britain, next to reasons for work, especially among French (36.2 per cent) and German respondents (33.5 per cent). ‘Study’ is a more common reason to migrate to Britain than to any other of the countries under inquiry (mainly for French and Spanish movers, 17.4 per cent and 15.4 per cent). Finally, ‘miscellaneous reasons’ have been indicated above all by Germans living in Spain (22.3 per cent), France (21.6 per cent) and Italy (16.6 per cent). Viewed in terms of periodisation of migration over the three decades (1974 to 2003) covered by our sample, reasons for mobility have in fact

62

Pioneers of European integration

undergone significant evolutions. On the whole, the love/family reasons, which prevailed in the 20 years following the ‘immigration stop’ policies enacted in West European countries in the early 1970s, have progressively lost momentum, while ‘quality of life’ has been on the rise. Work-related migration has ever picked up as expected by European institutions, despite the continued attempts to uphold directives and laws easing the mobility of workers. In the most recent decade, it has shown only a slight increase as compared to the previous period. After a significant increase at the beginning of the 1980s, mobility for study has also remained substantially stable. However, significant differences are found according to the country of destination. Taking into consideration the whole period, work mobility dropped in France and in Spain. In Italy, on the contrary, it increased, while in Germany it remains more or less constant at high levels. Mobility for family/love reasons has declined, leaving room for an increase of movements for quality of life, especially in Spain (59.7 per cent) and France (42.3 per cent) in the last decade. In this more recent period, migration for mixed reasons peaked in Germany (20.4 per cent), perhaps as an effect of the transnational movements of families who settled there in the years of industrial migration. Gender differences are also noteworthy. In the literature, the so-called feminisation of migration is well documented, understood not only as an increase in the quantity of migrating women, but also as a change in the reasons for mobility, that is, an effect of the increased level of female education and the transformations of women’s roles in society (Kofman 1999). For some women, the experience of migration seems to have assumed therefore an aspiration to better job opportunities and personal self-fulfilment, independent of family ties (Ackers 1998). In our sample, however, a quite classic pattern of mobility, which sees men emigrating mainly for economic reasons (33.1 per cent) and women for reasons of love/family (37.4 per cent), is confirmed (Krieger 2004). But also for women, differences in motivations are linked to the destination they choose. In particular, in Germany and Britain, Italian women have migrated more for work than for love/family reasons (53.8 per cent and 35.9 per cent respectively). This same country-based logic applies to male movers. Work mobility is particularly widespread among Italian men migrating to Germany (74.3 per cent of them mention ‘work’ and only 9.6 per cent ‘love/family’). On the other hand, in the case of Italian men in Spain and Spanish men in Italy, family reasons prevail significantly over work reasons (21.9 per cent against 13.7 per cent and 30.9 per cent against 25.9 per cent), whilst for the Germans, Italians and Spanish in France, for the Germans in Italy and the French in Spain, the number of labour-related and love/family-related

Deciding to move

63

movers are substantially the same (no more than three percentage points of difference). An intriguing association between sexual orientation and the reasons for migration is revealed by our data. In our sample, a full 6 per cent of respondents declared that they had a partner of the same sex. This is a quite considerable proportion, given the intrusive nature of the question and the instrument used to pose it (an impersonal phone call). While the nature and the scale of the survey does not permit a deeper investigation of the link between mobility and sexuality, we have to note that for the majority of this group of respondents the main reasons for mobility are either ‘miscellaneous’ or ‘quality of life’, responses which might contain some hint that mobility has been used to better enable these people to fulfil their own sexual orientations. Neo-classical economic theories of migration (such as Harris and Todaro 1970) underline the centrality of human capital as the trigger of geographic mobility. At the micro level, these approaches treat migration as an investment increasing the productivity of human resources, an investment which has costs and returns (Sjastaad 1962, 83). In particular, it is usually assumed that higher levels of education offer increased income returns for specific segments of the labour market (Krieger 2004). Other scholars, instead, reject this assumption given the prevalence of low-skilled jobs for migrants in many destination countries (Bauer and Zimmermann 1999b). Starting from these theoretical considerations, the survey explored the possible links between forms of mobility and two fundamental aspects of human capital, measured in terms of education and age at migration. On the whole, the reasons for migration do not seem to be significantly associated with levels of education, not even controlling for gender. It can be noted that 63 per cent of labour migrants in our sample, in the most recent decade (1994–2003), were in fact highly educated. This confirms the rise of highly skilled mobility involving not only multinational corporations, the banking and the finance industry, but also sectors such as research, IT, tourism and marketing (as has been discussed inter alia by Salt 1992; Rodríguez-Pose 2002; Ackers 2005; European Foundation for the Improvement of Living and Working Conditions 2006; Favell 2008a). While in Germany about half of the respondents who immigrate for work (the greater part of which are of Italian nationality) possess a low–medium education level, high-skilled individuals form the majority of mobile workers in all the other countries analysed. In particular, in Germany there is a large presence of high-skilled English workers (44.5 per cent); in Britain and France the largest numbers of high-skilled workers are Italians (37.1 per cent and 34.2 per cent); in Italy 35.6 per cent

64

Pioneers of European integration

of high-skilled movers come from Britain and 34.9 per cent from France; while in Spain, besides the French (43.8 per cent) there is a significant quantity of high-skilled Italians (31.3 per cent). High-skilled mobility in the EU is both centrifugal and centripetal, without a neat distinction between sending and receiving countries. However, Germany and Britain do stand out as the prevailing destinations of European high-skilled workers in our survey. It must also be noticed that among love/family and quality of life movers as well, the majority of French and British respondents have a tertiary education (56 per cent and 43.3 per cent respectively for the first type; 54.4 per cent and 44.3 per cent for the second). In contrast, more than half of the German interviewees who emigrated for quality of life have a vocationally oriented secondary level degree (53.1 per cent). The effect of education on the inclination to move is controversial because it also interacts with the influence of age (Kalter 1997; Bauer and Zimmermann 1999b). Generally speaking, though, the young show a higher tendency for migration than people over the age of 40. From an economic point of view, the idea is that the older the potential migrant, the fewer the number of years (s)he has to cash in on human capital investment (Krieger 2004, 84). In other words, older workers have a lower economic incentive to move, as the amortisation period for their investment is shorter. However, such a conclusion overlooks the impact of the non-monetary factors that have emerged as so important in the EIMSS survey. In our sample, then, the respective ages of migrants who moved for work and those who moved for love/family reasons is revealing. Labour-related movements in our sample took place when respondents were on average 31 years old – not as young as might be expected. Yet among all nationalities, work migrants leave their home country earlier than average migrants as a whole. Movers for love/family reasons packed their luggage slightly later, on average when they were 32 years old. At the opposite end of the scale, migrants in search of a better quality of life left when they were on average 45 and a half. Given their age, we can hardly reconcile their movement with a theory that understands this only in terms of human capital investment, even if indirect. Other returns – such as psychological and physical well-being – need be taken into account to make sense of their choice. Families, Friends and Migration Choices Our overview of migration motivations is complemented by information about the family and social support available at the time of migration. Even in a relatively mobility-friendly political space like Western Europe,

Deciding to move

65

migration is a ‘brave decision’ which is pondered and sometimes opposed by parents, spouses and friends. Among our respondents, slightly more than half the families of migrants (51.8 per cent) are reported to have favoured mobility, while 35 per cent were ‘neutral’ and one out of six opposed it. No significant deviation emerges when these figures are differentiated by motives for migration or country of origin. One interesting fact that did emerge however was that Italian parents were more likely to support daughters than sons when they moved – an anomaly which might be interpreted either culturally as an expression of the ‘possessiveness’ of Italian mothers over their sons, or structurally in the light of the particular hardships of the labour market for young women south of the Alps. The central role of migrant networks in aiding and even explaining migration via the effects of social capital is amply documented in the classic literature regarding migration, for instance between Central America and the US (Massey 1999). Other people’s experience with migration is usually considered as one of the main pull factors of mobility (Wallace 1998). In our study, the existence of a social network as a possible incentive for migration was examined by asking respondents if they knew co-nationals already residing in the country of destination before moving. More than one-third answered positively, with some significant differences among all nationalities. In particular, as shown in Figure 3.3, a pre-existing network of national acquaintances was mentioned mainly by the Italian and Spanish respondents who are resident in Germany (37 per cent and 40 35 30 COO Germany COO France COO Britain COO Italy COO Spain

25 20 15 10 5 0 France

Germany

Britain

Italy

Spain

Country of residence (COR)

Source:

EIMSS, N 5 1 899

Figure 3.3

Co-national acquaintances before settling by country of residence (COR) and country of origin (COO) (%)

66

Pioneers of European integration

31.1 per cent) and by Spanish and Germans in France (33.2 per cent and 32.3 per cent). No significant statistical association emerges between the existence of such a network and the gender, age, education and reasons for migration of respondents.

PREVIOUS MOBILITY EXPERIENCES Intra-EU mobility can clearly be considered a form of ‘long-distance’ mobility, in that it at least entails the crossing of a national state border. However, this type of mobility should not be considered as an isolated form, but rather as simply one aspect of people’s varied mobility options throughout their lives (European Foundation for the Improvement of Living and Working Conditions 2006, 7). The idea of taking into account the mobility history of each individual poses the question of whether people with more mobility experiences in the past will be more mobile in the future. Data from Eurobarometer 2005 (EB 64.1) certainly show that people who moved in the past have a higher propensity to move in the near future. This is especially so for people who have made long-distance moves. Plausibly, one migration experience ‘trains’ them to move again in the future. Overall, respondents to the EIMSS are indeed individuals with considerable mobility experiences. Approximately half of them reported that they had lived outside their country of origin for at least three months before resettling in the country where they now live. This may mean having lived already for a short period in the current host country (that is, for a study period) before going back to the home country for some time; or that people have circulated across different countries, having therefore lived in at least three different nations in their lives (in fact, a full 11.7 per cent of the sample). What is the profile of this high-mobility population? Some studies suggest that there is a relationship between age and previous mobility in that the older you are, the more probable it is that you have lived in another country or in another region (Krieger and Fernandez 2006, 6). However, we could not find a significant effect of age on the existence of prior movements in our sample. The same is true of gender. Men are in slightly larger numbers, but this difference is not significant. However, our data do confirm that highly educated persons have a high level of mobility compared to those with low or average levels of education in the EU overall, as Krieger and Fernandez (2006) claim. In fact, 56 per cent of movers with a tertiary level degree have some other migration experiences in their background. Considering all possible combinations of past

Deciding to move

67

Table 3.2 Previous migration experiences and higher education (column %) Higher education

Previous experience Multiple experiences Lived in third country Lived in COR before

Source:

Yes No Yes No Yes No Yes No

No

Yes

40.3 59.7 8.1 91.9 26.4 73.6 22.2 77.8

56.1 43.9 13.3 86.7 43.7 56.3 25.8 74.2

EIMSS, N (minimum) 5 4 838

migrations, education makes a difference (Table 3.2). ‘Super-movers’, as we dub them, are boosted by an edge in human capital, which can manifest itself in a number of different ways – either in terms of labour skills, language proficiency, capacity to retrain themselves, or sheer curiosity and passion for other cultures. The nature of previous experiences of migration can vary enormously. First, we find a type of migration we could call ‘direct’: individuals opt for a given country and settle down there. For instance, the large majority of English and German respondents living in Spain, and English respondents in France did not report previous experiences in other countries. What they have in common is that quality of life was their main reason for moving abroad. This pattern of low previous mobility to third countries can also be found among Italians in Germany (mainly labour migrants), and Spanish citizens living in Italy, who reported family/love reasons as their main motivation. The case of British nationals living in Spain and France, who are mainly low-skilled workers and retirees, is strikingly different from that of the British in Germany and Italy, who in fact report the highest levels of plurinational mobility. The international prevalence of the English language is a plausible explanation for their strong propensity to move and keep moving – after all, language is possibly the tallest cultural barrier inhibiting mobility (European Foundation for the Improvement of Living and Working Conditions 2006, 11). It should be remembered that, generally speaking, in the EU mobility is still largely an intra-national, or at best inter-regional, phenomenon. According to Eurostat and the European Commission (2002, 16), only 5 per cent of EU citizens who, over the last ten years, have changed their

68

Pioneers of European integration

Table 3.3

Relationship between reasons for previous and current migration to COR (%) Reasons for current migration to COR

Reasons for 1st migration to COR

Work Study Quality of life Family/love

Work

Study

Quality of life

Family/ Love

66.0 42.1 19.4 24.0

4.0 20.1 4.4 3.4

13.5 19.6 65.2 16.8

28.9 34.7 23.8 67.8

Notes: Row totals add up > 100 because they refer to the total number of answers, not to respondents. Source:

EIMSS, N 5 1 200

place of residence, have moved to another country within the European Union. Approximately the same percentage has moved to countries outside the EU. Perhaps not surprisingly, EU movers deviate dramatically from this general pattern. In our sample, 61.9 per cent of respondents have had previous mobility experiences in at least another EU15 country, mainly France (12 per cent), Germany (11.3 per cent), and Britain (10.6 per cent). Significant shares of movers have also lived in the United States (15.3 per cent), non-EU15 European countries (14 per cent, of which 3.6 per cent in the A10 countries), and other non-European countries (35.4 per cent).3 ‘Super-movers’ are predominantly work-related movers – both in the past and in the current period of migration. However, mobility driven by education was equally important in the past for them. And, noticeably, former students in a given country tend to move back to that same country at a later stage of their lifecourse. The fact of having studied in a foreign country helps to bring down the cultural barriers associated with learning a new language and new customs, and also helps to strengthen enduring personal networks there. Past experiences as students create stronger roots to a given country than other forms of migration. How do these reasons for mobility evolve over time? Are ‘super-movers’ persistent in their migration decisions? Or do they opt for different countries over time following differing life plans? Our data allow us to answer these questions, distinguishing people who have migrated repeatedly to the same country (Table 3.3) and others who have rather come from a third country (Table 3.4). In a sense, Tables 3.3 and 3.4 constitute ‘mobility tables’ of spatial mobility. The entries on the diagonal correspond to the proportion of

Deciding to move

Table 3.4

69

Relationship between reasons for previous migration to third countries and current migration to COR (%) Reasons for current migration to COR

Reasons for 1st migration to 3rd country

Work Study Quality of life Family/love

Work

Study

Quality of life

Family/ Love

46.3 40.5 25.7 24.7

6.0 19.3 12.2 7.2

30.2 23.9 45.7 25.2

29.4 29.4 30.6 52.7

Notes: Row totals add up > 100 because they refer to the total number of answers, not to respondents. Source:

EIMSS, N 5 1 852

respondents who reported the same motivation in their current and previous migration experiences. We can see at a glance that figures on the diagonal of Table 3.3 are higher than on the diagonal of Table 3.4. This suggests that the reasons behind previous and current migrations are more likely to be similar when both events take place in the same country rather than in third countries. In other words, people who are more open to ‘circulate’ across different countries tend to select their destination on the basis of varying criteria, rather than fine-tuning their initial choice. ‘Study’ is the most versatile motive in facilitating a later mobility experience to the same country: respondents who had already lived in a foreign country for education-related reasons moved back mostly for other reasons, especially professional ones (42.1 per cent), although motives related to family/love (34.7 per cent) or the search for a better quality of life (19.6 per cent) are also significant. Presumably, this mobility path describes the destinies of many Erasmus students who started putting their roots down abroad after their EU-funded exploration time in that country – either for emerging career prospects or for blossoming personal relationships.

CONCLUSIONS Over the last 30 years, three main motivations characterize intra-European mobility: work-related mobility, which dominated the European area after World War II and until the 1970s; migration related to affective reasons that has been present throughout; and the wish to live in a better social and

70

Pioneers of European integration

natural environment, which has grown in importance. Within the EU15, the reasons for setting up residence in another country are not only associated with an economic factor (that is, ‘work’), but also with important non-economic reasons such as ‘family/love’ and ‘quality of life’. Work-related migration has Germany and Britain as its main countries of destination, Italy being the main country of origin. This form of mobility has followed two main routes. On the one hand, migrants have kept moving from southern Europe to Northern and Central European countries. On the other hand, migration flows between Northern-Central European countries have gained momentum. One of the most significant findings of our survey is that ‘family/love’ and ‘quality of life’ are increasingly important motivations of mobility within the EU. With respect to the ‘family/love’ variable, it is worth noting that ‘love’ (that is, usually joining a partner already living in the destination country) is a much more significant reason than ‘family’ (that is, usually following a partner moving to that country) in all the countries we have studied. There are only two exceptions: Spanish nationals residing in Germany, among whom we find a small difference between the two dimensions, and Italian citizens in Spain, for whom ‘family’ is definitely a more important reason than ‘love’. In fact, English and German movers are mostly drawn by ‘quality of life’ wherever they settle down, but particularly in Spain and France. In addition to these heterosexual patterns of migration, our survey also suggested that a remarkable number of gay people have used mobility strategies as part of their lifestyle choices. Finally, reasons related to studying deserve to be mentioned, even if they do not appear to be widespread. The Erasmus programme seems to have had a marginal quantitative importance as the direct reason behind current migration decisions: only 1.1 per cent in our sample settled down for good abroad as an immediate consequence of Erasmus. However, having studied in a given country, even for a few months, facilitates subsequent movements towards the country where it took place, since it apparently tears down (at least in part) the cultural barriers associated with learning a foreign language and new customs, and helps to strengthen social networks which can be re-activated at a later stage in the destination country. Our findings thus show the relevance of previous mobility experiences, as about half of respondents have migrated before settling down where they live at the moment of the interviews. Past reasons for mobility tend to be different from current ones, especially when prior migrations took place in another country. To conclude, the motivations for intra-EU mobility seem to come close to the reasons for changing residence noted in broader surveys on spatial mobility (Eurostat and European Commission 2002), where family/

Deciding to move

71

personal reasons or motives associated with quality of life turned out to be as important as work-related reasons. For citizens who dare cross the frontiers of nation states, moving to another country within the EU is a decision essentially comparable, in terms of motivation, to moving within one’s own country. In this sense, then, in terms of the motivations of the actors involved, EU mobility is being redefined as something very different from classic migration, in which economic rationality prevails. Rather, intra-EU mobility begins to look more like a form of ‘internal’ migration, something which in itself can be seen as a step forward in the process of building an integrated Europe ‘from below’.

NOTES 1. Sections 1 was written by Albert, 2 by Baldoni, and 3 and 4 by Santacreu. 2. In the questionnaire it was possible to mention two possible prevalent reasons for mobility. 3. Since mobility experiences are not necessarily confined to a single country, these percentages add up to more than 100.

4.

The social mobility of mobile Europeans Ettore Recchi

SPATIAL AND SOCIAL MOBILITY: BETWEEN SPECULATION AND EMPIRICAL RESEARCH Social mobility is arguably the major spur for the spatial mobility of human beings. While it is true that some people relocate across borders regardless of occupational or economic reasons (be it to escape a war or retire in a sunny place), the bulk of migrants decide to move in order to improve their position in the social structure. Upward social mobility is the rationale for most migration projects. This chapter examines the intergenerational and intragenerational class mobility of EU movers. In particular, it is meant to outline the opportunities for social mobility in an almost completely open migration regime, such as the one created by free movement policies in the European Union. As Chapter 3 shows, intra-EU migration is driven by a wider canvas of individual strategies than social class advancement. People also move to join partners, follow spouses and enjoy a more peaceful or thrilling life. Yet such motives are no impediment to socioeconomic changes. And this is an even more plausible outcome for those who move in search of more rewarding jobs and careers. At least as a ‘null hypothesis’, the experience of EU movers might be assumed to conform to the baseline immigration story that classically finds a tight association of social and spatial mobility. Eventually, whether they fit in this story or not will help us assess the novelty of their migration pattern. Traditional migrants’ insertion into the social structures of destination countries does not unfold linearly. Many studies indicate that migrants seldom fulfil their aspiration to upward social mobility soon after their arrival. In fact, they tend to be subject to downward trajectories in the early stages of their occupational life in the host society (Piore 1979; Evans and Kelley 1991; McAllister 1995; Bauer and Zimmermann 1999b). This is particularly true during periods of saturation of the receiving labour market (Raijman and Semyonov 1998). The demand for foreign workers 72

The social mobility of mobile Europeans

73

is usually concentrated in so-called ‘3D’ jobs, involving ‘dirty, dangerous and dull’ tasks, that are unlikely to generate upward mobility.1 Still, migrants take these jobs, and are normally gratified by one basic improvement over their experience at home: a higher income. Even if placed at the bottom of the pay ladder, the average migrant experiences a significant increase in income due to the earnings differentials between origin and destination country. Moreover, the migrant career achievements are expected to be U-shaped, progressing with the passing of time in the host society (Chiswick 1979). To a certain extent, this expectation compensates for the possible decline of occupational status in the early stages of immigration. Nonetheless, downward status mobility is something many migrants have to cope with.2 In spite of all strategies for handling the status inconsistency that so frequently marks migrants’ lives, the aspiration to full-fledged social mobility – entailing an ascent in both income and status – remains at the heart of the immigration experience. But, while it is sometimes noted that migrants are more occupationally mobile than natives (for example, Bauer and Zimmermann 1999b), there is surprisingly little systematic research on the intergenerational class mobility of migrants (for a significant exception, see Yaish 2002). It is usually assumed that the ‘social mobility dream’ tends to be fulfilled by second and third generations – that is, by means of postponement or projection on to the children and grandchildren of first movers (Gordon 1971). A ‘peddler to plumber to professional’ upward trajectory is idealized thereafter (Perlmann and Waldinger 1997, 894). However, even in the US and Canada, the prototypical countries of social mobility opportunities for migrants, fine-grained evidence on the mobility patterns that substantiate such a linear intergenerational assimilation of migrant minorities in the social structure is quite scant (Boyd and Grieco 1998; Portes and Rumbaut 2005). In Europe, so far, information is limited to a dated German longitudinal study (Seifert 1997), a comparison of the early labour market experiences of young Turks and Moroccans in six countries (Cruel and Vermeulen 2003), an inquiry into the occupational achievements of Turkish, Moroccan and Portuguese second generations in France (Simon 2003), and a more recent analysis of British longitudinal data showing sharp differences of career trajectories by ethnic group (Platt 2005). Overall, however, speculation seems to prevail over evidence. For sure, no firm generalization has been made on the basis of comparative empirical analyses of migrant populations in European societies. In fact, it is not the lack of scholarly interest but the limits of existing surveys of the general population in most countries that explain the paucity of quantitative-oriented research focused on migrant social mobility. In nations where immigrants amount to less than 10 per cent of residents, their

74

Pioneers of European integration

number simply ends up being too small in average-size random samples. Not by chance, the more detailed studies in the field so far have been conducted in Israel, a country where over half the Jewish population was born abroad (Raijman and Semyonov 1998, 292). This permits the use of microcensus data for significant comparisons of natives and immigrants. An alternative strategy is to carry out surveys of immigrants parallel to surveys of the general population such as, for instance, the Multicultural Australia Survey of 1988–89 (McAllister 1995). This is exactly the strategy adopted by our European Internal Movers Social Survey (EIMSS). This chapter tries to fill this gap for that specific share of the mobile population – EU movers – who are simultaneously migrants and European nationals. The key question of the analysis is: does EU citizenship status make movers’ social mobility patterns akin to stayers’? The next section discusses hypotheses about immigrant mobility in such a social and institutional context. In section 3, the class compositions of EU stayers and movers are contrasted. The final two sections of the chapter illustrate and discuss findings on intergenerational and intragenerational mobility respectively.3

EASIER MIGRATION, EQUALIZED OPPORTUNITIES? THE IMPACT OF THE EU FREE MOVEMENT REGIME ON INTRA-EU MIGRANTS’ SOCIAL MOBILITY Immigrants’ face many disadvantages in their host societies. Economic theory imputes immigrants’ disadvantages to their normally worse command of country-specific labour skills (like language and social norms), while sociological perspectives stress discrimination in the access to jobs, housing and public services (in particular, due to institutional discrimination such as exclusion from citizenship rights). The two views are not mutually exclusive. Lack of market-valued resources (relative to natives) and discrimination can definitely combine to impair the social and economic achievements of immigrants. In an ideal world where both factors could be removed, differences in the chances of upward social mobility between immigrants and natives should disappear altogether. Is this the case in the EU? How does free movement designed by the European Union for the cross-border mobility of citizens of its member states – that is, a quite special migration regime – affect the social mobility regime of intra-EU migrants? Does being a EU citizen with the same rights as natives cancel out the impediments to upward social mobility usually suffered by immigrants? Some hypotheses can be formulated.

The social mobility of mobile Europeans

75

H1. As the elimination of barriers to mobility and settlement for EU citizens within the Union has made migration costs negligible (compared to the costs of resettlement of other world migrants), intra-EU migration has become more accessible to all social strata. While it is often found that migrants disproportionately originate from the middle class, who can afford the travel and opportunity costs of international resettlement (for example, Yaish 2002, 456; Fielding 2007, 123), in principle the lowering of migration costs for EU citizens should encourage mobility from one country to another irrespective of class origins. Therefore, H1 posits that the class origins of EU movers and EU stayers do not differ significantly. H2. Since discrimination against EU migrants is almost non-existent – institutionally but also culturally, as social distance among Europeans has largely waned (see Díez Medrano 2003) – EU movers should not face ‘glass ceilings’ and should enjoy career prospects similar to stayers. At the same time, since wage differentials between EU15 member states tend to level out4, workers’ migration should be driven by opportunities for upward class mobility rather than by the allure of higher salaries in status-degrading jobs (as is in fact typical of migrants from less-developed countries). In other words, it seems plausible that EU movers might replicate the kind of ‘spiralist’ trajectories of spatial mobility aimed at class mobility that are found in studies on internal migration (for example, in Britain, Fielding 1992 and 1995; the idea of ‘spiralism’ dates back to Watson 1964). New European movers might use their unusual opportunities for international movement to out-flank social mobility regimes that remain limiting at a national level. Therefore, H2 posits that EU movers are more frequently found in upper class destinations than EU stayers. H3. As cross-border mobility nonetheless remains an investment-like business, implying the endorsement of stronger risks in the labour market than in home countries, it can also be posited that, in spite of the decrease in the costs of intra-European migration, EU movers are more likely than stayers to change social class over their lifecourse. Thus, H3 posits that EU movers are more socially mobile than stayers – both upwardly and (possibly) downwardly. H4. Opportunities for absolute social mobility are contingent on the economic structure of the countries where people live. Therefore, individuals moving from relatively backward (that is, with a higher number of blue-collar jobs) to more service-based economies are more likely to experience upward social mobility. In the geographical area under scrutiny, France, Germany and Britain are known to have a larger proportion of the workforce in white-collar occupations

76

Pioneers of European integration

than Italy and Spain (Breen and Luijkx 2004a, 74). Hence, H4 posits that Italian and Spanish migrants have higher rates of upward social mobility than their French, German and British counterparts. The analyses that follow will determine the extent to which these ex ante suppositions describe the real trajectories of class mobility of intra-EU migrants. The dataset used in this chapter merges EIMSS and ESS for Germany, France, Britain, Italy and Spain. However, as the focus of this chapter is on social mobility achieved through occupations in a foreign country, EU movers without any job experience in the host country – such as students, non-working spouses and pensioners – are left out of the analysis, except when otherwise noted.

THE CLASS COMPOSITION OF STAYERS AND MOVERS Tables 4.1 and 4.2 describe the class positions of EU stayers and movers by country of residence (COR) and country of origin (COO), allowing for a first test of the hypotheses about the social origins and destinations of our sub-populations. Even at a rapid glance, Table 4.1 reveals the markedly more privileged family background of intra-EU migrants. In all countries, the proportion of movers from working class families is lower than that of natives of the host country, with the only exception being Italians and Spanish in Germany. On the other hand, more than one-quarter of respondents among movers to the country, but only one out of six among natives, originate from the bourgeoisie or the salariat (classes I–II of the EriksonGoldthorpe schema). Again, the Italians and Spanish in Germany (the latter also in Britain) are an exception. Nonetheless, in all the communities of Spanish abroad, and in four out of five of those of Italians, the proportion of movers with upper class origins is higher than among Spanish and Italian stayers respectively. In sum, the experience of resettling across EU borders is more frequent among the children of better-off families. It might be suspected that a spurious correlation with age intervenes here, given the lower age profile of working movers (on average they are 46, while stayers are 50). However, the class origins of natives and movers below and above 45 in fact reproduce approximately the same pattern (analysis not shown). Thus, intra-EU migration seems far from being uncorrelated with social origins, and appears to be skewed towards the middle–upper class. At the country level, the overall pattern of movers’ privilege can be found among EU movers residing in Britain, France, Spain and Italy. In the last country, the difference with the native population is at its apex

The social mobility of mobile Europeans

Table 4.1

77

Class origins of EU stayers and movers by country of residence and nationality (column %) Country of residence Nationality

Class of origin Bourgeoisie (I–II) Routine non-manual (III) Petty bourgeoisie (IV) High-skilled manual (V–VI) Low/non-skilled manual (VII)

Nationality Class of origin Bourgeoisie (I–II) Routine non-manual (III) Petty bourgeoisie (IV) High-skilled manual (V–VI) Low/non-skilled manual (VII)

Nationality Class of origin Bourgeoisie (I–II) Routine non-manual (III) Petty bourgeoisie (IV) High-skilled manual (V–VI) Low/non-skilled manual (VII)

Nationality Class of origin Bourgeoisie (I–II) Routine non-manual (III) Petty bourgeoisie (IV) High-skilled manual (V–VI) Low/non-skilled manual (VII)

Nationality Class of origin Bourgeoisie (I–II) Routine non-manual (III)

Germany FR IT

DE

GB

19.0 26.7 12.4 20.7 21.2

26.0 28.8 12.0 13.5 19.7

33.2 28.2 13.4 13.9 11.4

DE

GB

Britain FR

52.7 22.3 9.8 7.1 8.0

20.8 16.5 14.3 20.4 28.1

35.6 23.1 20.7 12.0 8.7

DE

GB

France FR

30.2 28.1 19.8 13.5 8.3

34.7 23.8 14.9 12.9 13.9

16.1 14.3 20.4 14.3 35.0

DE

GB

Italy FR

34.0 31.9 12.1 15.6 6.4

40.7 27.3 16.3 8.6 7.2

37.4 24.0 18.4 3.9 16.2

DE

GB

Spain FR

17.0 34.0

23.7 22.0

25.9 30.9

ES

3.3 5.2 31.0 4.8 55.7

13.4 8.8 32.4 10.6 34.7

IT

ES

28.5 11.3 31.2 7.2 21.7

18.4 13.5 27.0 8.5 32.6

IT

ES

22.0 10.4 28.6 9.3 29.7

22.9 13.1 24.6 11.4 28.0

IT

ES

8.3 15.5 35.6 9.2 31.4

22.5 17.5 34.3 8.1 17.5

IT

ES

20.4 12.4

8.3 8.8

78

Pioneers of European integration

Table 4.1

(continued) Country of residence

Petty bourgeoisie (IV) High-skilled manual (V–VI) Low/non-skilled manual (VII)

21.3 14.9 12.8

30.5 10.2 13.6

18.5 12.3 12.3

29.9 10.2 27.0

29.7 9.6 43.6

Notes: ‘Stayers’ data in italics. Source:

ESS 2004 and EIMSS, N 5 11 727

– more than 60 per cent of movers stem from classes I, II and III, whereas less than one quarter of Italians have a similar origin. Although movers in the EIMSS sample only include people who work or have worked in the host society, leaving out Goethe- or Stendhal-likes living for extended periods south of the Alps, EU residents in Italy seem to embody an elitebased immigration.5 Yet they are not as distant from the native population as EU movers in Spain. In particular, almost 35 per cent of the European migrants working in Spain are self-employed or owners of small businesses – some of them possibly trying to combine work and sunshine on the basis of prior professional experiences and savings. The situation is reversed in Germany, where there is a much higher proportion of movers with working class roots. Germany still qualifies as a country of traditional migration, offering economically rewarding opportunities to lowerclass migrants from the rest of Europe. This is in stark contrast also with Britain, which attracts a more select population of intra-EU migrants – a higher number originating from the upper class, as well as with higher educational credentials (see Recchi et al. 2006). In turn, the British economy seems to reward migrants from the other side of the Eurotunnel more than any other – to the point that 45 per cent of movers to Britain end up in the salariat. Since the bulk of European migrants in Britain live in the capital,6 Fielding’s (1992) argument about the London area as an ‘escalator region’ is corroborated on an international scale. But it is not only in Britain that privilege in class origins carries over to class destinations. Migration does not thwart the access of mobile Europeans to the higher ends of the social ladder. EU movers are overrepresented in both the salariat and the self-employed middle class (Table 4.2). The only exceptions to this picture are Italian and Spanish migrants in Germany and France, who are more likely than natives to take working class jobs (and are also less likely to get middle–upper class occupations in Germany). This is the outcome of two different processes: the recruitment of Southern European workers in industrial occupations (diminished but

The social mobility of mobile Europeans

Table 4.2

79

Class destinations of EU stayers and movers by country of residence and nationality (column %) Country of residence Nationality

Class of destination Bourgeoisie (I–II) Routine non-manual (III) Petty bourgeoisie (IV) High-skilled manual (V–VI) Low/non-skilled manual (VII)

Nationality Class of destination Bourgeoisie (I–II) Routine non-manual (III) Petty bourgeoisie (IV) High-skilled manual (V–VI) Low/non-skilled manual (VII)

Nationality Class of destination Bourgeoisie (I–II) Routine non-manual (III) Petty bourgeoisie (IV) High-skilled manual (V–VI) Low/non-skilled manual (VII)

Nationality Class of destination Bourgeoisie (I–II) Routine non-manual (III) Petty bourgeoisie (IV) High-skilled manual (V–VI) Low/non-skilled manual (VII)

Nationality Class of destination Bourgeoisie (I–II) Routine non-manual (III)

Germany FR IT

DE

GB

27.5 27.1 8.4 17.0 19.9

42.3 19.1 12.7 9.5 16.4

40.7 32.5 9.1 8.1 9.6

DE

GB

Britain FR

65.5 16.1 10.8 4.0 3.6

29.2 28.2 8.7 9.1 24.7

43.7 35.2 13.6 5.2 2.3

DE

GB

France FR

52.6 25.6 9.0 9.0 3.8

53.1 16.3 13.3 5.1 12.2

28.0 23.2 9.1 21.9 17.8

DE

GB

Italy FR

44.3 32.8 16.4 2.5 4.1

49.5 17.3 28.4 2.4 2.4

44.9 24.0 15.0 3.0 13.2

DE

GB

Spain FR

35.7 19.6

31.7 30.0

38.6 22.9

ES

3.5 11.1 12.4 17.7 55.3

23.9 15.7 7.4 12.6 40.4

IT

ES

35.2 23.2 13.0 7.0 21.7

32.2 20.3 14.7 14.0 18.9

IT

ES

34.9 18.3 13.1 12.0 21.7

37.7 20.5 7.3 9.3 25.2

IT

ES

18.4 20.9 25.0 9.0 26.6

36.0 22.8 16.2 6.6 18.4

IT

ES

22.2 19.3

20.6 11.1

80

Pioneers of European integration

Table 4.2

(continued) Country of residence

Petty bourgeoisie (IV) High-skilled manual (V–VI) Low/non-skilled manual (VII) Notes: Source:

30.4 5.4 8.9

23.3 6.7 8.3

25.3 8.4 4.8

37.8 5.2 15.6

16.3 13.4 38.5

‘Stayers’ data in italics. ESS 2004 and EIMSS, N 5 11 095

not terminated after 1973), and the prospering of ethnic niches, attracting a job-specific workforce of co-nationals (such as Italian restaurant and ice-cream businesses).7 Intra-EU migrants from Southern Europe are found in all social classes – in greater numbers in the working class among movers of the 1970s and 1980s, more in the middle-upper classes subsequently. In contrast, not only retirement migrants, but also workers heading to the south of the continent constitute a select population among which the highly skilled are overrepresented. In particular, German citizens on the move to other EU countries get predominantly upper-class employment. More than 50 per cent of German respondents to the EIMSS belong to the salariat, while less than 5 per cent have a non-qualified manual job. The class profile of British and French movers is similar, although not so skewed towards the upper end as that of Germans.

INTERGENERATIONAL MOBILITY: CROSSING STATE BORDERS, REPRODUCING CLASS BOUNDARIES? From this section on, I shift to an individual-based analysis of social mobility, decomposing absolute intergenerational mobility into different typical paths. In particular, four basic trajectories are distinguished: (1) Immobility: individuals who belong to the same social class as their dominant parent (that is, the parent in the highest social class); (2) Upward mobility: individuals who have moved hierarchically between social classes, from a lower to a higher class; (3) Downward mobility: individuals who have moved hierarchically between social classes, from a higher to a lower class; (4) Non-vertical mobility: individuals who have moved across classes but not hierarchically (Table 4.3).8 The results of this recode are presented in Figure 4.1, which compares

The social mobility of mobile Europeans

Table 4.3

Social mobility outcomes in a 5 3 5 mobility table

Origin

I–II III IV V–VI VII Notes: Source:

81

Destination I–II

III

IV

V–VI

VII

Imm Up Up Up Up

Down Imm Nonvert Nonvert Up

Down Nonvert Imm Nonvert Up

Down Nonvert Nonvert Imm Up

Down Down Down Down Imm

Imm 5 Immobile; Nonvert 5 Non-vertical Recode based on Erikson and Goldthorpe (1992, 124 and sparsim)

the proportion of intergenerationally mobile respondents (of the different kind) between subpopulations. A strong similarity between stayers and movers emerge. In both samples, the intergenerationally mobile amount to two-thirds of the total – a result very close to what is found in existing research on social mobility in our reference countries (Breen and Luijkx 2004a, 48). Among these mobile individuals, as with stayers, slightly less than half have gone up the social ladder, while about 30 per cent have gone down and 20 per cent have experienced non-vertical class mobility. This finding counters one of our initial expectations: in an intergenerational perspective, movers are not more socially mobile than stayers. At the country level, Spain appears to be the only labour market in which European working migrants enjoy a significantly higher mobility rate than natives. However, this is almost exclusively due to a higher rate of nonvertical mobility (in particular, 28.9 per cent of working Germans in Spain experience this form of movement in the social structure). Differences in such short-haul mobility are also the cause of the especially high intergenerational mobility rates of the Spanish abroad (but noticeably in Britain 43.7 per cent of them are upwardly mobile) and Germans at home. Overall, no consistent difference between movers and stayers shows up, suggesting only a loose, if any, relation between spatial and social mobility. Still, the possibility that this finding is spurious, as other exogenous factors mask the real impact of migration on intergenerational mobility, cannot be ruled out by the analysis undertaken so far. The similarity of the mobility paths of movers and stayers could be only apparent and in fact stem from their differing biographical profiles. A logistic regression of intergenerational class mobility (dichotomized) was therefore run to control for other likely determinants of social mobility such as age, gender, education, and class origin (Table 4.4). Two models are compared,

82

FR IN DE

GB IN DE

IT IN DE

ES IN DE

DE IN FR

FR IN FR

GB IN FR IT IN FR

ES IN FR

DE IN GB

FR IN GB

GB IN GB

IT IN GB

ES IN GB

DE IN IT

FR IN IT

GB IN IT

IT IN IT

ES IN IT

DE IN ES

FR IN ES

GB IN ES

IT IN ES

ES IN ES

Total

Immobile

NV mobile

UP mobile

DN mobile

Mobility rates decomposed into upward (UP), downward (DN) and non-vertical (NV) mobility by nationality and country of residence

ESS 2004 and EIMSS, N5 10 463

DE IN DE

Figure 4.1

Source:

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

The social mobility of mobile Europeans

Table 4.4

83

Logit models predicting intergenerational class mobility in the pooled ESS-EIMSS dataset Model 1 B

Age Age squared Gender Male Female Education University degree University entry qualification Upper vocational secondary Primary/lower secondary Nationality German French British Italian Spanish Country of residence Germany France Britain Italy Spain Migration status Mover Stayer Class of origin Bourgeoisie (I–II) Routine non-manual (III) Petty bourgeoisie (IV) High-skilled manual (V–VI) Low/non-skilled manual (VII) Class of origin by migration status (mover) Bourgeoisie (I–II) Routine non-manual (III) Petty bourgeoisie (IV) High-skilled manual (V–VI) Low/non-skilled manual (VII)

Model 2

Std Error

B

Std Error

20.02** 0.00

0.00 0.00

20.02** 0.00

0.00 0.00

20.09* –

0.04 –

20.09* –

0.05 –

0.80** 1.07** 0.47** –

0.09 0.10 0.08 –

0.81** 1.08** 0.49** –

0.10 0.10 0.08 –

0.11 0.16 0.34 0.05 –

0.23 0.20 0.20 0.19 –

0.18 0.22 0.43* 0.08 –

0.25 0.23 0.22 0.22 –

0.33 0.34 0.11 0.06 –

0.21 0.23 0.22 0.21 –

0.26 0.25 20.01 0.08 –

0.24 0.26 0.23 0.22 –

20.16 –

0.16 –

20.28 –

0.24 –

21.09** 20.01 0.71** 0.71** –

0.08 0.07 0.07 0.08 –

21.00** 20.07 0.61** 0.66** –

0.08 0.08 0.08 0.09 –

– – – – –

– – – – –

20.43* 0.50* 0.88** 0.60 –

0.21 0.25 0.25 0.33 –

84

Pioneers of European integration

Table 4.4

(continued) Model 1 B

Constant 22 Log Likelihood Degrees of freedom Pseudo R-square

0.45 9 850 19 0.12

Std Error 0.22

Model 2 B 0.45 9 814 23 0.12

Std Error 0.22

Notes: ESS: only respondents with work experience; EIMSS: only respondents with work experiences in country of residence. Reference category for dependent variable: class immobility. * Significant at 5%; ** Significant at 1%. Immobility 5 0; Mobility 5 1. Source:

ESS 2004 and EIMSS, N 5 8 426

the second controlling also for the interaction of class origin and migration status. In both models, the migration status of respondents (that is, being movers or stayers) does not affect respondents’ odds of intergenerational social mobility significantly. In the pooled ESS-EIMSS dataset, as could be expected, education, and especially upper-secondary qualifications, enhance the probabilities of social mobility considerably. In fact, country contexts and nationalities make no big difference, although British citizens tend to be slightly more socially mobile than their continental counterparts. Moreover, younger respondents and women have higher chances of absolute intergenerational mobility – including downwards, which is relatively more frequent for these social categories. Finally, and most importantly, odds of intergenerational mobility are contingent on class origins, being higher for lower middle class offspring. In particular, sons and daughters of the petty bourgeois self-employed and high-skilled manual workers qualify as the most likely candidates for intergenerational mobility, while people from the top of the class hierarchy are in fact more often bound to stay put in their class of origin. By introducing an interaction term of class and migration status, model 2 intends to control whether the effect of family class on the odds of intergenerational mobility varies for EU movers. While the main effect does not cease to be significant, the interaction term shows that the odds of being intergenerationally immobile are even higher among EU movers from the upper class than among their stayer counterparts: in the EIMSS, 59.1 per cent of them are immobile as opposed to 48.7 per cent in the ESS. Equally, the odds of intergenerational mobility are higher than average for individuals from classes V–VI, but even more so for EU movers with such

The social mobility of mobile Europeans

85

a family background – in the EIMSS 89.1 per cent of them are socially mobile as opposed to 80.7 per cent in the ESS. This means that crossnational migration in the EU, far from levelling out the odds of social mobility of individuals from different backgrounds, in fact amplifies existing, class-based divergences. The findings suggest that the free movement regime of the EU seems to offer additional opportunities of social closure for the European upper class, as well as extra chances of (mostly upward) mobility for children of the lower-middle class. At the start of this chapter it was argued that EU movers from Italy and Spain could be more intergenerationally mobile than their counterparts from Central-Northern Europe, since the social structures of their countries of origin one generation ago had a larger number of blue-collar workers (including peasants) than those of Germany, France and Britain. Moreover Germany, France and Britain still have a much higher number of jobs in the salariat. Therefore, the absolute intergenerational mobility for Italian and Spanish migrants should be boosted by both the backwardness of parental occupations when respondents were adolescents and today’s more upgraded occupational structure of the host countries. This hypothesis is tested in the logistic regressions of the odds of intergenerational mobility (and more precisely of upward, downward and nonvertical mobility) on a set of independent variables, including nationality and country of residence, in Table 4.5 (analysis limited to EU movers, both workers and non-workers). In fact, Italian and Spanish citizens living abroad in the EU are not significantly more mobile than the other nationalities of movers considered. However, a ‘Spain’ effect is worth mentioning. Spain is significantly more likely to host socially mobile European migrants, especially compared to Germany. In particular, the chances of non-vertical mobility are significantly higher for EU citizens who settle in Spain than anywhere else. Shifts to self-employment (class IV) are noticeably frequent. This social class is home to 31.4 per cent of working EU movers interviewed in Spain, some of them being semi-retirees who take (or keep) some part-time job during el buen retiro under the Spanish sun. Far from being a means to improve individual economic conditions, intra-European spatial mobility is perhaps often a benefit in itself – or a side-effect of an already ‘golden’ career (like, typically, for expats and their families: see Wagner 1998). Among all other predictors included in the model, the impact of education is strong and in line with what could be expected: the lowest level of credentials puts at stake the chances of upward mobility and heightens tremendously the risk of downward mobility among EU movers just as it does in the rest of the population. The effect of class origin is highly significant as well. For all EU movers, regardless of their working status,

86 20.00 0.16 0.19 20.16 –

0.54** 0.71** 0.27 – 0.13 0.13 0.12 0.12 –

0.15 0.16 0.15 –

0.05 –

0.23 20.06 0.14 20.05 –

1.93** 1.18** 0.56** –

0.44** –

20.41** 20.17 –

0.12 0.10 –

20.17 20.10 – 20.12* –

0.06* 0.00

B

0.02 0.00

Std error

0.14 0.13 0.13 0.13 –

0.17 0.17 0.17 –

0.08 –

0.12 0.10 –

0.02 0.00

Std error

Upward vs others

20.02 0.00

B

Mobility vs immobility

20.50** 20.09 20.14 0.38* –

21.92** 21.21** 20.95** –

20.33** –

0.18 0.09 –

20.04 0.00

B

0.16 0.14 0.14 0.15 –

0.18 0.18 0.19 –

0.09 –

0.14 0.12 –

0.02 0.00

Std error

Downward vs others

20.06 0.03 20.05 20.20 –

20.39 0.07 0.22 –

20.37** –

0.02 20.18 –

20.05 0.00

B

0.17 0.16 0.16 0.17 –

0.20 0.20 0.20 –

0.10 –

0.15 0.13 –

0.03 0.00

Std error

Non-vertical vs others

Logit models predicting intergenerational class mobility in the EIMSS dataset (only EU movers)

Age Age squared Period of migration 1974–83 1984–93 1994–2003 Gender Male Female Education University degree University entry qualification Upper vocational secondary Primary/lower secondary Nationality German French British Italian Spanish

Table 4.5

87

* Significant at 5%; ** Significant at 1%

0.02 0.25 0.30* 0.11 – 0.07* 0.11 – – – – – – 23.03 3 693 18 0.17

0.13 0.13 0.14 0.14 – 0.03 0.11 – 0.11 0.12 0.12 0.16 – 0.57

– – – – – 0.60

0.11 –

0.14 0.13 0.15 0.14 – 0.03

– – – – – 1.17 3 191 18 0.13

20.21 –

0.26 20.05 0.15 0.22 – 20.08*

– – – – – 0.64

0.13 –

0.16 0.16 0.16 0.16 – 0.04

– – – – – 0.89 2 482 18 0.04

20.07 –

20.48** 20.38* 20.33 20.45** – 20.04

– – – – – 0.72

0.14 –

0.17 0.16 0.17 0.16 – 0.04

Source: EIMSS, N5 4017 (model 1); 2963 (model 2, excluding respondents from class I–II); 3170 (model 3, excluding respondents from class VII); 2113 (model 4, excluding respondents from classes I–II and VII)

Notes:

Country of residence Germany 20.42** France 20.24 Britain 20.12 Italy 20.18 Spain – Knowledge COR lang. at migration 20.05 Working status Never worked in COR 0.03 Works/worked in COR – Class of origin Bourgeoisie (I–II) 21.57** Routine non-manual (III) 20.29* Petty bourgeoisie (IV) 0.73** High-skilled manual (V–VI) 0.72** Low/non-skilled manual (VII) – Constant 1.34 22 Log Likelihood 4 423 Degrees of freedom 22 Pseudo R-square 0.19

88

Pioneers of European integration

the probability of changing class as adults depends heavily on the class in which they were born. Children of the service class cross national borders almost only if they can preserve their original class location, while everybody else moves spatially with a view to some status upgrading. However, Table 4.5 (first column) indicates that the capacity of upper class movers to be intergenerationally immobile is considerably higher than that of migrants from all other social classes to be mobile.9 Perhaps intra-EU migration projects are more easily accessible to individuals from the upper class who are bound to stay in that same class than to others aiming at a class upgrade by resettling abroad. On the other hand, it is true that more recent movers have higher probabilities of upward mobility. This may well be a sign of growing equalization of opportunities in a time perspective. However, a significant gender gap shows up, with men enjoying 1.5 times higher odds of moving upwardly in the social ladder than women, who are, in fact, more commonly found among intra-EU migrants who experience non-vertical intergenerational mobility in the social structure.

INTRAGENERATIONAL MOBILITY: MAKING A CAREER IN ANOTHER COUNTRY? Among scholars of social stratification, intergenerational mobility is more widely studied than career mobility (Morgan 2006, 7).10 Using crosssectional data, the only viable analytical strategy consists in identifying individuals’ work positions at key stages of their life course. In our case, three different points of migrants’ work histories are held to be critical to their intragenerational mobility: the last job held in the country of origin (Coo0), the first job in the country of residence (Cor1) and the final job in that same country (Cor2).11 The following possible combinations of work trajectories can be found: a. Job immobility (Coo0 5 Cor1 5 Cor2); b. Job mobility at migration (Coo0 ≠ Cor1 5 Cor2); c. Job mobility after migration (Coo0 5 Cor1 ≠ Cor2); d. Recurrent job mobility (Coo0 ≠ Cor1 ≠ Cor2). In our sample, the two extreme cases (a and d) are relatively more frequent, accounting for almost two-thirds of work trajectories: 31.7 per cent of respondents who have worked in both the sending and the receiving country are occupationally immobile migrants, while 31.9 per cent have changed job upon their arrival in the country of current residence and also at a later stage in that same country. In fact, 23 per cent of respondents

The social mobility of mobile Europeans

89

changed job only at the first one of the two transitions considered (that is, when resettling abroad), whereas 13.4 per cent changed only at the second possible transition. These options are contingent on a number of different factors operating at different levels (Table 4.6). At the macro level, history shapes opportunities and work trajectories markedly. Intra-EU migrants who left their home country between 1974 and 1993 are definitely more likely to have experienced some job mobility than movers of the last period considered. Of course, they have also had a longer time to ‘shop’ for job switches, but these movers also have significantly higher probabilities of having changed occupation at their first employment in the host country (that is, of falling into the ‘always mobile’ category). Contexts count, too. Job mobility after migration is overall more frequent for movers to Central-Northern European countries. Germany and Britain also offer significantly higher opportunities for changing jobs at both transitions. Of the individual characteristics studied, gender and education can be seen to play a relevant role. Women change job at migration, and possibly also at a later stage, much more commonly than men. This might reflect the dependency of their migration choice on their male partners’ work choices – a ‘tied-mover effect’ which emerged in a Dutch study on internal migration (Van Ham 2001). In other words, women who migrate following a first-earner partner are likely to be less demanding of the labour market and thus be willing to leave their former occupation. But a higher propensity to face migration as an international career detour – that is, to start a new career – on the part of women cannot be ruled out on the basis of this evidence. What is for sure, is that women associate geographical mobility with job mobility more than men in the EU. A similar interpretative conundrum about the nature of career changes at migration – seen as outcomes of involuntary constraints or first-order preferences – emerges when considering the influence of education. Movers with secondary school credentials (or higher) are less likely to change job at and after migration. This is all the more so for EU movers who had a service-class job or were self-employed in their country of origin. Occupational mobility tends rather to be the story of the labour market integration of lowskilled European migrants. Thus, stronger socioeconomic resources (like gender, education and prior class positions) seem to be used to safeguard occupational stability when moving across national borders in Europe. In other words, it turns out to be the weakest workers who change employment as a consequence, or simply after, relocating in another country. Nationality does not make a real difference. But language knowledge does. Interestingly, not proficiency at migration but current fluency in the host country language is a better predictor of job mobility after migration

90 20.11 – 20.08 20.09 20.17 – 0.42 0.23 0.17 20.05 –

0.14 – 0.31 0.31 0.30 – 0.24 0.21 0.20 0.20 –

20.36 20.63* 20.47 – 20.15 20.25 20.03 20.31 –

1.07** 0.56** –

0.21 0.16 –

20.11 0.03 – 20.43** –

0.03 0.00

B

0.05 0.00

Std Error

0.28 0.26 0.25 0.24 –

0.37 0.37 0.35 –

0.17 –

0.26 0.21 –

0.06 0.00

Std Error

Mobile after migration

0.01 0.00

B

Mobile at migration

Multinomial logit predicting trajectories of job mobility among EU movers

Age Age squared Period of migration 1974–83 1984–93 1994–2003 Gender Male Female Education University degree University entry qualification Upper vocational secondary Primary/lower secondary Nationality German French British Italian Spanish

Table 4.6

20.03 0.13 20.07 20.15 –

20.45 20.48* 20.55 –

20.63** –

0.83** 0.51** –

20.06 0.00

B

0.23 0.20 0.20 0.19 –

0.30 0.29 0.28 –

0.10 –

0.21 0.17 –

0.04 0.00

Std Error

Always mobile

91

EIMSS, N 5 1 956

0.27 0.27 0.40 0.30 – 1.59

Source:

20.96** 20.41 20.99* 20.64* – 21.02

0.24 0.23 0.32 0.25 – 1.21

21.10** 20.43 20.83** 20.38 – 1.61 5 198 66 0.16

0.34 0.34 0.33 0.35 – 0.07 0.12

Reference category for dependent variable: job immobility. * Significant at 5%; * Significant at 1%.

0.67* 0.70* 0.70* 0.12 – 20.05 0.38**

0.24 0.25 0.24 0.24 – 0.06 0.09

20.23 20.42* 20.12 20.22 – 20.01 0.19*

Notes:

Country of residence Germany France Britain Italy Spain Knowledge of COR language at migration Knowledge of COR language at interview Class in COO Bourgeoisie (I–II) Routine non-manual (III) Petty bourgeoisie (IV) High-skilled manual (V–VI) Low/non-skilled manual (VII) Constant 22 Log Likelihood Degrees of freedom Pseudo R-square 21.24** 20.23 20.92** 20.31 – 3.13**

0.72** 0.30 0.65** 20.04 – 20.07 0.63**

0.22 0.22 0.31 0.24 – 1.14

0.25 0.26 0.25 0.26 – 0.05 0.10

92

Pioneers of European integration

Table 4.7

Paths of intragenerational class mobility of EU movers (%) Transition 2

Transition 1

Source:

Immobile NV mobile UP mobile DOWN mobile Total

Immobile

NV mobile

UP mobile

62.1 4.2 6.7 7.7 80.7

1.8 0.9 0.5 0.4 3.6

5.8 0.6 0.2 6.1 12.7

DOWN Total mobile 1.7 0.1 1.1 0.1 3.0

71.3 5.8 8.5 14.3 100.0

EIMSS, N 5 2 180

and both at and after migration. In fact the opposite cause–effect relation seems more plausible: job mobility trains language skills. Perhaps the propensity of the highly educated movers to be occupationally immobile can lead them to live more secluded lives in expat circles, while less-educated movers are forced to improve their knowledge of the host country language as they transfer from one workplace to another. Job-to-job mobility forms the basis of movements between social classes – inasmuch as occupations are taken as proxies of class belonging. Of course, job mobility does not automatically lead to shifts from one class to another, as each class comprises several different occupations. Therefore, intragenerational class mobility rates are inherently lower than job mobility rates in any given population (Mayer and Carroll 1987). Indeed, 71.3 per cent of respondents to the EIMSS did not change social class when taking up their first job after migration. Moreover, 80.7 per cent held the same class position in the transition between first and current job in the host country. As might be expected, the work-with-migration transition can bring about downward mobility (this is the case for 14.3 per cent of respondents, while 8.5 per cent are in fact upwardly mobile), but the subsequent career in the host country is more likely to be on the upside (12.7 per cent) than on the downside (3.0 per cent). A cross-tabulation of class mobility patterns at each transition (that is, Coo0 to Cor1 and Cor1 to Cor2) helps describe all the possible career paths across these three points of intra-EU migrants’ work histories (Table 4.7). Some of these intragenerational mobility patterns are quite unusual. In particular, only two respondents (less than 0.1 per cent) experienced downward mobility both at migration and in the following step of their career. But an extremely successful career (upward class mobility at both transitions) is equally exceptional (five cases, 0.2 per cent of the sample).

The social mobility of mobile Europeans

93

Class immobility at both transitions, regardless of migration and possible subsequent career moves, is overwhelmingly the rule, reported by 62.1 per cent of respondents. For the sake of simplicity, these 16 possible combinations of transitions are grouped into two broad typical class trajectories: immobile and mobile paths (respectively, pale grey and dark grey cells in Table 4.7). The effects of the usual series of contextual and individual characteristics on the log odds of being intragenerationally mobile rather than immobile are estimated in Table 4.8. The findings are not dramatically different from those concerning job mobility. Again, time and space effects are noteworthy. Migrants of the earlier periods (that is, those who moved before the introduction of European citizenship in 1993) are more likely to have experienced some career mobility. Where individuals move to affect the chances of facing intragenerational class mobility as well, with Britain and Italy exposing them to higher ‘risks’ of class shifts, perhaps for opposite reasons – Britain due to a relatively more volatile labour market that does not inhibit upward and downward job spells, and Italy to a tighter labour market forcing newcomers into underemployment. But once again, the lion’s share of the accounted variance is explained by respondents’ premigration social class. Quite straightforwardly, low/non-skilled manual workers are much more likely than other EU movers to experience class mobility – necessarily upward, given their starting point – when resettling abroad, net of all other effects. Since most of the working class intra-EU movers belong in the older cohorts of migrants, the association between spatial and intragenerational class mobility in Europe turns out to be on the decline at a time of easier movements across national borders. The higher accessibility of international job trajectories in Europe does not bring about a higher rate of class mobility by default.

CONCLUSIONS Traditional migrants move in space in order to move in the social structure, particularly across generations. In highly fluid societies, opportunities for work, education, career and the bettering of lifestyles have led geographic mobility to be a kind of escalator up the social ladder, often implying moves from periphery regions to core cities. This has always held for internal migration, but European integration poses the question of whether geographical mobility within the unified space of the EU may offer new possibilities to ambitious social climbers escaping entrenched national social hierarchies. Moreover, in a social stratification perspective, the process of European

94

Pioneers of European integration

Table 4.8

Logit models predicting intragenerational class mobility among EU movers

Age Age squared Period of migration 1974–83 1984–93 1994–2003 Gender Male Female Education University degree University entry qualification Upper vocational secondary Primary/lower secondary Nationality German French British Italian Spanish Country of residence Germany France Britain Italy Spain Knowledge COR language at migration Class in COO Bourgeoisie (I–II) Routine non-manual (III) Petty bourgeoisie (IV) High-skilled manual (V–VI) Low/non-skilled manual (VII) Constant 22 Log Likelihood Degrees of freedom Pseudo R-square

B

Std Error

20.04 0.00

0.03 0.00

0.38* 0.40** –

0.17 0.14 –

20.24* –

0.10 –

20.08 20.10 20.08 –

0.22 0.21 0.20 –

20.29 20.04 0.16 20.14 –

0.19 0.16 0.16 0.15 –

0.29 0.35 0.58** 0.48* – 20.03

0.20 0.21 0.20 0.20 – 0.04

21.29** 20.59** 21.31** 20.43* – 1.45 2 482 21 0.12

0.17 0.16 0.26 0.18 – 0.86

Notes: Reference category for dependent variable: immobility. * Significant at 5%; ** Significant at 1%. Immobility 5 0; Mobility 5 1. Source:

EIMSS, N 5 2 154

The social mobility of mobile Europeans

95

integration ‘adds a further potential challenge to the association between class structures and national states’ (Breen and Rottman 1998, 16). European citizens who take jobs abroad in another EU member state, enjoying EU-based rights and facilities (like the recognition of foreign diplomas and free access to national health services), ought to be found on the cutting edge of the transnationalization of social mobility patterns. These people thus form a strategic population for researchers interested in the globalization of systems of social stratification (see Weiss 2005; Berger and Weiss 2008). The European Internal Movers Social Survey (EIMSS) permits a systematic comparison of the social mobility patterns of EU movers and stayers, using the European Social Survey (ESS) as a reference. At the beginning of the chapter, four hypotheses were formulated. In brief, it was posited that the class origins of movers do not differ substantially from those of stayers (H1); in regards to their destination, movers are disproportionately represented in the upper class (H2); movers experience higher intergenerational mobility than stayers (H3); and Southern European movers are more likely to be upwardly mobile than their Northern European counterparts (H4). With the exception of H2, no hypothesis stands our empirical testing. Analyses presented in this chapter found that EU movers are overrepresented in the upper class and underrepresented in the working class, both in their family of origin and in their current situation, contradicting H1 about class origins and confirming H2 about class destinations. On the other hand, this finding matches fairly closely those of studies on interregional migration in countries like Britain (Fielding 2007) and the US (Lin and Christiadi 2002). Contrary to H4, nationality does not affect either intergenerational or intragenerational social mobility significantly, once personal factors are controlled for. H3 is falsified as well. Perhaps the single most important finding of this chapter is that the overall intergenerational mobility of movers does not significantly differ from that of stayers in European societies. However, such a finding needs to be qualified. While upper class movers show a higher propensity than upper-class stayers to maintain their family class position, lower class movers are in fact more intergenerationally mobile than their stayer counterparts. Thus, the free-movement regime of the EU is an alternative route to social mobility for the less privileged young who are ready to leave their home countries, working also as an additional means of class reproduction for the transnational fraction of the European bourgeoisie. The latter group deserves closer scrutiny in future research. Whether migration across Europe serves the purpose of improving the prestige and income of upper-class cadets, or that of defending their ascribed privilege,

96

Pioneers of European integration

is an emerging question that needs to be addressed with some fine-tuning of career movements and occupational rankings (see work in progress by Le Galès and Andreotti 2006). For the time being, social mobility tables inform us that one-fifth of working movers in the EIMSS circulate within the upper class. To this number one must add middle–upper class pensioners who have migrated to the south of the continent – 36.5 per cent of the retirees in our sample (see also King et al. 2000). As a result, a sizeable proportion of EU movers embody a world-regional variant of the global, cosmopolitan upper class. In conclusion, while it cannot be denied that intra-continental free movement is a unique historical experiment with potential far-reaching effects in the long run, so far it has not reshaped the social structure of the EU. Much to the dismay of EU institutions and economic gurus who have long conceived of the European free-movement area as grease in the wheels of the continent’s labour markets (see Recchi 2006 and 2008), cross-state mobility in Western Europe turns out to be quite weakly associated with job-to-job mobility, and even when this is the case, it does not seem to bring about dramatic shifts across class boundaries.

NOTES 1. 2.

3. 4.

5.

6.

Even high-skilled migrants, if not in top managerial positions, are likely to experience some status downgrading in the workplace as they move from less-developed to developed countries (Alarcon 1999; Cornelius and Espenshade 2001). Indeed, one of the functions of immigrant ethnic communities lies in the safeguarding of migrants’ original prestige in social relations outside the workplace in the face of occupational degradation. Co-ethnics acknowledge migrants’ ‘real’ status – that is, the one they had in the local communities from which they came. Unfortunately, while forms and determinants of intergenerational mobility can be studied comparing movers and stayers, data availability only permits a focus on the intragenerational mobility of movers. The process of price and salary homogenization in the EU is still under way. In 2004, when EIMSS was carried out, the median hourly cost of labour in the private sector was 32 euros in Denmark and 10.3 euros in Portugal; within the five countries of our survey, the average worker in Germany earned 28.2 euros and his/her counterpart in Spain 16.3 (Statistisches Bundesamt Deutschland 2007). But even between these two extreme poles of the spectrum of Western European wages, migration flows are apparently insensible to variations in income and unemployment differentials (see Mouhoud and Oudinet 2004; Puhani 2001). Demographically, Italy is the country where the number of EU15 movers is the lowest of the four examined, and the presence of foreign-born workers from other EU15 countries the lowest in the entire EU15 (0.11 per cent of the workforce: Van Houtum and Van der Velde 2004, 102). It is estimated that about 40 per cent of all Europeans living in Britain are in fact in London, gravitating around its global city economy (Piracha and Vickerman 2003). Ethnographic accounts of the careers and prospects of EU movers in the City and other highly qualified sectors of the London economy can be found in Favell (2008a).

The social mobility of mobile Europeans 7.

8. 9. 10. 11.

97

Notoriously, Germany hosts significant numbers of another category of intra-EU movers: posted workers from fiscally less onerous countries (for instance, workers in the construction sector from the UK). However, at least legally, posted workers in the EU are entitled to no more than a six-months stay, and are therefore excluded by the EIMSS sampling criteria (at least one year of permanent residence). The hierarchical arrangement of social classes that generates these four types follows Erikson and Goldthorpe (1992). More precisely, the odds of intergenerational immobility for upper class movers are 2.5 times those of mobility for movers originating from classes IV and V–VI. However, it is true that the spread of longitudinal studies and the development of event history analysis have revived research on intragenerational mobility in connection with life course approaches (for an overview, see Mayer 2005). EU movers who never worked in the destination country are excluded from this analysis, as well as those who took their first job in life only after they migrated, as it is impossible to assess the impact of migration on their career mobility. Moreover, due to a lack of comparable data, findings presented in this section refer to EU movers only.

5.

Living across cultures in a transnational Europe Antonio Alaminos and Oscar Santacreu

ACCULTURATION OR INTEGRATION? THEORETICAL PREMISES As other chapters have documented, resettling to another country entails various degrees of mobility, both in terms of time and previous movements. Some migrants decide to remain indefinitely in the destination country, while others keep alive the wish to return to their country of origin as soon as possible. Some have had significant mobility experiences before, with long stays in several countries. For others, migration is in fact a new and one-shot experience. Whatever, the case, for all migrants, moving internationally is a cumulative and dynamic life choice that can lead to great changes in the individuals and (cumulatively) in the societies where they choose to settle. These questions can best be approached with reference to the well-developed literature in immigration studies on ‘acculturation’ or ‘integration’. Before tackling the data revealed by EIMSS as regards this issue, however, we need first to clarify the key concepts used in this chapter. Acculturation According to a classic definition: acculturation comprehends those phenomena which result when groups of individuals having different cultures come into continuous first-hand contact, with subsequent changes in the original culture patterns of either or both groups. Under this definition, acculturation is to be distinguished from culture change, of which it is but one aspect, and assimilation, which is at times a phase of acculturation. (Linton et al. 1936, 145–6)

In an updated version of this perspective, the most well-known contemporary theorist of acculturation, John W. Berry (1997), outlines that the acculturation process is associated with psychological and physical changes 98

Living across cultures in a transnational Europe

99

in immigrants due to adaptation to a new diet, climate and housing, and the conventions of social interaction, rules and values of the new society. It is important to highlight the differences between acculturation and assimilation. Assimilation is not the inevitable consequence of acculturation. For instance, in certain circumstances, such as when migrants are spatially concentrated, assimilation will hardly be a result of acculturation. On the other hand, acculturation takes place both on an individual and a group level. This means that, even among individuals from the same cultural group living in the same geographical area and with similar socioeconomic characteristics, the acculturation process can produce different outcomes because of differing personal experiences (Kelly et al. 1994). A key formulation such as Hofstede’s acculturation curve model (2003) is premised on such an individual perspective. The psychological states experienced by migrants during their acculturation tend to follow a fourstage sequence. The first stage is usually short, and it is characterised by a state of euphoria: the excitement of travelling and living in a different society and culture. The second stage is the period of culture shock, when real life starts in the new environment. The third stage is a period during which migrants slowly begin to function in the new context, learn some local values, find increased self-confidence, and choose their own lifestyle in the new social environment. The acculturation process can have different outcomes that lead to a fourth stage. During this last stage, migrants adopt a stable psychological disposition that may be different from the one they had in their society of origin. Some might find it difficult to live with locals in their new society. In other cases, however, foreign residents over-assimilate and become ‘happy natives’. The acculturation process entails, to a great extent, the acquisition of intercultural competences. The first approaches to this basically distinguished three areas: affective (attitudes), knowledge (cognition) and skills (behaviours) (Bloom 1956). ‘Awareness’ was later recognised as another important element for migrants to be able to develop an adequate intercultural competence. Stevens (1971), Curle (1972) and Gattegno (1976), amongst others, mention self-awareness and awareness of others as keys to being able to interact adequately in another culture. More recent models of intercultural competence consider these five dimensions of learning to be relevant: awareness, attitudes, skills, knowledge and linguistic competence in the language of the host society. In any case, intercultural competence does not develop in a synchronic or integrated way, both in the number of dimensions it comprises and the degree of competence achieved. Acculturation also raises significant measurement issues. Empirical scales to measure acculturation are often plagued by the pitfall of unidimensionality. Generally speaking, linear unidimensional scales conform to

100

Pioneers of European integration

an assimilationist approach: the more American a, say, Pakistani migrant becomes, the less Pakistani he or she is. This means that the acquisition of new cultural competences is associated with the loss or weakening of the original ones. Many linear unidimensional scales have been constructed, such as those by Marín et al. (1987), Ghuman (1991), Cortes et al. (1994), Landrine and Klonoff (1994), Cuellar, Arnold and Maldonado (1995), Dawson et al. (1996) and Hofstede (2001). However, measurements over just one continuum impose unnecessary theoretical restrictions since this reduces culture to a single dimension. Several researchers responded by developing multidimensional orthogonal scales (Berry 1970; Zak 1973; Teske and Nelson 1974; Clark et al. 1976; García and Lega 1979; Padilla 1980; Szapocznik et al. 1980; Sánchez and Fernández 1993; Suinn et al. 1995; Ryder et al. 2000). Multidimensionality allows us to understand the acculturation process as a hybridisation process in which the acquisition of intercultural competence does not necessarily imply accepting the new culture (language, values, food, clothes, and so on) as a single and indivisible package. Some authors have observed that unidimensional scales tend to be more successful in the United States. The assimilation literature in the social sciences from Gordon (1964) to Alba and Nee (2004) has clearly conceptualised this process as one that measures structural convergence with middle and upper class norms (of education, housing, social mobility, and so on). However, in the Unites States there has been a normative tendency to blur this process with cultural issues of becoming ‘American’, where political, cultural and ethnic proofs of assimilation to the dominant ‘white’ majority are considered a goal – whether positively or coercively – for migrants. In Canada, though, scales tend to be multidimensional, in accordance with public immigration policies and what is considered socially acceptable (see Berry 1984; Taylor and Lambert 1996). As with the concept of assimilation, disagreements on the concept of acculturation thus revolve around the best way of operationalizing it, basically because this involves a political stance regarding what is expected from migrants. Our view is that there are various ways of living abroad, which can hardly be captured by unidimensional scales, and that a first step consists in defining the critical dimensions of acculturation. Berry (2003) suggests a double dimensionality, based on orientations towards one’s group and towards other groups (that is, the host society), as expressed in social contacts and collective actions. The importance of Berry’s approach lies in the fact that it ascribes a strategic action to migrants. Padilla (1980) also proposes two dimensions. One dimension, called ‘ethnic loyalty’, considers the migrants’ cultural orientation, according to their preference for the culture of origin or the culture of the society of destination. The other dimension is ‘cultural awareness’, which takes into account the cultural

Living across cultures in a transnational Europe

101

competence of migrants from the viewpoint of both the culture of origin and the culture of destination. The first dimension is psychological. The second one considers the intercultural competence of migrants – that is, their knowledge of both cultures (origin and destination). In this sense, Padilla introduces a qualification about migrants’ ability (their intercultural competence) to adjust preferences and actual behaviours. Berry’s and Padilla’s scales are clearly complementary. Both share a dimension that refers to ‘preferences’ or ‘cultural orientations’, and include a second dimension – ‘social participation’ for Berry, and ‘intercultural competence’ (knowledge of language, norms, values and so on) for Padilla. These may be interdependent formulations, although Berry’s emphasizes behaviour as the basis for measurement, while Padilla’s highlights knowledge of the environment. Both these alternatives are important. It is difficult to have an intense social participation in the society of destination without an acceptable knowledge of its language and social codes. Intercultural competence encompasses several skills, some of which are harder to achieve than others, for instance, the command of etiquette, when to bring a present, congratulate or greet, appropriate clothing or language tones. Learning the underlying values or interpreting situations or behaviours correctly is much more difficult. Feeling disoriented about those meanings can cause acculturation stress (uncertainty, anxiety and depression), which may lead to hostility towards the new social environment and cultural rejection. In this sense, intercultural competence is essential in order to facilitate participation or create a preference for a different lifestyle. Integration Although there are parallels in the application of these concepts to the immigration process, a distinct line of approach to the same questions is taken if the concept of ‘integration’ is adopted. For example, Schoorl (2005) prefers to use this concept when referring to the process through which migrants become part of the social, cultural and institutional fabric of the society that hosts them. The concept of integration thus pays attention not only to migrants’ adaptation, as does acculturation, but also to the structural effects on the society in which they reside and its more or less welcoming orientations. Schoorl identifies four significant forms of integration: socioeconomic integration (structural), social integration, cultural integration and identification. The structural integration dimension refers to immigrants’ equal access to services and institutions: education, labour market, housing, political system and health care (equality versus discrimination). Social integration refers to the degree of interaction

102

Pioneers of European integration

between immigrants and the local population (segregation versus mixing). Cultural integration refers to the degree to which different groups share the same norms, values and preferences (homogeneity versus heterogeneity). Finally, from the viewpoint of integration, the identity dimension refers to the degree to which immigrants identify with the society in which they live, on the one hand, and, on the other, the degree to which immigrants are accepted by that society (inclusion versus exclusion). Inclusion does not necessarily mean at first the loss of identification with the culture of origin. Obviously, there is not a synchrony between these different dimensions of integration. Socioeconomic integration or cultural integration can be interdependent, but this is not necessarily the case. Again, overlaps in these distinctions can be detected with the dominant US paradigm of assimilation. Thus, the segmented assimilation theory (Portes and Zhou 1993; Portes and Rumbaut 2001) suggests that a high degree of structural (socioeconomic) integration can coexist perfectly with a strong orientation towards one’s own ethnic or cultural group. Time also plays a paramount part in the dynamics of the immigration process. O’Reilly (2000) classifies migrants into ideal types taking into account two dimensions: orientations towards one’s country of origin and the duration of residence in each country. He applies this model to intra-European migrants, particularly residents in Spain. This classification model shows several limitations, which come from the combination of two highly interdependent dimensions. Migrants’ orientation towards the country of origin and the country of destination change over time. Moreover, the time factor should include not only the past (the time an individual has been living in the country), but also migrants’ expectations regarding the future. The (explicit or implicit) intention of returning to the country of origin has frequently been used to explain the persistence of ethnic differences or the unwillingness to integrate into the new social environment (for example, Bonacich 1973; Anwar 1979; Anthias 1992; Uriely 1994). Starting from the theoretical models that we have just reviewed, we will adopt the integration rather than the acculturation terminology, while using some tools from the acculturation literature. In our view, there are two dimensions that adapt particularly well to the reality of intra-European migration. The first dimension refers to what we think of as social integration. It takes into account Berry’s and Schoorl’s notions of social participation and Padilla’s view of intercultural competence. In short, it looks at migrants’ social networks, as well as their ability to coexist effectively with the local population. This dimension is operationalized by measuring on the one hand the number and nationality of friends, and linguistic competence on the other. Both indicators are closely correlated; the preponderance of

Living across cultures in a transnational Europe

103

friends from the country of residence implies a higher level of linguistic competence. Among the different dimensions of intercultural competence, we consider linguistic skills to be basic and fundamental. The possibilities of integrating into the everyday life of the new social environment are seriously impaired by language weakness. The second dimension that can be used to conceive how migrants live in and perceive their new country of residence is that of cultural integration. It includes a widespread idea in the literature reviewed above: that is, the preference for the culture of origin or the culture of destination makes a difference in immigrants’ strategies for dealing with the new society of residence. Together with the possible cultural discontent or well-being of migrants, it is also important to bear in mind their expectations of future mobility. The way migrants experience a new social situation depends on their expectations or wishes to stay in it. Those migrants wishing or expecting to return soon to their country of origin will show more indifference towards the society they live in. Preferences for life in the country of origin or that of destination and the wish to return to the country of origin are the indicators used in our analysis. These two dimensions characterize differently the way European migrants live in their new social environment, and constitute a satisfactory synthesis of existing theoretical models. As stated above, the immigration process – whether conceptualized as acculturation, assimilation or (as we do) integration – is an opaque process, the pace of which depends on the individual, that cannot be seized clearly in a single survey. However, the distinctiveness of EU movers as regards other kinds of migrant raises intriguing questions about the kind and degree of social and cultural integration they experience in the society in which they have chosen to live.

LINGUISTIC COMPETENCE AND SOCIAL NETWORKS Language is one of the critical issues of living in a foreign society, as it mediates almost every daily life situation. EU movers’ command of the language of their country of destination is overall quite poor. English is the only language in which practically all migrants groups show a high level of competence. Particularly noteworthy is the modest knowledge of the host-country language shown by German, British and French nationals upon their arrival in Spain or Italy, and the equally weak control of German declared by the Spanish, Italians and British moving to Germany. See Table 5.1 for full details. Quite surprisingly, in the light of the changes in social composition and

104

Pioneers of European integration

Table 5.1

Knowledge of COR language of EU movers at time of migration (row % )

Country of origin and country of residence

Almost as well as native language

Quite well

Just soand-so

Poorly

No knowledge

6.7 12.2 5.6 5.7 10.4 23.5 2.4 4.0 7.2 3.1 4.0 1.2 15.7 0.8 3.2 15.4 6.8 0.8 5.1 2.4

22.7 44.5 15.7 8.5 14.0 47.8 11.6 3.2 16.8 11.8 11.6 3.7 21.0 2.8 28.7 12.6 15.2 5.9 25.5 15.2

17.6 20.2 10.4 17.1 30.0 20.0 18.4 8.0 21.6 9.4 10.8 11.0 10.9 2.8 23.9 12.2 13.2 9.5 29.3 12.0

16.9 12.6 12.9 26.0 25.2 7.8 32.0 18.1 35.6 24.8 17.1 40.8 18.1 11.8 19.9 23.2 22.0 15.4 19.1 23.6

36.1 10.5 55.4 42.7 20.4 0.8 35.6 66.7 18.8 50.8 56.6 43.3 34.3 81.9 24.3 36.6 42.8 68.4 21.0 46.8

French in Germany French in Britain French in Italy French in Spain Germans in France Germans in Britain Germans in Italy Germans in Spain British in France British in Germany British in Italy British in Spain Italians in France Italians in Germany Italians in Britain Italians in Spain Spanish in France Spanish in Germany Spanish in Britain Spanish in Italy Source:

EIMSS, N 5 4 901

education levels of the population at stake, more recent migrants are not significantly more proficient in the language of the destination country than their predecessors. Only the Spanish and Italians who moved to Britain in the 1990s and after fare linguistically better than older cohorts of co-nationals crossing the Channel. Nevertheless, EU movers of all generations are a select group also in terms of their original linguistic competence: the proportion of them claiming to have had a good command of the language of the place where they have resettled is definitely higher than that of their compatriots at home (Table 5.2). With regard to the current knowledge of the language of the country of residence, British nationals show the lowest level of knowledge of foreign languages. However, we should bear in mind that English is the most widespread language, which reduces the impact of this lack of knowledge.

Living across cultures in a transnational Europe

Table 5.2

105

Stayers’ and EU movers’ knowledge at migration of other EU languages by country of residence (‘quite well’ or ‘almost as well as native language’) (%) Language° French

French Germans Italians British Spanish

Movers Stayers Movers Stayers Movers Stayers Movers Stayers Movers Stayers

German 29.4 7.0

24.4 12.0 36.7 11.0 24.0 14.0 22.0 8.0

3.6 3.0* 14.9 6.0 6.7 1.0*

English

Italian

Spanish

56.75 34.0 71.3 51.0 31.9 29.0

21.3 4.5* 14.0 2.1*

14.2 10.0 7.2 2.5* 28.0 4.0 4.9 3.2*

30.6 20.0

15.6 1.6* 17.6 1.3*

Notes: ° This is the main language spoken in the COR of EU movers in each column (i.e., 24.4% in the first column refers to the proportion of German movers living in France speaking French well when they migrated, compared to 12% of the German population with a similar knowledge of French). Source: Data on movers are from EIMSS. Data on the general population are taken from Eurobarometer 63.4; * from Eurobarometer 55.1, N 5 4 997

Particularly noteworthy is the low level of knowledge of the Spanish language shown by British and German residents in Spain; a relatively modest proficiency in French is also the norm among Germans living in France (Table 5.3). The duration of stay abroad, in these groups of EU movers, makes little difference: German and British residents in Spain who admit to not speaking Spanish at all average 13 years of residence in this country. As a matter of fact, British and German residents in Spain are the migrants who least improve their linguistic competence. In contrast, Spanish and Italian nationals are the migrants who most actively seek to improve their linguistic competence in all countries of destination. In all groups, there is a marked improvement of linguistic competence over time. Another key aspect of social integration has to do with the friendship relations of movers (Table 5.4). The general trend is for the percentage of migrants with friends of a different nationality to be noticeably higher than those with friends of the same nationality. In contrast, German and British residents in Spain have the highest number of friends of the same nationality, confirming O’Reilly’s (2000) evidence. A similar case of an ethnic segregation of personal networks is found among Italian residents

106

Pioneers of European integration

Table 5.3

Knowledge of language at time of survey (row %)

Country of origin and country of residence

Almost as well as native language

Quite well

Just so-so

Poorly

No knowledge

42.4 72.3 59.8 25.6 30.0 83.5 37.6 12.4 25.3 34.6 49.4 7.7 53.6 15.0 57.4 57.7 37.3 18.7 54.1 42.4

32.9 23.9 30.1 41.9 28.0 14.5 32.8 18.9 36.5 40.6 34.3 20.3 35.9 51.2 38.2 26.4 38.6 40.9 40.8 43.2

14.1 3.4 6.4 18.7 28.4 2.0 22.4 33.7 28.9 17.7 11.6 43.9 8.5 24.8 2.8 9.8 14.9 29.8 3.8 11.2

7.5 0.4 2.8 11.0 10.8 0.0 6.4 24.9 8.4 6.7 3.2 23.6 1.6 8.3 0.8 2.0 6.4 9.9 1.3 3.2

3.1 0.0 0.8 2.8 2.8 0.0 0.8 10.0 0.8 0.4 1.6 4.5 0.4 0.8 0.8 4.1 2.8 0.8 0.0 0.0

French in Germany French in Britain French in Italy French in Spain Germans in France Germans in Britain Germans in Italy Germans in Spain British in France British in Germany British in Italy British in Spain Italians in France Italians in Germany Italians in Britain Italians in Spain Spanish in France Spanish in Germany Spanish in Britain Spanish in Italy Source:

EIMSS, N 5 4 899

in Germany. On the other hand, the EU movers with the highest number of friends from their country of residence are Italians living in Spain and Britain, and Spanish and French residents in Italy. Overall, linguistic competence and the density of the network of friends in the country of destination show a very significant covariation, indicating their commonality in expressing the degree of social integration of EU movers in the host societies.

NATIONAL ORIENTATION AND MOBILITY EXPECTATIONS Which society – that of origin or that of destination – do EU movers prefer? We would hypothesize that those movers who feel closer to their

Living across cultures in a transnational Europe

Table 5.4

Density and nationality of network of friends (row %)

Country of origin and country of residence French in Germany French in Britain French in Italy French in Spain Germans in France Germans in Britain Germans in Italy Germans in Spain British in France British in Germany British in Italy British in Spain Italians in France Italians in Germany Italians in Britain Italians in Spain Spanish in France Spanish in Germany Spanish in Britain Spanish in Italy Source:

107

Many more from COO

More from COO

Same from both countries

More from COR

Many more from COR

2.4 1.7 1.2 0.8 1.2 0.8 2.4 20.2 0.8 1.2 2.0 6.6 1.2 2.8 2.8 0.0 2.0 3.2 0.6 0.0

16.2 12.2 3.2 17.3 16.4 9.1 17.6 25.8 10.4 16.5 6.4 38.3 13.8 35.8 21.9 6.7 25.3 22.7 14.0 9.2

35.6 37.6 24.7 35.8 22.8 24.4 20.8 41.1 47.8 21.7 26.3 39.9 34.8 41.3 33.1 24.2 22.9 29.9 35.7 20.5

22.9 29.1 34.0 34.2 37.2 46.1 37.6 9.7 31.7 32.3 39.0 14.0 36.0 18.5 31.9 55.0 30.9 33.1 32.5 40.6

22.9 19.4 36.8 11.9 22.4 19.7 21.6 3.2 9.2 28.3 26.3 1.2 14.2 1.6 10.4 14.2 18.9 11.2 17.2 29.7

EIMSS, N 5 4 877

country of origin will find it more difficult to integrate into the society they live in. The four response categories in Table 5.5 correspond to the acculturation strategies put forward by Berry (1997): assimilation, integration, segregation and marginalization. First of all, we find that the most frequent ‘national orientation’ is twofold: towards the country of residence and the country of origin. Going into detail we see that a predilection for the country of origin is highest among Spanish nationals living in any country of destination, followed by Italians in any country of destination except Spain, and French citizens living in Germany or Britain. On the other hand, the highest level of preference for the country of residence can be found among British residents everywhere. ‘Sympathy’ with the country of residence is also high among German residents in Italy and France, and among French and Italian residents in Spain. Generally speaking, EU movers from the south of the continent appear to stick

108

Pioneers of European integration

Table 5.5

National orientation of EU movers (row %)

Country of origin and country of residence French in Germany French in Britain French in Italy French in Spain Germans in France Germans in Britain Germans in Italy Germans in Spain British in France British in Germany British in Italy British in Spain Italians in France Italians in Germany Italians in Britain Italians in Spain Spanish in France Spanish in Germany Spanish in Britain Spanish in Italy Source:

Towards COR

Either

Towards COO

Neither

8.3 9.8 13.8 20.4 20.1 17.8 24.9 9.3 36.3 18.2 24.7 34.2 5.3 6.7 7.6 16.5 6.5 7.1 10.3 4.5

67.9 66.2 69.6 68.6 67.9 60.9 62.9 76.6 57.7 58.1 64.1 57.1 72.0 73.4 62.0 75.3 74.6 66.8 63.2 75.3

20.2 17.1 7.3 4.5 5.2 13.0 5.7 10.1 2.8 12.6 7.6 4.6 18.7 17.1 26.0 4.9 14.5 22.5 23.2 17.8

3.6 6.8 9.3 6.5 6.8 8.3 6.5 4.0 3.2 11.1 3.6 4.2 4.1 2.8 4.4 3.3 4.4 3.6 3.2 2.4

EIMSS, N 5 4 859

to their roots more than their counterparts from Central and Northern Europe. Expectations regarding the future are also a component of cultural integration into the host society. Migrants who wish to return to their country of origin as soon as possible convey considerable indifference towards the society they are currently living in. Table 5.6 shows how nostalgia is higher among Southern Europeans. A strong wish to return is also shown by French residents in Germany and Britain, and British in Germany. As we have just seen, there is a close relationship between these two measures, in the sense that a preference for life in the country of origin or the country of residence is associated with the wish to return to the former or remain in the latter. Almost 80 per cent of migrants who prefer their current country of residence do not expect to move in the future. On the other hand, 69 per cent of those preferring life in their country of origin nurture the idea of going back home.

Living across cultures in a transnational Europe

Table 5.6

109

Wish to return to the country of origin of EU movers (row %) Wish to return

French in Germany French in Britain French in Italy French in Spain Germans in France Germans in Britain Germans in Italy Germans in Spain British in France British in Germany British in Italy British in Spain Italians in France Italians in Germany Italians in Britain Italians in Spain Spanish in France Spanish in Germany Spanish in Britain Spanish in Italy Source:

No

Indifferent

Yes

31.4 37.8 46.2 63.4 60.8 43.9 63.2 51.4 75.2 52.0 52.6 80.9 39.5 23.6 33.1 65.0 37.2 21.3 26.1 20.4

9.4 10.9 7.2 8.9 9.2 18.0 11.6 14.9 8.4 7.5 11.2 9.3 7.3 7.9 10.4 14.2 11.6 7.5 13.4 6.0

59.2 51.3 46.6 27.6 30.0 38.0 25.2 33.7 16.4 40.6 36.3 9.8 53.2 68.5 56.6 20.7 51.2 71.1 60.5 73.6

EIMSS, N 5 4 901

CULTURAL AND SOCIAL INTEGRATION: TWO LATENT DIMENSIONS A factor analysis of the indicators illustrated so far shows that they revolve around two independent dimensions.1 These can be thought of as the latent dimensions that we conceptualize in terms of integration. We name the first cultural integration. It measures the degree of affinity of individuals with their cultural environment – that is, their acceptance of values and lifestyles different from those of their country of origin. The other dimension, which we name social integration, deals with participation, through meaningful social relations, in the host society. Both dimensions enable us to profile how migrants live in the society of residence. An important aspect to bear in mind is that social integration shows only a weak correlation with cultural integration. In other words, both dimensions, although associated, act independently.

110

Pioneers of European integration

Social integration, as measured by the factor scores of this analysis, is very low among German and British residents in Spain. Italians and Spanish in Germany also have difficulties in integrating socially. On the other hand, a particularly high level of social integration is found among movers to Italy – whatever their nationality. Germans and French show a fairly good social integration in Britain as well. Cultural integration encompasses important elements of psychological closeness. Unlike social integration, which emerges through objective indicators of social participation and intercultural competence, cultural integration has to do with strongly subjective orientations. The weakest cultural integration, or psychological closeness, is found among Spanish migrants in any of the countries analysed in this study. Italian migrants also show low cultural affinity with all host societies except Spain. A slightly higher level of cultural integration can be found among French migrants, particularly those who live in Germany and Britain. Southern Europeans are especially at odds in Northern Europe, whereas German and British movers are more at ease with the countries they have migrated to. Integration takes place on different levels, which are not necessarily parallel. The two dimensions we have taken into account help us understand situations that would at first appear to be contradictory. Thus, we can find high integration in social terms (good intercultural competences and networks of friends from the country of destination) coupled with low cultural integration (that is, detachment from the values and culture of that country). Some groups of migrants are poorly integrated, both culturally and socially. This is the case for Italians and Spanish in Germany. Their migration is basically work-related and, in this sense, forced by the search for better living conditions. In fact, the integration of British and German residents in Spain and France offers an interesting paradox in that it is strong culturally but weak socially. The situation of British residents in France is similar, although not so polarized. In these categories of movers, we find a positive subjective perception of belonging to the places where they live, but with a limited participation in their social life. The reverse situation is the rule in Britain: all groups of movers (except Germans) show good social integration together with the lowest levels of cultural integration. We may ask ourselves how some movers can feel culturally integrated if they cannot speak the language or do not have friends among the locals. A first answer is that it is a question of ‘awareness’ – that is, of recognition of the depth and nuances of a culture. Migrants can achieve basic competences in the lifestyle of the country of residence, such as dress and eating codes or greeting formulas. However, they do not acquire cognitive

Living across cultures in a transnational Europe

111

Spanish in Italy Spanish in Britain Spanish in Germany Spanish in France Italians in Spain Italians in Britain Italians in Germany Italians in France British in Spain British in Italy British in Germany British in France Germans in Spain Germans in Italy Germans in Britain Germans in France French in Spain French in Italy French in Britain French in Germany –1.5

–1

–0.5 Social integration

Source:

0

0.5

1

Cultural integration

EIMSS, N 5 4 880

Figure 5.1

The social and cultural integration of EU movers: averages of factor scores

competence in the values and meaning of what happens around them. That is part of the acculturation process. The result is a superficial coexistence. Figure 5.1 shows the group differences on these two dimensions of integration. Bars growing in the same direction show coincidence (towards high or low levels of integration) between social and cultural integration – that is, high participation and competence together with affinity and psychological closeness. Bars growing in opposite directions show situations in which social and cultural integration do not coexist. It is worth mentioning the lack of symmetry in cultural integration (Table 5.7). One might think that the poor adjustment of Italian residents in Germany would imply reciprocity – that is, low integration of German residents in Italy. However, that is not the case. Thus, for instance, we find that British migrants in Germany do not on average feel comfortable, whereas German migrants in Britain do. This might have to do with the particular national environments that migrants find themselves in; or with the kinds of migrants who choose to move to particular countries.

112

Pioneers of European integration

Table 5.7

Mean scores for the cultural integration dimension Cultural Integration

Italians in Germany Spanish in Germany Italians in Britain Spanish in Britain French in Britain Spanish in Italy French in Germany Italians in France Spanish in France British in Germany Source:

20.44 20.42 20.41 20.29 20.26 20.25 20.23 20.20 20.16 20.12

Cultural Integration Germans in Italy Germans in Spain British in Italy British in Spain British in France Italians in Spain Germans in France French in Italy French in Spain Germans in Britain

0.33 20.04 0.16 0.50 0.73 0.25 0.43 0.05 0.23 0.11

EIMSS, N 5 4 880

In order to better understand the complex interplay between cultural and social integration, we need to know not only how migrants feel, but also how they feel they are perceived by the society around them. Even in Western European societies, where cultural distance is not dramatic (that is, there is no strong ‘ethnic’ prejudice against other fellow Western Europeans), EU citizens living abroad may still feel discriminated against (Table 5.8). Such perception of being discriminated against seems to depend on two key elements: intercultural competence (being able to perceive the nuances of social reality), and social interactions from which the occasion to perceive oneself as being discriminated against could arise. We found that those EU movers who claim to have a good knowledge of the language report to a greater degree that they feel discriminated against, and vice versa. In particular, British and German migrants in Spain, who are less proficient in the local languages, might be said to feel more culturally integrated in Spain because they do not feel discriminated against by the society in which they live. Researchers who study the migration flows from Northern Europe towards Spain or other Mediterranean countries stress the fact that these migrants tend to live in secluded areas. This is the actual situation of a proportion of the British residents in Spain. It is a free choice, yet their lifestyle in a foreign country seems not to cause conflicts, especially because they do not have the perception of being discriminated against by Spanish society. All this leads to their high cultural integration on these measures together with their modest level of social interactions with locals. On the other hand, Spanish migrants with high social integration do feel

Living across cultures in a transnational Europe

Table 5.8

113

Perception of being discriminated against because of nationality (row %) Feels discriminated against because of nationality

French in Germany French in Britain French in Italy French in Spain Germans in France Germans in Britain Germans in Italy Germans in Spain British in France British in Germany British in Italy British in Spain Italians in France Italians in Germany Italians in Britain Italians in Spain Spanish in France Spanish in Germany Spanish in Britain Spanish in Italy Source:

No

Yes

65.5 62.4 75.0 76.8 78.4 69.0 80.4 92.4 72.0 64.6 59.8 84.1 66.9 75.1 72.5 79.2 62.7 62.9 64.3 76.8

34.5 37.6 25.0 23.2 21.6 31.0 19.6 7.6 28.0 35.4 40.2 15.9 33.1 24.9 27.5 20.8 37.3 37.1 35.7 23.2

EIMSS, N 5 4 893

discriminated against in Germany, Britain and France, which affects their cultural adjustment. The subjective perception of discrimination is a relevant factor in understanding the degree of satisfaction with the life migrants lead in a foreign country. Interestingly, there is some correspondence between the perception of discrimination and societal resistance to immigration from other EU member states, as measured by Eurobarometer. Data from EB53 of 2000 show that, among the countries included in our survey, those with less resistance to intra-European mobility are Spain (55 per cent) and Italy (54 per cent), where public opinion considers that there should be unrestricted freedom of mobility. In France, most of the public (51 per cent) is in favour of mobility with some restrictions, whereas in Germany and the United Kingdom 60–62 per cent of citizens state that they would like to see restrictions placed on mobility between European Union countries. Hence,

114

Pioneers of European integration

there appears to be some objective basis for the subjective impressions of European migrants.

PSYCHOLOGICAL ADAPTATION Taking a lead from the acculturation literature, we shall now examine the concept of psychological adaptation to assess the quality of life achieved by migrants in their country of destination on the basis of the integration strategies they have adopted. An important gauge in assessing the well-being of foreign residents is how satisfied they feel with their life as a whole. In this sense, we have chosen a direct approach and taken into account movers’ own assessment of their quality of life. Then, we have looked more thoroughly into their emotional situation through an indirect approach based on the notion of homesickness. Satisfaction with Life On a scale from 0 to 10, ‘satisfaction with life as a whole’ was measured among the general public by the European Social Survey of 2004. In Spain and Britain, the average score was 7; it was 6.9 in Germany and Italy and only 6.4 in France. Generally speaking, the mean level of satisfaction reported by EU movers, regardless of the country of residence or origin, was higher than the level indicated by the public opinion in their country of origin. The lowest satisfaction, with a mean of 7.3, was reported by Spanish residents in Germany. The highest level was reported by British in France (8.3) and Spain (8), and by Germans in France (8.2). Satisfaction with life has a direct and very important relationship with cultural integration (Table 5.9). It seems reasonable to assume that higher psychological closeness results in higher levels of satisfaction with life in that social environment. Indeed, the satisfaction of movers who report lower-than-average levels of cultural integration in the destination country is even lower than the satisfaction of stayers in their country of origin. On the other end of the cultural integration scale, migrants who report higher cultural integration or affinity with the host society reach very high means of satisfaction with their life: between 8 and 9 points, which is practically a ceiling effect considering that this is a mean. While satisfaction with life among movers is clearly associated with cultural integration, it has hardly any relationship with social integration. Table 5.10 shows the mean satisfaction of migrants according to their level of social integration – that is, the density and nationality of their network of friends, and the linguistic ability to get to know and communicate with

Living across cultures in a transnational Europe

Table 5.9

Mean satisfaction with life and levels of cultural integration of EU movers (0–10 scale) Total

French in Germany French in Britain French in Italy French in Spain Germans in France Germans in Britain Germans in Italy Germans in Spain British in France British in Germany British in Italy British in Spain Italians in France Italians in Germany Italians in Britain Italians in Spain Spanish in France Spanish in Germany Spanish in Britain Spanish in Italy Source:

115

7.61 7.47 7.65 7.70 8.21 7.95 7.96 7.62 8.30 7.25 7.52 8.02 7.91 7.43 7.54 7.99 7.60 7.34 7.57 7.60

Cultural Integration Low

MediumLow

MediumHigh

High

5.7 5.5 4.4 4.2 4.5 5.5 4.2 5.5 5.2 4.9 4.4 4.2 5.9 5.2 6.3 4.9 5.2 5.3 5.7 4.9

7.4 7.3 7.1 7.0 7.2 7.3 7.1 7.0 7.0 7.0 6.9 7.2 7.7 7.5 7.4 7.3 7.2 7.3 7.4 7.4

8.5 8.2 8.3 8.3 8.3 8.4 8.3 8.5 8.3 8.0 8.2 8.1 8.8 8.7 8.4 8.4 8.6 8.6 8.6 8.5

8.8 9.1 9.2 8.6 9.6 9.4 9.3 9.6 9.1 9.1 9.1 8.9 9.6 9.5 8.9 9.0 9.6 9.5 9.2 9.5

EIMSS, N 5 4 880

their social environment. No clear-cut pattern can be detected. In fact, a higher satisfaction with life is associated with lower social integration only among German and British residents in Spain. Overall, satisfaction with life is a subjective feature that is strongly associated with cultural integration (0.77**), and poorly correlated (and negatively) with social integration (−0.07**).2 Such a negative association, where the increase of intercultural competence and high social presence is associated with lower satisfaction with life, is particularly noticeable in certain groups of migrants: Germans in Italy (−0.27**), French in Italy (−0.14*), French in Germany (−0.27**) and British in Spain (−0.14*). Both dimensions are clearly differentiated, as we can see below. Living in another country is a very special situation that enables migrants to compare their experiences and their knowledge of the country of origin

116

Pioneers of European integration

Table 5.10

Mean satisfaction with life and the social integration of EU movers (0–10 scale) Total

French in Germany French in Britain French in Italy French in Spain Germans in France Germans in Britain Germans in Italy Germans in Spain British in France British in Germany British in Italy British in Spain Italians in France Italians in Germany Italians in Britain Italians in Spain Spanish in France Spanish in Germany Spanish in Britain Spanish in Italy Source:

7.61 7.47 7.65 7.70 8.21 7.95 7.96 7.62 8.30 7.25 7.52 8.02 7.91 7.43 7.54 7.99 7.60 7.34 7.57 7.60

Social Integration Low

MediumLow

MediumHigh

High

8.4 8.3 7.9 7.7 7.5 8.6 8.4 7.5 8.0 7.0 7.8 8.0 8.1 7.7 7.7 8.1 7.8 7.0 7.8 7.3

7.5 7.3 7.9 7.8 8.5 7.7 8.2 7.6 8.4 7.5 7.8 8.3 7.7 7.3 7.3 7.9 7.6 7.3 7.2 7.4

7.7 7.5 7.7 7.7 8.4 8.2 7.9 8.3 8.4 7.5 7.6 7.5 8.1 7.4 7.7 8.1 7.8 7.7 7.7 7.9

7.0 7.7 7.3 7.3 8.1 7.4 7.2 7.7 7.9 6.3 7.1 6.3 7.3 6.5 7.6 7.7 6.9 7.4 7.9 7.2

EIMSS, N 5 4 880

with their experiences and knowledge of the country of destination. In our survey, we asked respondents to compare the quality of different dimensions of life in the countries of origin and destination: education, bureaucracy, health care services and public safety. Two scales have been built from the number of times a service or institution is believed to be better in the country of origin and the number of times it is said to be better in the society of the country of residence. Cultural integration shows a positive correlation with a preference for the services and institutions of the country of residence (0.04**). The correlation is negative in the case of preference for services and institutions in the country of origin (−0.09**). Thus, cultural integration is associated with a higher preference for services in the destination country. However, when it comes to social integration, the trend is quite the opposite. As the level of social integration increases, the preference for services and institutions in the country of residence lowers

Living across cultures in a transnational Europe

117

(20.04**), and the preference for those in the country of origin rises (0.12**). This trend emerges in all groups of migrants consistently. Homesickness and Integration To conclude, we examine another indicator of psychological adaptation: homesickness. Any novel or folk song on migrants revolves around this feeling – nostalgia is the psychological outcome of migration par excellence. Here we shall consider homesickness as a multidimensional phenomenon. Our analysis of open answers to the question: ‘What do you miss the most from your country of origin?’ shows that there are three main dimensions of homesickness. Firstly, there is the yearning for primary groups: family, friends and food. The last seems to be a common element in the social act of meeting with family or friends. Secondly, there is the yearning for the civic culture of the migrants’ country of origin, that is, the social norms (for instance, ‘how things work’, ‘how people respect you’), the way the civil service deals with citizens, and so forth. Finally, there is culture as such, the yearning for lifestyles (traditions, customs, folklore, celebrations). Each of these three dimensions is more or less important according to different groups of migrants and their social or cultural integration. EU movers that report feeling more yearning for their primary groups are Spanish residents in any country of destination, Italian residents in Germany and Spain, and British and French residents in Spain. When it comes to the longing for lifestyles and traditions, Spanish migrants in any country again show the most yearning for their home, followed by Italian migrants in Germany, France and Britain. In contrast, the lifestyle and way of life in Mediterranean countries are valued positively by German and English movers to the south of the continent, which is in fact one of the main reasons for migrating. A contrast of the means representing the yearning for home civic culture reported by EU movers shows two main aspects. First of all, Italy has the least valued or longed-for civic culture, according to French, British and Germans living in Italy, as well as Italian residents in other countries. The second aspect is the non-transitive nature of the ordered sequence of civic cultures. The British are nostalgic for their civic culture when they live in Germany, Germans when they live in France, and the French when they live in Britain. In this sense, it is not possible to order the desirability of different civic cultures using the comparison established by residents from other countries. However, neither Italian nor Spanish nationals long for their civic culture when they reside in other countries. Different forms of homesickness are also clearly related to cultural and social integration. The higher the social integration is, the lower is the

118

Pioneers of European integration

yearning for the primary groups of the country of origin. However, the higher the social integration, the higher the yearning for the civic culture of origin. This means that having more social contact and knowledge of how everyday life works in the country of destination makes movers critical of that country. In fact, the higher cultural integration is, the lower the yearning for the primary groups and the lifestyles of the home country. Cultural integration thus testifies to a neat psychological detachment from migrants’ roots. Overall, higher levels of integration mean less homesickness, with the exception of the aspects of civic culture we have just discussed. Degrees and forms of homesickness vary also on the basis of individual differences. Thus, the longing for primary groups is slightly higher among movers who expect to return soon to their country of origin (0.06**) and among women (0.08**). The period of time spent in the country of destination also shows an effect. The migrants who have lived there the longest are less affected by this type of homesickness (−0.07**). With regard to civic culture, homesickness is higher among the oldest migrants (0.03*), men (−0.04*), and those who are less satisfied with their life in the country of residence (−0.05**). The longer migrants have lived abroad, the more they yearn for their civic culture (0.05**). The yearning for lifestyles affects more those who expect to return soon (0.08**), especially the younger they are (−0.07**). Therefore, the length of time migrants spend abroad reduces their homesickness regarding family and friends but increases their yearning for social aspects like civic culture or lifestyles.

CONCLUSIONS Living in a different society is the only experience shared by all the populations studied in this book, who differ in many other ways. In spite of their individual diversity, EU movers (like all migrants) have the same, everyday concern: that is, how to adapt to a new social and cultural environment. In other words, they may have followed completely different paths taking them to live abroad, but all of them must adjust to a changed milieu. Their past, present, and especially their future, play an important role in the ways they choose to organize their existence. In many regards, EU movers differ quite notably from more familiar types of immigrants. Taking account of the existing literature on acculturation and integration, we sought to make sense of their distinctively patterned strategies of adjustment by devising two indexes on the basis of a series of attitudinal and behavioural items. These two indexes refer to what we call social and cultural integration. Basically, the first index relies

Living across cultures in a transnational Europe

119

on the nationality of friends and the linguistic competence of respondents; the second on their preference for the country of origin or destination and their wish to return to the home country (or not). Overall, EU movers’ personal networks comprise a larger number of non-nationals than nationals. Their command of the language of the country of destination is fairly good – almost three out of four movers know the language ‘quite well’ or ‘almost as natives’. Still, this was not the case at migration, when 40 per cent of movers did not speak a single word of the host-country language. Even though their linguistic skills were overall higher than those of the general population in the country of origin, knowledge of the COR language is far from being a necessary condition to move within Europe. This, in turn, does not jeopardize their overall sense of well-being. EU movers’ cultural integration – as captured by our indicators – is more nuanced. Movers appear to be happier than stayers, and this happiness with their life is even higher when it goes hand in hand with the conviction of preferring the place where they actually reside over their home country. But, to be more precise, the bulk of respondents skipped the simple and perhaps simplistic alternative between the home and the host country as ‘ideal’ places, preferring to indicate both as plausible and partially agreeable locations where they would spend their lives. Equally, EU movers express homesickness in a quite elaborate form, making sense of different pros and cons of their mobility choice. At the end of the day, all this works to single out EU movers as a discerning transnational population (see also Chapter 8 on their transnational media consumption), trying to capitalize on European citizenship to get the best of both worlds – as much as this is possible.

NOTES 1. Confirmatory factor analysis, with orthogonal and non-orthogonal rotations, has been used. The extraction of two significant, differentiated dimensions has been the result of this analysis. As a second method of analysis, we have used non-metric scaling. The results also show two dimensions, differentiated and significant, and group together the indicators considered to be theoretically representative of each dimension. 2. [**] Correlation is significant at the 0.01 level (bilateral). [*] Correlation is significant at the 0.05 level (bilateral).

6.

More mobile, more European? Free movement and EU identity Nina Rother and Tina M. Nebe

FREE MOVEMENT AND EUROPEAN IDENTITY: WHICH RELATIONSHIP? European citizens who live in an EU country other than their native one come in closer contact with many of the EU’s policies than those who stay at home. These EU movers, as we have called them, can benefit from their French health insurance in Germany, shop with the Euro in a wide range of countries and pay reduced home-student tuition fees at British universities. Movers can experience European integration first-hand, be it at the dinner table with friends in the country of residence, at the workplace, or in everyday interaction in a supermarket or at a bus stop. Experiences related to European Union policies or contact and exchange with other EU citizens – if experienced as positive – may affect pro-European attitudes and identities.1 Movers, in short, might differ from stayers regarding their level of Europeanness. But is this necessarily true? Couldn’t positive experiences and contacts in Italy simply make a German Italophile rather than pro-European? Much has been written about the absence of Europe in the lives of ordinary EU citizens (Shore and Black 1994; Meinhof 2004). Maybe moving from Germany to Italy simply makes movers feel at home in two societies, without them developing a new tier of identification with a supranational entity. Maybe movers shop with the Euro and use the health insurance of their country of origin while resident elsewhere, yet stay unattached to the European Union that claims to have made these things possible. On the other hand, developing a European identity might be a convenient way around the tensions and clashes possibly associated with holding two distinct territorial or national identities. In this chapter, we will adopt a social-psychological approach to answer three interrelated questions. First, do movers position themselves differently towards the European Union than stayers? Do they know more about the EU? Do they have a more positive image of the EU? Do they 120

More mobile, more European?

121

feel more European and less Spanish, Italian or British? Are they more attached to the European Union than those who have stayed in their country of origin (COO)? Second, what kinds of territorial identities do movers hold? Do most movers identify with their country of residence (COR) and with the European Union on top of their COO identity? How many movers assimilate entirely and lose their COO identity? Do movers tend to become bi-cultural, holding COO and COR identities, but feeling no attachment to the European Union? Does a bi-cultural COO and COR identity constitute the best base for Europeanness? And third, which factors are conducive to the development of European identities in movers? What is the impact of education, professional status, language knowledge or contact with COR nationals on the European identities of free movers? Before we provide empirical answers to these questions, we will consider their theoretical underpinnings. Despite enduring criticisms of the European Union’s supposed ‘democratic deficit’ and the lack of identification with the EU among European citizens (Scharpf 1999; Siedentop 2001), ambitious research on the emergence of European identity has continued to appear (Gabel 1998; Duchesne and Frognier 2002; Diez Medrano 2003; Herrmann et al. 2004; Hooghe and Marks 2004; Bruter 2005; Gillespie and Laffan 2006; Checkel and Katzenstein 2009). Other scholars are sceptical of the very concept of European identity, pointing to the often absurd proliferation of ways of talking about identity or operationalizing the concept in research (Brubaker and Cooper 2000; Favell 2005). Yet, whether or not European identity has an empirical base or is merely a matter of Bruxellois propaganda is an empirical question. It is the question this chapter seeks to answer for the population of European internal movers. We ask if these movers have a stronger attachment to the EU compared to stayers as a relative question, independent of absolute levels of European attachment. By defining precisely what it is we are studying, by being explicit about our methodological assumptions and simplifications, and by staying clear of any kind of partisanship, we hope to avoid the ‘European identity trap’. The usage we shall make of the term ‘identity’ for the purpose of this chapter will only contain a small and well-defined fraction of its potentially manifold meanings. We take identity to refer to the contextual and dynamic ‘part of the individual’s self-concept which derives from his knowledge of his membership of a social group (or groups) together with the value and emotional significance attached to that membership’ (Tajfel 1981, 255). In other words, we are not interested in the essence of individuals, but rather in changing perceptions of group membership that are induced by changes in the social context in which the actor moves. Identity then is a dynamic process, not a profound and hidden part of the self, a

122

Pioneers of European integration

definition and redefinition of the boundaries between ‘us’ and ‘them’ in particular situations. It follows that identity is multiple (it differs according to context), relational (only because ‘we’ are not ‘they’, we know who we are) and social (it refers to the group membership of individuals). The actor’s self-defined identity is in constant interaction with the group membership attributed to the actor by others (‘internal’ versus ‘external identification’: Jenkins 2000). Via the act of identification, the group is defined and thus invented or reinvented (Brubaker 2004). European identity then is linked to the awareness of one’s feeling of belonging to an imagined category of ‘Europeans’.2 Note that European identity does not designate that feeling of belonging or attachment itself but rather the awareness of it. While attachment is affective, carried out by what G.H. Mead (1934) calls the ‘I’ (the active part of the Self), identity is reflexive and linked to Mead’s ‘Me’. For example, I can be attached to the fate of the pygmies in Papua New Guinea without identifying with them. A more complex case would be to identify as an Alsatian but not be particularly attached to Alsace. We can say that, all in all, attachment is somewhat easier to achieve than identity. De facto, attachment and identity often overlap even if they are analytically distinct. In most cases, the psychological process of ‘ingroup favouritism’ (Turner 1978) makes us evaluate the ingroup higher than any outgroup and produces some form of affect towards it (or ‘value and emotional significance’ according to Tajfel). For the purpose of this chapter, it is important to remember that identity is more reflexive and hence potentially less widespread than affect, but affect is the best proxy available to measure the complex concept of identity. Much has been written about the different ways in which European identities and national identities can co-exist (Duchesne and Frognier 1995; Marks 1999; Kohli 2000; Citrin and Sides 2004; Risse 2004). The most common argument is that territorial identities are potentially additive: one can be aware of belonging to Wales, the United Kingdom and the European Union without experiencing any conflict between these identities. Herrmann and Brewer (2004, 8) propose three configurations summing up how multiple identities work: 1.

Nested Identities: The ‘Matruska doll’ or ‘onion’ model conceives of identities as co-existing one inside the other. That is, I can simultaneously be Basque, Spanish, European and a global citizen. According to this configuration, all European citizens could hold a European identity on top of their existing territorial identities. This model is praised as something akin to the ‘subsidiarity principle’ on the level of identities; it is particularly widespread in the neofunctional literature.

More mobile, more European?

2.

3.

123

Cross-Cutting Identities: Here, the idea is that some, but not all members of one identity group (say French citizens) also feel they belong to another identity group (say Jewish people). In turn, some members of the latter group (Jewish people) identify with professional categories (such as doctors) rather than with Frenchness. If European identity functions according to this principle, some but not all people in Europe might subscribe to a European identity; Europeanness could then overlap with being Protestant, Catholic or Muslim, male, female, gay, lesbian or straight and so on. Political scientists hold that crosscutting identities are a precondition for democracy (Lijphart 1999); nested identities in contrast do not allow for political opposition to develop. Separate Identities: This final configuration is a special case as the level of analysis is the individual not the group. Although an individual can hold two or more identities without problem, there is no group that shares both of these exact two identities. Imagine an Italian teacher in France whose friends are all either teachers or Italian but not both. In this example, there is no cross-cutting group of Italian teachers. If Europeanness took this form, it would not overlap with national identities.

All in all, the debate about multiple territorial identities leaves little room for potential conflict between two or more identities. Research has consistently shown that identities concerning different levels of territorial attachment do not clash with one another; European, national, regional and local identities are not experienced as being in competition. However, little is known about possible inconsistencies between two or more territorial identities that operate on the same level (for example, Tuscan versus Sicilian identity, German versus British identity). When it comes to the territorial identities of internal migrants in the European Union, it is precisely this problem that arises: how do migrants go about feeling both ‘German’ and ‘British’? Can two identities on the same territorial level co-exist? A further point is that perceptions of reference and membership group affiliation are altered when the social context changes. If a revolution occurs in my country, I have to redefine who I am with and position myself and my group vis-à-vis the new set-up. Likewise, if I move from one context to another, I have to adjust to a changed environment and find my place and my ingroup within it. This process of adaptation and redefinition is known as ‘psychological acculturation’ (Graves 1967). It can be understood as a kind of resocialisation as it generally entails not only identity changes but also modifications in attitudes and values as well as the acquisition of new social skills and norms (Berry 1992 and 1997). Adapting Berry’s (1997, 10)

124

Pioneers of European integration

Table 6.1

Four ideal types of migrant identities Are the cultural characteristics and identities of the country of origin maintained? Yes

Are the cultural characteristics and identities of the country of residence taken on?

No

Yes

Hybridization: Bi-cultural identities

Assimilation: COR identity only

No

Segregation: COO identity only

Marginalization: Loss of identity

famous model of interethnic integration to the question of migrant identities, four ideal-typical scenarios arise (see also the use of this typology by Alaminos and Santacreu in Chapter 5). As ideal-types, they momentarily fix otherwise fluid and changing identities for the sake of clarity. In everyday life, a migrant can hold both a COO and a COR identity in the workplace but experience strong COO identification when watching a football match. Likewise, a progression over time is probable: we can expect COO identification to decrease over time in most cases but gain momentum in certain crucial periods (for example, homesickness in the first months, on Christmas day, in hard times and so on). To simplify matters analytically, let us consider the ideal-types shown in Table 6.1. As this table shows, migration within the EU can lead to four distinct responses regarding the migrant’s territorial identity. Far from mirroring the consensual additive models found with EU, national and regional identification, conflict is an option when it comes to two national identities. Here, we have to distinguish between identities that are experienced: 1.

2. 3.

in terms of conflict, as a zero-sum game, that is, ‘the more French I become, the less Italian I am’ or ‘the more Italian I stay, the less French I can become’ (top right and bottom left); as additive and non-conflicting, that is, leading to bi-cultural French and Italian identities (top left cell); as losing a sense of belonging through the migratory experience (bottom right cell).

Let us look at the likelihood of conflicting and non-conflicting identities in turn. The foundations of national identities have always been forged in opposition to the Other (often the neighbours) throughout European history. It is likely that such oppositional thinking still exists among some

More mobile, more European?

125

Europeans despite the existence of the EU. Now imagine it were part of my Spanish identity ‘not to be French’. I eat different food, I speak a different language, my culture is distinct and all my friends are Spanish. By moving to France, a conflict is built up between my Spanish identity and my everyday life where I have to interact with French society, at least to some degree (speak French, eat French food and so on). Thus, segregation and assimilation of identities occur when the group membership proposed by the country of residence is in dissonance with the values, attitudes and identities the actor has been socialized into in his/her country of origin. According to Festinger’s Cognitive Dissonance Theory, individuals suffer from unpleasant psychological tension (‘dissonance’) when two pieces of knowledge or ‘cognitions’ are experienced as discrepant (Festinger 1954): How can I be Spanish (hence ‘not French’) yet behave as if I were French? People prefer cognitions that fit together and dislike dissonant cognitions. Indeed, psychological dissonance has drive-like properties that are much like those of hunger and thirst: the actor wants to get rid of it. One way of overcoming dissonance is to change cognitions: a migrant might simply deny his/ her COR values, identities and attitudes and harden his/her position on the COO cognitions. In this case, the migrant would feel Spanish only and reject everything French – leading to a segregation of identities. The same scenario is possible with COO values, identities and attitudes being rejected and only COR cognitions being kept – an assimilation of identities. On the other hand, it is also possible that the mover does not experience conflict between the COO and COR identities. In that case, the actor acculturates to the country of residence while keeping the links to the country of origin – a hybridization. The migrant thus develops a new, bicultural, mixed identity (for example, Spanish–French). Both Spanishness and Frenchness are now part of the individual’s self-concept, s/he is aware of his/her identification with French and Spanish people alike and attaches emotional significance to both memberships. The question then becomes: under which conditions are movers likely not to experience cognitive dissonance between the COR and the COO identities? Here, two aspects are relevant. On the one hand, the availability of and access to COR identity; on the other, categorical markers having to do with the migrant herself. First, according to Festinger, dissonance increases as the degree of discrepancy among cognitions increases and as the number of discrepant cognitions increases. In other words, if COR and COO identities are experienced as rather different (say, Spanish versus British as opposed to Spanish versus Italian), dissonance is higher and hybrid bi-cultural identities are less likely to occur. Furthermore, sociodemographic markers that can be expected to facilitate the hybridization of COO and COR identities are: knowledge of the COR language; sustained, non-hierarchical contact

126

Pioneers of European integration

with COR citizens (for example, partners and friends, also colleagues and neighbours); level of education (linked to the ability to see overarching similarities); and the year of migration (COO identification is likely to decrease over time) (Recchi and Nebe 2003, 16–17). In sum, we suspect that conflict is likely to occur where (perceived) differences between the COR and the COO countries are great, and where the migrant has arrived only recently, lacks contact with the country of residence and its citizens, is poorly educated and lacks COR language skills. These two configurations regarding the interplay between national identities – the conflict and the hybridisation models – open up two very distinct possibilities for the creation of a European ‘third layer’ of identification. Another possibility is one in which the experience of two cultures leads to a loss of identity with both – a marginalization of the individual from both societies. This completes the typology of possible psychological outcomes. Of special interest to us, though, is the possibility of an emergent Europeanness. If heightened levels of European identity in movers are indeed connected with the experience of moving from one European country to another, such Europeanness must be connected with the development (or transformation) of the COR identity. The (potential for a) COR identity is thus what distinguishes stayers from movers. We will argue that there are two potential paths that lead to the development of a heightened European identity in movers. According to the cognitive dissonance model, the COR identity is in conflict with the COO identity, and one of the two or both must be transformed to develop a European identity. In contrast, the interculturation model holds where bi-cultural COO1COR identities have been accepted. Here, Europeanness can emerge as a tertiary identity that encompasses both national identifications. Let us look at the models in some more depth. The cognitive dissonance model proposes to resolve or reduce the incompatibility between a migrant’s COO and COR identities by adding a new cognition (Festinger 1954). In other words, when the clash between ‘feeling Spanish’ and ‘feeling French’ becomes unbearable, a new cognition, ‘feeling European’, can alter the relative weight of what it means to be Spanish versus. French. This process is different from the segregation/ assimilation model (see above) where one of the two or more dissonant cognitions were simply denied. Where dissonant identities are however transformed, several distinct outcomes are possible: 1.

European identity 1 COO or COR identity: When a European identity is added to the COR and COO identities and is given a higher relative weight than either of them, the degree of dissonance experienced by

More mobile, more European?

2.

127

the individual may be reduced. This is so because European identity does not clash with national identities as it is inclusive by definition and constructed on a different level of abstraction. If the European identity is more important than the COR identity, the individual will experience a feeling of belonging to the COO and to Europe. If the European identity is more important than the COO identity, the individual will chose the COR and European identities. European identity only: It is theoretically possible that European identity is used to reduce the valence of both the COO and the COR identities. The two discrepant cognitions are weakened until cognitive dissonance is no longer experienced. The newly-formed European identity can then be said to have replaced both national identities.

According to the interculturation model, tertiary multi-cultural identities are developed on the basis of existing positive relationships with both national cultures when the individual does not have to choose between them (LaFromboise et al. 1993, 131). The concept of interculturation has been developed in order to understand the ethnogenesis of new cultural identities that cannot be explained with bi-directional models (such as the one proposed by Berry 1997). Examples for such tri-directional or tripartite identities are ‘Chicano’ or ‘Asian American’ identities in the USA that are based on bi-cultural Mexican/Asian and US-American identities, supplemented by an overarching ‘umbrella identity’ that encompasses both national identities. In other words, people who already hold hybrid, bi-cultural identities will find it easier to open up to further tiers of identification. Identification here is a cumulative enterprise, a matter of learning: the individual feels part of several universes and finds it easy to add on others because (perceived) conflict is absent. We are then left with two questions that ultimately require empirical answers but can be raised here. First, in which cases do bi-cultural, nonconflicting identities take on the extra European tier and in which cases do they simply stay bi-cultural? Second, which national identity (COR or COO) is likely to prevail in the case of conflict? Which identity is more likely to be rejected or replaced by an EU identity when cognitive dissonance arises? Scenarios that might lead to interculturation and the development of a European identity could be: (a) a migration history that involves several European countries or; (b) having a partner from an EU country other than one’s own and living in a third EU country. While a bi-cultural British–Italian identity is probably still manageable, a quatri-cultural British–Italian–Spanish–Czech identity might be more demanding. Here, an EU mover might simply prefer to refer to herself as European. Regarding the relative strength of COO versus COR identities, we

128

Pioneers of European integration

suspect that the COR effect will be overall weaker than the COO effect. This means that: (a) where an identity clash occurs between two national identities, the COR identity is more likely to disappear than the COO identity; and that (b) in terms of attitudes towards the EU, movers’ attitudes will be similar to stayers’ attitudes in the country of origin, not those in the country of residence; that is, Italians in Germany will be more similar to Italians in Italy than to Germans in Germany. In other words, primary socialization can be expected to have a stronger impact on identities than psychological acculturation as a consequence of the migration experience. Where one comes from will be more pertinent than newly experienced identities, even if these are explicitly chosen by intraEU migrants. However, once again, the characteristics of the country of residence and its migration regime will have an impact here. Concretely, the strength of the COR effect will depend on: (a) the Europe-orientation of the country of residence (that is, moving to a pro-European country makes one more highly Europeanized, moving to a country with higher levels of Euroscepticism decreases feelings of belonging to the group of Europeans); and (b) the assimilation pressure of the receiving society (that is, where assimilation is fostered, COR effects are stronger, where segregation is tolerated, COR effects are weaker). Bringing together all the models reviewed in this section, we are therefore left with eight options of identity development for the internal EU migrant (Table 6.2). Essentially, these are equivalent to Berry’s categories with an additional layer added to them, that is, whether or not the migrant has also developed a European identity. The -ing form is employed to signal our dynamic and relational reading of identity as ‘identification’. Our hypothesis is that most intra-EU migrants will be found in the first three categories (integrating European, assimilating European, self-segregating Table 6.2

Eight identity outcomes of internal migration in the European Union COO not attached 1 COR attached

COO attached 1 COR not attached

COO 1 COR not attached

EU attached Integrating European

Assimilating European

Self-segregating European

Selfmarginalising European

Not EU attached

Assimilating non-European

Self-segregating non-European

Selfmarginalising non-European

COO 1 COR attached

Integrating non-European

More mobile, more European?

COO-identity

COR-identity

Assimilating non-European

Self-segregating European Self-segregating Non-European

COO-identity

Integrating European

COR-identity

Integrating Non-European

Assimilating European

Self-marginalising European

Figure 6.1

129

Self-marginalising non-European

Models of European identification depending on conflict or compatibility between COO and COR identities

European). The integrating European is the person with bi-cultural identities who develops an interculturated European identity in the absence of any tension between COO and COR identification. The assimilating European is the migrant who experiences cognitive dissonance between the COO and the COR identity and resolves it by adding a new non-conflicting cognition (European identity) to ‘swallow’ the COO identity. The opposite scenario (the COR identity is absorbed by the newly-developed European identity) is the case for self-segregating Europeans (Figure 6.1).

DIFFERENCES IN EUROPEANNESS AND TERRITORIAL IDENTITIES BETWEEN MOVERS AND STAYERS We will now try to answer the first of our three questions: whether movers position themselves differently towards the European Union and their countries of origin and residence than stayers do. We will therefore compare stayers’ and movers’ attitudes towards the EU and their territorial attachments. While we will use the EIMSS to obtain measures for movers, for stayers here we refer to Eurobarometer results.3 Attitudes Towards the EU The first important variable measuring attitudes towards the EU is the image the EU has in the eyes of respondents.4 Figure 6.2 shows how many movers and stayers indicated that they associate the EU with something positive, allowing for different levels of comparison. First, we note that movers have a far more positive image of the EU than stayers. Movers more often hold ‘very positive’ images of the EU, regardless of their country of

130 40 35

Pioneers of European integration

Movers - COR Movers - COO Stayers (EB)

30 Average movers

25 20 15 10

Average stayers

5 0 Germany

Italy

Spain

France

Britain

Notes: The graph must be read as follows: the left bar refers to movers living in Germany, the central bar to Germans living abroad, and the right bar to Germans living in Germany (stayers). Source:

EB 61.0 and EIMSS, N 5 10 901

Figure 6.2

EU-image: ‘very positive’ by COO and COR (%)

origin and their country of residence. This pattern holds for each and every country as well as for the category ‘positive image’ (not shown here). Second, out of all groups of stayers, Spanish and Italians have the most positive image of the EU while Germans have the least positive image. By their country of origin (COO), French and Spanish movers think most highly of the EU. British movers least often associate the EU with something positive. Looking at movers by country of residence (COR), it is movers to Spain that have the most positive image of the EU. These very impressive differences between stayers and movers also apply to self-perceived knowledge of the EU: movers overall and in every country say they know more about the EU than the respective group of stayers.5 Either because they are a self-selected group (that is, more informed people are more likely to move) or, more plausibly, because moving and the greater exposure to the EU and its policies make them more knowledgeable, movers are more aware of the role of the EU and its institutions. Territorial Identities and Attachments When it comes to territorial identities and the feeling of national and/or European identity, a slightly different pattern emerges.6 Again, the percentage of movers feeling ‘European only’ is higher than that of stayers. This is true in each country: while about 5 per cent of all stayers feel European

More mobile, more European?

131

40 Movers - COR Movers - COO Stayers (EB)

35 30

Average movers

25 20 15 10

Average stayers 5 0 Germany

Italy

Spain

France

Britain

Notes: The graph must be read as follows: the left bar refers to movers living in Germany, the central bar to Germans living abroad, and the right bar to Germans living in Germany (stayers). Source:

EB 61.0 and EIMSS, N 5 7 859

Figure 6.3

National versus European identity: ‘European only’ by COO and COR (%)

only, almost 30 per cent of all movers do (Figure 6.3). The gap between stayers and movers is especially large among French and Germans: about 25 percentage points. Breaking down our figures by country of residence, the differences are not as large as between countries of origin. Movers going to Germany and Britain feel least often European only (about one in four) in contrast to almost 40 per cent of movers going to Spain. Apart from looking at the percentages of respondents feeling European only, it is also worth having a look at the percentages of respondents feeling that they are a ‘home country national (HCN) only’ (Figure 6.4). The percentage of movers feeling HCN only is smaller in all countries than that of stayers. This is again especially true for British and German stayers and movers. Whereas 62.1 per cent of British stayers say that they feel British only, this is only true for 18.7 per cent of British movers. The question is whether the moving experience makes British feel more European or whether those British who feel European are more likely to leave their country. Unfortunately, only panel data would permit us to adjudicate between these two possibilities. By COO, 20.2 per cent of Italian movers feel Italian only, but only 9.9 per cent of German movers. Hence, it might be assumed that feeling German is a less stable feeling than feeling Italian and can easily be influenced by the moving experience. In fact, by COR, it turns out that movers

132

Pioneers of European integration

65 60 55 50 45 40 35 30 25 20 15 10 5 0

Movers - COR Movers - COO Stayers (EB)

Average stayers

Average movers

Germany

Italy

Spain

France

Britain

Notes: The graph must be read as follows: the left bar refers to movers living in Germany, the central bar to Germans living abroad, and the right bar to Germans living in Germany (stayers). Source:

EB 61.0 and EIMSS, N5 7 859

Figure 6.4

National versus European identity: ‘Home country national only’ by COO and COR (%)

going to France are the least likely to show high nationalist feelings (8.7 per cent); this is a more common experience for intra-EU migrants living in Germany (22.2 per cent). The next battery of questions asks for the respondents’ attachment to their COO and COR and to the EU.7 These questions have the big advantage of not asking for a ranking of single attachments but for a separate rating of each item. They will therefore be central for all the following types of analyses. When comparing EU attachment between stayers and movers, we again find very clearly the heightened Europeanness of all groups of movers (Figure 6.5). Only 10.3 per cent of all stayers taken together feel very attached to the EU whereas 24.5 per cent of all movers taken together feel so. Out of all groups of stayers, British citizens say least often that they feel very attached to the EU (5.0 per cent); Italian (14.4 per cent) and French stayers (13.0 per cent) on the other hand say it most often. These differences still prevail when the category ‘fairly attached’ is considered apart from ‘very attached’. Movers from Britain, which has traditionally posted high levels of Euroscepticism in surveys, show the least attachment to the EU (only 17.5 per cent feel very attached), while the other four groups of movers show about the same level of attachment to the EU. In terms of COR, however, no big deviations across countries can be found, which makes us assume that the effect of the country of origin is somewhat stronger on

More mobile, more European?

133 Movers - COR Movers - COO Stayers (EB)

30 25

Average movers 20 15 Average stayers 10 5 0 Germany

Italy

Spain

France

Britain

Notes: The graph must be read as follows: the left-wing bar refers to movers living in Germany, the central bar to Germans living abroad, and the right-wing bar to Germans living in Germany (stayers). Source:

EB 61.0 and EIMSS, N 5 10 928

Figure 6.5

Feelings of attachment to EU: ‘very attached’ by COO and COR (%)

Europeanness than that of the country of residence – a hypothesis that will later be tested in more detail. Apart from EU attachment, having a look at COO and COR attachment is also worthwhile. As movers have not lived in their country of origin for at least one year, one might assume that stayers – who still live in their country of origin – feel more attached to this country than movers. In general, however, it turns out that the experience of moving does not decrease the level of movers’ attachment to their countries of origin. Out of all groups of stayers, Italians and French are those who feel most attached to their country of origin – 57.9 per cent and 54.1 per cent are very attached to their COO – in contrast to only 39.5 per cent of Spanish stayers (Figure 6.6). But Italian and French movers are very attached to their countries of origin, too. British movers feel least attached to Britain (29.4 per cent), followed by German movers (32.7 per cent). These rather low values are understandable when the migration motives of British and Germans – primarily quality of life (for example, weather and food) – are taken into account. Figure 6.6 further shows that 60.2 per cent of Spanish movers feel very attached to Spain but only 39.5 per cent of Spanish stayers do. The hypothesis that leaving Spain makes Spanish feel even more attached to Spain fits very

134

Pioneers of European integration

65 60 55 50 45 40 35 30 25 20 15 10 5 0

Movers - COR Movers - COO Stayers (EB) Average stayers Average movers

Germany

Italy

Spain

France

Britain

Notes: The graph must be read as follows: the left bar refers to movers living in Germany, the central bar to Germans living abroad, and the right bar to Germans living in Germany (stayers). Source:

EB 61.0 and EIMSS, N 5 11 238

Figure 6.6

Feelings of attachment to COO: ‘very attached’ by COO and COR (%)

well into the stereotype of the proud Spanish, but it is also very possible that Spanish stayers (in contrast to Spanish movers) prefer to indicate a strong regional attachment instead of a national attachment. When looking at COR differences, it is evident that movers going to Germany continue to feel very attached to their country of origin. Germans in Spain feel more attached to Germany than German stayers, which might be due to strong German communities on the Spanish coasts, where most of the German retired movers settle. Apart from attachment to their country of origin, movers were also asked to rate their attachment to their country of residence. However, comparisons between stayers and movers are not easy because the stayers’ country of origin and residence have always been the same. Figure 6.7 shows that, as expected, stayers feel more attached to their country of residence than do movers to their target country with the exception of movers to Spain. A high percentage of British movers in general feel very attached to their COR, whereas only few Italian movers feel so. Breaking figures down by country of residence, it is movers to Spain who feel most attached to Spain (41.6 per cent), whereas only 23.0 per cent of movers going to Britain feel very attached to this country. Movers living in Germany also indicate low feelings of attachment to Germany. Clearly, movers have by far more positive attitudes towards the EU and a higher attachment to the EU than stayers. We also saw that the experience of

More mobile, more European?

135

60 Movers - COR Movers - COO Stayers (EB)

50

Average stayers

40 Average movers

30 20 10 0 Germany

Italy

Spain

France

Britain

Notes: The graph must be read as follows: the left-wing bar refers to movers living in Germany, the central bar to Germans living abroad, and the right-wing bar to Germans living in Germany (stayers). Source:

EB 61.0 and EIMSS, N 5 11 241

Figure 6.7

Feelings of attachment to COR: ‘very attached’ by COO and COR (%)

moving influences attachments to COO and COR – two important moderating variables which themselves influence EU attachment. The question now is whether such a heightened Europeanness can be found among all movers or whether different types of movers exist with regard to their territorial attachments and attitudes towards the EU. For all of these probably existing mover groups, it would then be interesting to find out which variables influence territorial attachments and EU attitudes. In other words, which variables affect whether a mover feels attached to COO, COR and the EU? Is it more (or only) context effects like the socialization effect of the country of origin or the influence of attitudes in the new country of residence? Or is it more (or only) individual variables like gender, age, year of migration, marital status or daily contacts? This is what we will now analyse, changing our focus from the comparison of stayers and movers to movers only.

A TYPOLOGY OF TERRITORIAL IDENTITIES OF MOBILE EUROPEANS As elaborated above and summarized in Table 6.2, our theoretical considerations led to identifying eight possible different identity outcomes

136

Pioneers of European integration

Table 6.3

Eight identity outcomes of internal migration in the European Union (%) COO 1 COR attached

EU attached 49.7 [G1] Not EU attached 17.7 [G2] Total 67.5 Source:

COR attached 1 COO not attached

COO attached 1 COR not attached

9.8 [G3] 5.4 [G4] 15.1

6.9 [G5] 5.4 [G6] 12.3

COO 1 Total COR not attached 2.5 [G7] 2.6 [G8] 5.1

68.9 31.1 100.0

EIMSS, N 5 4 458

for internal movers. From an empirical point of view, however, some of these possible configurations might not actually occur in movers. We will now try to answer our second question: the kinds of territorial identities that movers hold. To address this question, we will again use the EIMSS dataset and analyse movers’ ratings of their attachment to their COO, COR and the EU, using attachment as a proxy for identity. As the question battery for COO, COR and EU attachment offers only four alternative answers and no middle category, a split into ‘attached’ and ‘not attached’ movers can be realized quite easily by assigning ‘very’ and ‘fairly’ attached movers to the ‘attached’ category and ‘not very’ and ‘not at all’ attached movers to the ‘non-attached’ category. Table 6.3 shows the resulting empirical distribution of this typology. It turns out that about two-thirds of our movers feel attached to both COO and COR and thus have developed (at least) bi-cultural identities without feeling conflicts emerging from this multiple identity (G11G2). However, about 27 per cent of movers (15.1 per cent and 12.3 per cent respectively) did feel an identity conflict and dissonance when faced with two different national identities. How did they resolve this conflict? While one group (G31G4) assimilates to the COR identity and drops their COO attachment, another group (G51G6) rejects the COR identity and sticks to COO attachment only. Interestingly, even if the difference between the two groups is small, more people adapt their identity to their entourage rather than thinking in terms of their ‘roots’. This finding clearly contradicts our hypothesis that COO effects would be larger than COR effects. Only 5.1 per cent of movers feel attached to neither COO nor COR (G71G8). Marginalization thus seems to apply to only a small minority. Table 6.3 also speaks to the interculturation hypothesis cited above: the development of EU attachment is more common when movers do not have to chose between COO and COR attachments but feel at ease with both of them. Almost half of our movers belong to the group of

More mobile, more European?

137

integrating Europeans (G1) and feel attached to COO, COR and the EU. Europeanness for them thus serves as an overarching umbrella identity that encompasses both national identities. It is therefore very likely that the heightened Europeanness of movers in comparison to stayers is indeed due to the confrontation with a second (perhaps third or fourth . . .) national identity. We will develop this in the next section. The second most common identity type is ‘integrating non-Europeans’ (G2): 17.7 per cent of movers prefer to feel attached to both COO and COR, but not to the EU. The question posed in the theory section comes into play here: why do some bi-cultural movers develop a sense of Europeanness while others do not? In our sample, the 5.4 per cent who were ‘assimilating non-Europeans’ (G4) and the 5.4 per cent ‘self-segregating non-Europeans’ (G6) resolved their feeling of identity conflict and cognitive dissonance by reducing the valence of either the COO or the COR identity and only felt attached to either the COR or the COO. Another 9.8 per cent were ‘assimilating Europeans’ (G3) and 6.9 per cent ‘self-segregating Europeans’ (G5); they reduced their feeling of dissonance by adding a new cognition, namely EU attachment. As types 3 and 5 – movers with EU attachment – are more frequent than types 4 and 6 – movers without EU attachment – it seems to be easier for movers to reduce dissonance by adding new cognitions than by reducing the weight of existing ones. But this point is for experimental psychologists to develop further. Overall, in any case, it is most likely for internal movers in the European Union to develop a non-conflicting bi-directional identity with EU attachment as an additional umbrella identity.

ACCOUNTING FOR DIFFERENT TERRITORIAL IDENTITIES AMONG MOVERS It remains to determine which factors are conducive to the development of EU attachment in movers. In other words, which contextual and individual factors make it more likely that a mover will develop one out of the eight theoretically possible identity outcomes? Do our hypotheses hold in different social contexts: that is, do variables like the COO’s Europeanness and individual variables like education or language knowledge influence movers’ territorial identities? In order to single out important predictors for affiliation to one of the eight types, we conducted a multinomial regression analysis with ‘selfsegregating non-Europeans’ (type 6) as the baseline against which all other types will be contrasted. Using this type of analysis, we will be able

138

Pioneers of European integration

to conclude which independent variables differentiate best between each type and type 6. Concerning the list of independent variables, three classes can be distinguished. The first class consists of contextual variables – the country of origin and the country of residence. For both variables, Spain serves as a baseline. As typical socio demographic variables, we included age at migration date, length of stay, gender with female being the baseline, education level with all degrees less than a university degree as baseline, and marital status with no partner as baseline.8 The third class of variables consists of individual variables related to the mover’s character or behaviour: the mover’s migration motive (that is, work, study, family and love reasons, quality of life, or others), past migration experience (whether the respondent has lived in the COR or a third country prior to the last move), daily contacts (whether she has friends from the COO, the COR or third countries), and language knowledge (good or bad, upon arrival in the COR and at the time of interview). Table 6.4 shows the B-coefficients and their significance as well as the pseudo-R-square for the multinomial regression model. Although the pseudo-R-square remains relatively low and suggests that other variables such as the mover’s personality may also play a role in predicting territorial identities, several important predictors can be singled out. Table 6.4 thus shows what makes movers most likely to belong to one of the eight types and not to another. This needs to be done by contrasting the different types with a baseline, that is, we learn how type 1 to type 8 movers differ from type 6 movers, which constitute the baseline. As we can see, this baseline category consists mainly of recently arrived labour migrants who feel attached to their COO but not to the COR and to the EU. They can therefore be seen as the most traditional type of immigrants, less likely to experience the distinctive effects of being a free-moving European. Each of the eight types of movers will be now described in detail. We correspondingly offer eight labels to identify these types, which might be compared and contrasted to the four basic clusters identified earlier in the book (see Chapter 2). Type 1 (Integrating Europeans) – the Euromasters ‘Integrating Europeans’ (49.7 per cent of our sample) are not only the most frequent but also the most ‘ideal’ – from a pro-European integration perspective – kind of internal migrants in the European Union: they feel attached to their country of origin, to their country of residence and to the European Union without experiencing any conflict between these identities whatsoever. They are model Europeans who seem to live the slogan ‘unity in diversity’. Understanding what distinguishes type 1 movers from

139

Type 2 vs. type 6

Type 3 vs. type 6

0.295

20.069 20.176 20.276 0.468 – 20.412* 0.031 0.154 0.542*** 20.042 0.358 0.513*

20.142 20.045 20.403 0.264 – 20.123 20.011 0.256 0.704*** 0.140 0.515* 0.575*

0.238

0.326

0.438*

20.461 20.232 20.700* 0.649* – 0.110 20.014 20.184 1.203*** 0.143 0.140 0.667**

0.515

20.014

0.009 20.002 0.001 0.035*** 0.034*** 0.026* 0.011 20.161 0.002 0.511** 0.155 0.378* 0.194 0.030 20.033

Type 1 vs. type 6

20.126 0.079 20.493 0.765* – 20.335 0.009 20.002 1.213*** 20.182 0.298 0.104

0.639

0.181

20.012 0.033** 20.364 20.051 0.032

Type 4 vs. type 6

20.113 20.219 20.161 0.218 – 0.259 20.154 0.296 20.236 0.183 0.401 0.520*

0.256

0.182

20.002 0.001 0.142 0.579** 20.034

Type 5 vs. type 6

0.457

0.119 0.581 0.467 1.036* – 20.164 0.169 0.059 0.556* 0.143 0.171 0.567

0.470 20.057 0.335 0.849 – 0.044 20.358 20.053 0.874** 0.425 0.415 0.562

20.302

20.003 0.010 0.430 0.173 20.319

0.900*

0.346

20.007 20.006 0.770** 0.454 20.093

Type 7 vs. Type 8 vs. type 6 type 6

Characterization of identity types 1 (integrating Europeans) to 8 (self-marginalising non-Europeans) in contrast to type 6 (self-segregating non-Europeans): multinomial regression analysis (B coefficients)

Age at migration Years of stay Gender (0 5 female) University diploma Partner from COO (0 5 no partner) Partner from COR (0 5 no partner) Partner from 3rd country (0 5 no partner) Migration motivation Work Study Personal reasons Quality of life Miscellaneous Lived in 3rd country before Lived in COR before Several friends from COO Several friends from CO R Several friends from other countries Language knowledge at migration date Language knowledge now

Table 6.4

140

(continued)

20.120 20.866** 20.270 – 0.549 12 918.877 182 0.201

20.036 0.015 20.099 –

Type 1 vs. type 6

Source:

EIMSS, N 5 4 458

Type 3 vs. type 6

20.105 20.901** 20.418 – 21.415

0.306 0.608* 1.079*** 1.733*** 0.479* 20.043 – –

Type 2 vs. type 6

20.390 20.733* 20.027 – 20.310

Notes: Reference category for dependent variable: type 6. * Significant at 5%; ** Significant at 1%; *** Significant at 0.1%

Country of Origin France Britain Italy Spain Country of Residence France Britain Italy Spain Constant 22 Log Likelihood Degrees of freedom Pseudo R- square (Cox and Snell)

Table 6.4 Type 5 vs. type 6

20.495 20.599 20.425 – 21.129

0.787 1.208 0.756 – 21.414

0.799* 0.019 2.040*** 20.256 0.534 0.128 – –

Type 4 vs. type 6

0.079 20.351 0.107 – 23.768

1.184** 1.667*** 0.546 –

20.085 20.129 20.096 – 22.707

0.882* 1.206** 0.284 –

Type 7 vs. Type 8 vs. type 6 type 6

More mobile, more European?

141

other kinds of internal EU migrants is central to our analyses. Why do some movers construct identities according to this politically desirable scheme COO1COR1EU? And why do others not develop the EU tier or reject one of the national identities? Several variables help to explain the emergence of this tripartite attachment: out of the sociodemographic variables, the length of stay in the COR, the education level and having a partner from the COR have a highly significant impact and make type 1 differ from type 6 movers. We can assume that the longer the mover has already lived in the COR, the more s/he feels attached to it – especially when the partner is a COR national. Holding a university degree may increase a person’s ability to see what unites the countries of Europe and what the abstract EU construct is all about and therefore foster the development of an umbrella identity. Feeling attached to COO, COR and Europe in comparison to the baseline category is furthermore strengthened when the mover has several friends from the COR and when s/he had and has a good command of the COR’s language. Having COR friends and being able to communicate right from the start seem to be key for taking on the COR identity. The COO attachment however is not reduced by this strong link to the COR society; these movers still have several friends from the COO (although the coefficient of the analysis does not reach a significant level). On the contextual level, we clearly see an impact of the country of origin as well as of the country of residence. In comparison to our baseline category type 6, type 1 movers are more likely to be from Germany than from Spain. Moving to Britain and Germany (in comparison to moving to Spain) on the other hand has a negative effect on the EU attachment of movers – a clear COR effect as both countries are typically seen as Eurosceptic. Apart from territorial attachments, it would also be interesting to know what attitudes towards the EU are held by type 1 movers, as well as if they can be regarded as ‘successful movers’ in terms of a high life satisfaction. Integrating Europeans have the second highest knowledge of the EU as well as the second most positive image of the EU of all of the eight types of movers. Regarding their life satisfaction, however, type 1 movers come only third – although the mean differences here are relatively small and the general level of life satisfaction is quite high. The following quote, taken from qualitative interviews conducted with internal movers, illustrates how Euromasters feel about their national and European identifications. It is from a French mover in Germany. Interviewer:

Do you sometimes think about where you belong to, I mean, what your national identity is?

142 Jean: Interviewer: Jean:

Interviewer: Jean: Interviewer: Jean:

Pioneers of European integration Yes, often. [quickly and spontaneously] And do you think you are more French or . . . I’m more French. But sometimes I think, uh, could I live at all without any involvement of Germany? I mean, um . . . When I was in France for that short period, I missed Germany a lot, I mean, I missed it. I tried, uh, to find a job, in which I was able to talk German, in which I have contact with Germany. And I think I would miss that most definitely. And do you feel a little bit ‘German’? [Pause] Or . . . [Jean interrupts] Some parts . . . That is some parts of Germany I simply like. And do you also feel as a European, sometimes? Yes, Hmhm. So, I, I could, if I was offered a great job, in Italy, I would go at once.

INTEGRATING EUROPEANS . . . (a) have lived in COR for a longer time (b) are highly educated (c) have COR and COO friends (d) have a COR partner (e) spoke and speak the COR language (f) typically left Germany and moved to Spain (g) are not movers to Germany and Britain (h) know a lot about the EU and have a very positive image of the EU ➔ the Euromasters

Type 2 (Integrating Non-Europeans) – the Bi-cultural Movers The second type of movers are the ‘integrating non-Europeans’ (17.7 per cent of all movers) who feel attached to their COO and their COR but not to the EU. Regarding their sociodemographic profile, integrating non-Europeans differ from type 6 movers only in terms of their length of stay in the COR. Movers are more likely to belong to type 2 than to type 6, the longer they have already lived in the host country. When it comes to the character and behaviour of type 2 movers, we see that the variable ‘lived in a third country before’ has a strong negative impact. Thus movers who have never lived in a third country before are more likely to develop attachments to two countries only. Type 2 movers also tend to be more integrated into the COR in terms of having friends from that country. When it comes to the impact of contextual factors on the affiliation to type

More mobile, more European?

143

2 versus type 6, it turns out that integrating non-Europeans are mainly British, German and Italian movers. Hence, an ‘early socialization’ effect of the two more Eurosceptical COOs (Britain and Germany) is again salient. Again, as with type 1 movers, type 2 movers are likely not to settle in Britain and are more likely to settle in Spain. As may have been expected, the integrating Non-Europeans’ attitudes towards the EU are less positive than those of integrating Europeans. Note also that the life satisfaction of type 2 movers – which differ from type 1 movers only by the fact that the latter do not feel attached to the EU – is somewhat smaller. Charles:

I’ll never be a perfect German or a . . . Perfect meaning hundred percent German. I’ll never speak Bavarian [smiles], like my colleagues here are speaking [smiles]. And, ah, you always keep your accent and your – way of thinking. On the other hand, ahm, you notice, you’re not as French as your friends who stayed home. – But . . . You notice it and it . . ., it either it bothers you or it doesn’t. And after that I would say, for me, the question is: Am I happy or not? And if you say I’m happy, because I have a job, I have, a – home . . . I have – everything I need. Then it doesn’t bother you. Interviewer: So would you say that you are first, European and then French and then German? Or if you . . . would put those three words in an order, what would you say? Or are they all on the same level? Is this possible, do you know that? Charles: Ah . . . [pause]. So . . . Somehow you . . . I do have the impression is a mix between French and German. Or you would say that what you have to start, you know, think, ah, European. European – no. I mean I’ve never lived in Spain or lived in Italy or somewhere else. I mean that’s what really, ah, ‘Euro(!)-pean’ meaning – than everywhere. In fact I could only really compare those two countries.

NON-INTEGRATING EUROPEANS . . . (a) have lived in their COR for a longer time (b) have not lived in a third country (c) have COR and COO friends (d) speak the COR language (e) typically come from Germany, Britain and Italy (f) are not movers to Britain (g) have the least positive image of the EU ➔ the Bi-Cultural Movers

144

Pioneers of European integration

Type 3 (Assimilating Europeans) – the Lifestyle Movers Almost 10 per cent of all internal movers feel attached to the COR and the EU but not to their COO anymore. Basically, they have taken on the COR identity and transformed their clashing COO identity into a European one, according to the dissonance model. These movers are also different from type 6 movers regarding their length of stay and their education level. Type 3 movers are more likely to have lived longer in the new country and to have a university diploma. Regarding their migration motives, friendships and language knowledge, ‘assimilating Europeans’ are also different from type 6 movers. Type 3 movers mainly moved to the COR to improve their quality of life, which results in the combination of a high COR attachment and low COO attachment as movers might want to forget the unpleasant COO and feel happy where they live. They probably also feel attached to the EU as the EU made this win–win migration possible. Thus assimilating Europeans reduce their identity conflict by giving up their COO attachment and adding a new attachment to the EU. Type 3 movers seem to comprise both retirement migrants as well as young, adventure-seeking pioneers. Regarding their friendships, a clear picture arises: the COO attachment stays low as assimilating Europeans tend to have only a few friends from the country they left. But then again the fact that they have several friends from the COR explains the high COR attachment. Rather unsurprisingly, type 3 movers have a quite good command of the COR language now in comparison to type 6. Again this good language knowledge might thus be a reason for a strong COR attachment. When looking at the contextual variables explaining the affiliation to type 3, we see a strong effect of the country of origin: it is either Germans or British abroad in contrast to Spanish who tend to assimilate to the COR, but also feel European. Assimilating Europeans have very positive attitudes towards the EU, their self-perceived EU knowledge comes in second place and their EU image in third place. But apart from having positive attitudes towards the EU, type 3 movers are also most satisfied with their life as a whole. Hence, having gone through the identity conflict by resolving it with a reduction of COO attachment and adding EU attachment seems to make movers most satisfied with their present life. Interviewer: Guillermo:

Does it happen to you to think of yourself as Italian? Of course! I am . . . I’ve been here for almost nine years. I don’t know whether I will ever take the Italian citizenship because I am not interested in it . . . I don’t believe in nationalities, I think they are an add-on, a matter of chance. Whether you are

More mobile, more European?

Interviewer: Guillermo:

Interviewer: Guillermo:

Interviewer: Guillermo:

145

French, Spanish, Italian or English doesn’t depend on you. I would like to have, one day, a European identity card or a European passport. That’s what I would like! What do you think you share with other Spanish? The lived experiences, childhood, the cultural background . . . Up to a certain point, obviously. For other things, I don’t think it depends on nationality to be able to share something with another person; indeed, the only thing I share is the imprint of the first years, the childhood, we’ve had similar trajectories, that’s the only thing. What in contrast do you think you share with Italians? With Italians . . . Everyday life, from politics to economic life. I even get angry when certain things happen in Italy, I am puffed up . . . It makes me slightly bitter to not be able to participate in political elections, because for better or worse, I am involved in Italian politics in the sense that whatever the Italian state decides will concern me . . . Hence, everyday life: public transport that doesn’t work, garbage that maybe doesn’t get picked up every day . . . What do you think you have in common with other Europeans? This culture that belongs to all of us. The European institutions . . . I’ve always been a Europe-fan, that’s why I always identify a lot with the institutions. The fact of having a fixed reference point in Brussels that protects you from injustice . . . The fact of having for better or worse the same problems in everyday life, the same money . . . And then this cultural process that influenced everyone, from Greece via Rome to the European Enlightenment of the 18th century.

ASSIMILATING EUROPEANS. . . (a) are highly educated (b) tend to be older at migration date (c) have many COR friends (d) speak the COR language (e) moved for quality of life reasons (r) typically come from France, Germany and Britain (t) typically moved to Spain (h) are most satisfied with their lives ➔ the Lifestyle Movers

146

Pioneers of European integration

Type 4 (Assimilating Non-Europeans) – the Carefree Movers ‘Assimilating non-Europeans’ constitute our fourth type of internal movers and comprise only 5.4 per cent of all movers. This type can be characterized by a high COR attachment and low COO and EU attachments. Thus movers of this type have dealt with their feeling of cognitive dissonance by assimilating more or less completely to the new country without adding a new supranational identity. The sociodemographic profile of type 4 movers again shows that, in contrast to type 6 movers, they have stayed longer in the COR, which explains their attachment to it. Although not significant, we can see a tendency for type 4 movers to be younger and less educated than type 6 movers. It seems that leaving one’s COO relatively early also means that a COO attachment has not yet been fixed and can easily be given up, especially when the mover is perhaps not familiar with other means of dealing with cognitive dissonance due to the lack of a higher education. To give in to external pressure and assimilate into the new context is also the easiest way to deal with identity conflict. It also turns out that assimilating non-Europeans left their COO mainly for a higher quality of life, which again explains a high COR attachment. The same mechanism should be true for the high number or COR friends that type 4 movers have. As for the typical COOs and CORs of type 4 movers, we find a clear COO effect for the British: movers from the other side of the Channel are likely not only to belong to type 3 but also to type 4, so whether a British person develops an EU attachment or not depends on other factors. But movers from Germany and France are also more likely to belong to type 4 than the Spanish. Movers to Germany, on the other hand, are less likely to develop a strong attachment to Germany than movers to Spain.

ASSIMILATING NON-EUROPEANS . . . (a) have lived in COR for a longer time (b) tend to be younger and less educated (c) have many COR friends (d) moved for quality of life reasons (e) typically come from France, Germany and Britain (f) are not movers to Germany (g) are very satisfied with their lives ➔ the Carefree Movers

More mobile, more European?

147

Assimilating non-Europeans hold less positive attitudes to the EU. Their life satisfaction, however, is almost as high as that of assimilating Europeans. Being able to reduce one’s attachment to the COO thus seems to be a crucial factor for a satisfied life without regrets and homesickness and external pressures in the new country. Type 5 (Self-Segregrating Europeans) – the Homesick Movers The fifth type of internal movers was labelled ‘Self-segregating Europeans’ and is formed of 6.9 per cent of our movers. Type 5 consists of movers with a high COO and EU attachment but a low COR attachment. The multinomial regression analysis shows that type 5 movers have a high level of education – as do all other types of migrant identities that include a European dimension. However, type 5 movers do not differ from the baseline, the type 6 movers, regarding their migration motivation or individual variables, the only exception being their current command of the COR language. Although (statistically) not significant, it seems that many self-segregating Europeans have lived in a third country before – an event that probably leads to higher EU attachment as the pool of ‘collected’ European experiences is larger. However, having lived in another EU member state does not foster COR attachment as the new country might be regarded as being just another one on the list (which will perhaps even be expanded later on in their life course). Unsurprisingly, self-segregrating Europeans have very positive attitudes towards the EU, both for EU knowledge and EU image. Hence, they do not differ much from the other European-oriented types of movers. Regarding their life satisfaction, clearer differences from the first four types can be found: movers not attached to the country they live in feel less satisfied with their life as a whole. Alberto: Interviewer: Alberto:

Interviewer: Alberto:

Um, I think, I am, I am a European. I feel extremely . . . I have this European identity. Um, I feel I am a European. An Italian European. That’s all. European but with a link to Italy. How do these things fit together? Um, It’s like a, Matrioska, it’s like Chinese boxes. You have the world, you have Europe, you have Italy, you have Lombardi, you have Como, you have Montrasio. You know, it’s . . . I . . . identity can live together. It’s not, ahm, I don’t see. I think, nationalism is over. It’s not over, but it should be over. Yes. You can be Italian and European. Why not?

148

Pioneers of European integration

SELF-SEGREGATING EUROPEANS . . . (a) are highly educated (b) speak the COR language (c) have COO friends (d) moved for miscellaneous reasons (e) are not movers to Spain (f) hold positive attitudes towards the EU (g) are least satisfied with their lives ➔ the Homesick Movers

Type 6 (Self-Segregating Non-Europeans) – the Recent Labour Migrants The sixth type – ‘Self-segregating non-Europeans’, 5.4 per cent of all movers – feel highly attached to their COO, but not attached to the COR or the EU. This group cannot be described in a statistically detailed way as it served as the baseline category for the multinomial regression analysis. However, when comparing the results for the other five types, it can be assumed that type 6 movers are very likely to have lived in the new country for only a short time. It is possible therefore that these movers simply have not yet had enough time to develop a strong COR attachment – maybe because they still feel homesick and are concerned by that. As for the migration motives, it seems that type 6 movers mainly came for work or personal reasons rather than for lifestyle reasons. Their command of the COR language is quite poor, explaining why their attachment to the COO still prevails. Lastly, type 6 movers have more friends from the COO than any other types of movers, again explaining the low COR attachment. It also turns out that self-segregating non-Europeans, like all other nonEuropean types, have a less positive image of the EU and also show the least knowledge of the EU. Similar to self-segregating Europeans, their life satisfaction is – in relation to the other types – rather low. Feeling attached to the COR and (also but with a smaller impact) to the EU can thus be seen as a relevant component of a high life satisfaction. Guido:

I feel as an Italian in any case. I’ve been, ah, in Italy for 27 years. I’ve . . . I was born there, I grew up. There are also some things that simply belong to my culture. Ahm, but, ah, I have to say, if I say now where I belong to, I say I belong to here, to my, yes, ah, since I . . ., ah, live here in Germany. Not something that I left in Italy.

More mobile, more European?

149

Interviewer: Do you sometimes also feel as a European? Or rather less? Guido: Yes . . . Actually . . . Yes, ‘less’ I would have said. Yes. Also because I have the feeling that you get a bit less of something, that is somehow a ‘European, ah, feeling’, here in Germany. For example I was quite often in Belgium, ahm, in Brussels, where a friend of mine lives. There you get a much better impression of what it means to be European. And therefore . . . Here in Germany it is a bit less, yes.

SELF-SEGREGATING NON-EUROPEANS . . . (a) are less educated (b) did and do not speak the COR language (c) have COO friends (d) moved for work and personal reasons (e) are not movers to Spain (f) know least about the EU (g) are less satisfied with their lives ➔ the Recent Labour Migrants

Type 7 (Self-Marginalising Europeans) – the Cosmopolitans The seventh type of movers consists of individuals who identify neither with their COO nor with the COR, but only with the EU. These movers thus choose a rather strict way to deal with the identification problems resulting from two opposed national identities – they deny both of them. This type was therefore labelled ‘self-marginalising Europeans’ and is the smallest group in our sample as it consists of only 2.5 per cent of all movers. The sociodemographic profile shows that in contrast to all other types discussed before, self-marginalising Europeans are mainly male and have a partner from a third country, which arguably may have weakened the importance of national identities. Interestingly, for this group of proEurope movers, education does not play a significant role, which is in contrast to all the other pro-Europe types. We can also see that no special migration motive is dominant for this group in comparison to the baseline of type 6. The distribution of friends, however, is special for this group as they tend to have no friends from the COO but many from the COR and – although not significant – from third countries as well. Thus it seems that, for this type of movers, personal connections to other countries are quite important. In sum, they look like

150

Pioneers of European integration

transnational migrants. In this sense, self-marginalising movers are not ‘marginalised’ at all. Rather, these cosmopolitans most closely resemble the denationalised ‘Eurostars’ described by Favell (2008a) in his study of free moving professionals in European metropolies. Regarding the national background of type 7 movers, we face a clear North–South difference: cosmopolitans come mainly from the three northern countries but do not necessarily move southwards; they can be found in all of the other countries, underlining that the quality of life in terms of better weather is less important for this group than for types 3 and 4. In line with the picture established by sociodemographic and contextual variables, analyses show that type 7 movers hold the most positive attitudes towards the EU and have the highest knowledge of the EU. We can assume that the concentration on one single identity – the European one – makes movers of this type feel so positive about the EU. When it comes to life satisfaction, however, they can be found somewhere in the middle of all types. Interviewer: Arthur: Interviewer: Arthur: Interviewer: Arthur:

And . . . What are your . . . your thoughts about that? Do you . . . do you think of yourself more of German or more of English or more of . . .? I think of myself as rather displaced. Aha. Do you think of yourself as European? Yes. More. . . more than you think of yourself as either German or English? Yes.

Type 8 (Self-Marginalising Non-Europeans) – the Individualists

SELF-MARGINALISING EUROPEANS . . . (a) are male (b) have COR and third country friends (c) have a partner from a third country (d) typically include movers from France, Germany and Britain (e) exclude movers from Spain (f) can be found in all countries (g) have the most positive image of the EU (h) know most about the EU ➔ the Cosmopolitans

More mobile, more European?

151

The ‘self-marginalising non-Europeans’ constitute the last type of movers. Type 8 movers can be seen as the opposite of type 1 movers as they feel neither attached to their COO nor their COR, nor do they feel attached to the EU. Thus the rule here is to resolve cognitive dissonance by withdrawing from all kinds of national or supranational identification. However, this strategy is not a very common one as this type of movers is only 2.6 per cent of the sample. The analysis of the sociodemographic profile shows that self-marginalising non-Europeans do not differ significantly from the baseline, the type 6 movers. The only exception is a non-significant prevalence of men. Regarding their migration motivation, however, a clear difference arises as type 8 movers leave almost exclusively for quality of life reasons. Like type 7 movers, type 8 movers are also well integrated into the new country as they have a lot of friends from the COR. What is more, the same distribution of origins arises: the bulk of self-marginalizers come from the three northern countries included in our analyses. However, they tend to settle in Spain rather than in any of the other four countries. Out of all non-European movers, type 8 movers are those with the most positive image of the EU and also the highest knowledge of the EU. There is perhaps a rationalised, instrumental understanding of the benefits of the EU, rather than an identity-based one. Taking all of these rather scarce findings together, it can be concluded that no clear pattern can be found in this group but that individual motivations or lifestyles prevail. These individual choices tend to cancel themselves out so that no clear profile of this group can be constructed. Thus, individuality is the common characteristic of this group.

SELF-MARGINALISING NON-EUROPEANS . . . (a) are more likely to be male (b) have COR friends (c) moved for quality of life reasons (d) include movers from France, Germany and Britain (e) include movers to Spain (f) hold relatively positive attitudes towards the EU ➔ the Individualists

CONCLUSIONS In this chapter, we posed the question of what happens to internal EU migrants and their group identification when they move from one

152

Pioneers of European integration

country to another. In order to study the possible mixing or conflict between two or more identities due to a change in context (that is, moving away from one’s country of origin to a country of residence), two relational identity models were constructed. Model one assumes that cognitive dissonance is likely to occur wherever two national identities have to be brought into unison, especially if these two identities are dissimilar and if the actor is not positively disposed towards changes of residence (for example, little COR language skills, few COR friends). The second model argues that bi-cultural identities will emerge where assimilation pressures are not too high (loss of COO identity) and not too low (segregation), and where the migrant interacts with the country of residence. Both identity models (conflict versus consonance) present different paths to the development of European identities. Where conflict is the starting point, Europeanness may become a surrogate identity, something that helps transform one or both of the inconvenient national identities. Where bi-cultural identities are taken up, Europeanness is developed as a new, overarching tier of belonging that encompasses the well-entrenched two national identities. More concretely, this chapter answered three questions. First, do internal migrants in the EU hold more pro-European attitudes and identities than stayers? Second, how do movers juggle and combine the three potential attachments that are available to them: that is, attachment to the country of origin (COR), to the country of residence (COR) and to the European Union? And third, which factors explain the different identity configurations of EU movers? Our findings strongly confirm the hypothesis that EU movers are more highly Europeanized than stayers. Movers know more about the EU, have a more positive image of it, often have European identities and feel more attached to the EU than stayers do. Indeed, almost half of our sample of free movers fall into the category of those holding tripartite identities, that is, strong COO identities, strong COR identities, and strong European identities. They are the group of ‘perfect Europeans’ who do not experience conflict between COO and COR identities and who find it easy to develop an ‘interculturated’ European identity layer on the basis of their bi-culturalism. For these movers, identities are additive and consonant. Migrants who fall into this group of highly Europeanized citizens tend to hold a university diploma, speak the COR language, engage in political discussions and, most importantly, have several friends from both the country of origin and the country of residence. The other half of our sample divides up into three blocks that are roughly equal in numbers: (a) those who are attached to both the COO and the COR but not to the EU; (b) those who hold European identities

More mobile, more European?

153

but only one national identity (COO or COR); and (c) those who hold one identity only (be it COR, COO or EU) or no identity at all. All of these three groups are likely to have experienced some kind of identity conflict (‘cognitive dissonance’) and to have chosen between unharmonious identities. Although sharing the experience of a conflict, these groups differ in many ways: age at migration, length of stay, gender and migration motives; most significantly, discrimination and integration experiences as well as language knowledge play important roles in predicting movers’ ways of dealing with identity conflict. Finally, in terms of the individual factors that account for heightened ‘Europeanness’ among EU movers, we note the following: COR versus COO effects. The primary socialization one has received usually has a larger effect on movers’ identities and national or European attachment than the psychological acculturation undergone in the country of residence. However, it is precisely this COR acculturation that makes free movers special (that is, different from stayers), and that accounts for the development of a European identity in about 60 per cent of them. British effect. Migrants from Britain are a special kind of European movers. On the one hand, they are unlikely to develop a tripartite identity with a strong European element. Feeling British and feeling European seem to exclude one another. On the other hand, they are most likely to assimilate to the country of residence – either with or without nurturing a feeling of Europeanness. Education. The fact that those intra-EU migrants who identify with the EU are also the most educated ones leaves one wondering about the nature of ‘Europeanness’. Is European identity a civic and elitist rather than affective and accessible concept that addresses the ‘minds’ rather than the ‘hearts’ of Europeans? The role of ‘contact’ in the form of friends or partners. The best indicator for the development of a COR identity in addition to the COO one (and with it, increasing one’s chances for developing an interculturated European identity) is contact with COR citizens. As we know from Allport’s contact hypothesis (Allport 1954), not just any contact (say with shopkeepers or bosses) will do. Rather, only personal and non-hierarchical relationships that take place at regular intervals are conducive to decreasing identity conflict and enhancing affect for the COR. Having friends or a partner from the COR country is thus the deluxe road to bi-cultural identities. Our results have several implications for EU policies which aim at an amplification of the identification with Europe of EU citizens. We showed that those Western Europeans who live in a European Union country other than their native one identify more strongly with the EU, know

154

Pioneers of European integration

more about it and feel more attached to it. This implies that in order to strengthen identification with Europe and the EU, policies promoting free movement (Erasmus/Socrates, transferability of welfare benefits, recognition of diplomas and so on) should indeed be reinforced and promoted further. Identification with the EU can be considered at least partly a spillover effect of such policies. Furthermore, our analysis showed that there are different ways of developing a feeling of belonging to the European Union. Therefore, the Commission’s message regarding the possible co-existence of regional, national and supranational forms of belonging should be strengthened. Research, on the other hand, should focus more on the different ways and predictors of co-existing identities than on the question of whether identities are exclusive or not. Concerning possible predictors of the development of a European identity we also found that less-educated internal European migrants showed lower levels of attachment to the EU than highly skilled movers. The European Commission should therefore consider promoting a European identity that is based on affect (a feeling of solidarity, shared cultural heritage and so on) rather than on civic elements only (institutions, flag, European passport and so on), that presuppose a certain level of ‘knowledge’. A more affective, culture-based European identity might be more easily accessible to all tiers of European society.

NOTES 1. We are arguing here that behaviour influences attitudes and identities. When we are unsure of attitudes, we infer them from behaviours (Bem 1972), not least as a function of legitimation of past activities (Jost and Major 2001). Note however that other research has brought evidence that causality runs from attitude to behaviour rather than vice versa (Ajzen 1991). Along these lines, one could maintain that those EU citizens who are most positively disposed towards the European Union are also those who move to live and work in European countries other than their native ones. 2. According to the context in question, ‘Europeans’ can of course be variously defined as European Union citizens, Western Europeans, or those living within geographical Europe (Wallace 1990). The kind of attachment measured in the EIMSS is attachment to the European Union. 3. EB 61.0 from 2004 or EB 58.1 from 2002 when items were not included in EB 61.0. 4. Question text: ‘In general, does the European Union conjure up for you a very positive (1), fairly positive (2), neutral (3), fairly negative (4) or very negative (5) image?’ 5. Question text: ‘Using this scale, how much do you feel you know about the European Union, its policies, its institutions?’ (Scale 0–10). 6. Question text: ‘In the near future, do you see yourself as . . .?’ 1. ([COO] Nationality) only; 2. ([COO] Nationality) and European; 3. European and ([COO] Nationality); 4. European only. 7. Question texts: ‘How attached do you feel to the European Union?’ Very attached (1), fairly attached (2), not very attached (3), not at all attached (4). ‘How attached do you feel to [COO]?’ Very attached (1), fairly attached (2), not very attached (3), not at all

More mobile, more European?

155

attached (4). ‘How attached do you feel to [COR]?’ Very attached (1), fairly attached (2), not very attached (3), not at all attached (4). 8. Although important, we decided not to include the respondent’s social class in our analyses for statistical reasons as this variable showed a high multicollinearity with educational level.

7.

EU movers and politics: towards a fully-fledged European citizenship? Anne Muxel

INTRODUCTION As an economic, social and cultural entity, the European Union has made much progress in recent decades, becoming a visible and objective reality in both the representations and practices of the inhabitants of its member countries. Nevertheless, Europe as a political space is not so well advanced. The political construction of the EU has not only encountered formal and institutional obstacles – such as the referenda rejections of European constitutional reform in France, the Netherlands and Ireland – but also problems of recognition by its own citizens. In the space of 20 years up to 2004, while the powers and prerogatives of the European Union have not ceased to expand, the abstention rate in European elections has gone up by 17.3 points on average in all countries and, in many of them, comprise half the electorate. This paradox clearly reveals how difficult it is to implement the idea of European citizenship and mobilize the genuine consciousness of a European political space. Research devoted to European citizenship – which reports on national public opinion concerning Europe – shows a fairly sizeable imbalance between, on the one hand, the relative acceptance of the Union and the largely positive image it enjoys; and on the other, the weakness of effective practices in this new field of civic expression and intervention (Cautrés and Reynié 2001; Duchesne and Frognier 2002). The European Union has suffered a ‘democratic deficit’ from the beginning of its history (Lindberg and Scheingold 1970; Siedentop 2001; Déloye 2005). Time has not reduced this situation. On the contrary: Europeanized national elites continue to drive European integration with little input from mass publics (Haller 2008). For some years now, scholars have been developing two major types of hypothesis to interpret the apparent disaffection of European citizens towards the EU. The first suggests a kind of cognitive deficiency. European institutions are misunderstood, and this hampers any possibility of giving concrete expression to the immediate and familiar exercise of citizenship 156

EU movers and politics

157

through political action. Thus, political Europe remains an abstract, remote entity that enables citizens neither to engage collectively nor raise their level of consciousness. The second type of hypothesis emphasizes the persistence of national identities and idiosyncrasies. The citizen’s realm of political involvement is first and foremost national and the identified interests or aims of contention – even if they have a supranational dimension of preoccupation – pass first through the filter of the nation and the political power struggle that prevails there. This strong ‘nationalization’ that generally weighs heavily on all European ballots, and shapes all political discourse about Europe, clearly results from the difficulty of constructing and establishing stakes that are specific to the European political context alone (Koopmans and Erbe 2004; Perrineau 2005; Schmidt 2006). European elections are above all used as second-order elections and midterm national contests, which first and foremost reflect domestic political concerns (Reif and Schmitt 1980; Van der Eijk and Franklin 1996; Marsh 1998; Schmitt 2005). More recently, some political scientists have shown a more consistently distinctive EU voting, more related to European than national politics (Carruba 2001; Tillman 2004). Nevertheless, the preoccupation with the ‘democratic deficit’ remains. Some commentators have argued that the EU is as democratic as it could or should be and that there is no real problem of democratic legitimacy (Majone 1998; Moravcsik 2002). Multi-level governance, pluralism in the policy process, growing non-state interest representation and the sophisticated system of comitology within EU institutions, can all be underlined to suggest how there are other ways than electoral participation to gain power and influence at the European level, as at the national ones, and also to gain more acceptance among the European citizens themselves (Schmidt 2006). But there remains a legitimacy problem. Perhaps the EU requires, as Follesdal and Hix (2006) have suggested, a more open contest for political leadership and more parliamentary control to reliably ensure that European policies are responsive to European citizens’ preferences and requests. There has been a great deal of literature on ‘European citizenship’ in terms of political and legal theory (see for example, Meehan 1993; Wiener 1998; Magnette 1999; Bellamy et al. 2006), but there is still a relative lack of studies on European citizens’ political attitudes and behaviour. Comparative studies undertaken in Europe – particularly those produced using Eurobarometer – have made it possible to isolate significant factors in the inventory of attitudes and behaviours of Europeans, which follow their development over several years and identify certain specific national characteristics (Herrmann et al. 2004). But these studies cannot capture European citizenship in action because the contours are generally always

158

Pioneers of European integration

perceived and evaluated from within a national context (see also Bréchon and Cautrès 1997; Favell 2005). Our study of EU movers allows us to assess the possible emergence of a European citizenship in a very different fashion from conventional studies. From the outset, the study proposes a transnational context – a new European social space – and in doing so throws a novel light on an experimental population that has chosen migration within Europe. Their mobility, as Rother and Nebe explore in depth in Chapter 6, enables a directly experienced and possibly interiorized assimilation of the European project. So are these European citizens like the other ‘nationalized’ ones, or does their relationship with politics have specific characteristics? And are they less encumbered by the overcautiousness that characterizes their national counterparts in recognizing and expressing European citizenship? The aim of this chapter is, on the one hand, to draw a political portrait of these mobile European citizens, whom we expect to be close to a transnational view of politics; on the other, to grasp the possible premises of their ‘Europeanization’ – here understood in the sociological sense of the term (see also Koopmans and Erbe 2004; Fligstein 2008) – on the civic and political level. In order to understand EU movers’ relationship to politics we will test two hypotheses. The first is related to classical sociological factors which explain political participation, such as education levels and social status, as well as political ones, such as the ideological position of respondents. Are these factors once again decisive? That is, do we find the usual correlations between social integration and political integration, or do we find a specific effect of the experience of mobility? The second concerns political socialization and the specific impact of national political cultures. Do EU movers share particular political attitudes and behaviours, or do we recognize traces of their distinct national origins and identities?

A POLITICAL PORTRAIT OF EU MOVERS To draw the political portrait of our respondents we have used a set of classical indicators on political attitudes and behaviour. Our basic results for EU movers show a high level of interest in politics but a poor level of political participation. A large majority (56 per cent) – in all countries and of all nationalities – admits being interested in the politics of their country of origin; 24 per cent are even very interested. Almost one-third of them frequently discuss politics within their circle (29 per cent). Men are more interested than women, and older more than younger people. Two out of ten have already contacted a politician or signed a petition. Yet only 15

EU movers and politics

159

per cent have taken part in a demonstration. Membership is weak: 8 per cent belong to a trade union, 3 per cent are members of a political party. Taken aggregately, EU movers are thus politically concerned but not so active. This is the first picture we get. The consistency of their political participation is not so different from the rest of the population in Europe and in relation to some aspects, even appears weaker. The comparison, of course, has to be taken carefully given the nature of the EIMSS sample. EU movers are more interested in politics, demonstrate a little bit more, belong more often to a trade union, but have less often signed a petition, and vote less at general elections. These differences are not significant. A comparison of their electoral involvement, however, shows significant discrepancies depending on the type of election. Only 40 per cent took part in the last national general election in their country of origin while, according to official statistics, an average of 70 per cent of stayers took part.1 However, their participation in European elections is equivalent to, or even slightly higher than, that of stayers (51 per cent versus 49 per cent), a sign that may reveal a certain Europeanization of their political involvement. We shall return to this in the last part of this chapter. Studies in political sociology often highlight the importance of the relationship between individuals’ conditions of social integration and their attitudes and political behaviour. In particular, the level of education is one of the most determining factors. This holds for EU movers. The more qualified our respondents are, the higher their level of interest. Of those with university degrees, 68 per cent say they are very interested in politics, whereas only 30 per cent of those with no qualifications say the same thing. A similar difference can be found when we look at the impact of professional status: 69 per cent of those in the upper-class category show a high level of interest, whereas the same applies to only 56 per cent in the routine white-collar group, 54 per cent of the petty bourgeoisie, 44 per cent of the qualified blue-collar group and 35 per cent of the non-qualified working-class group. To make an overall measurement of EU movers’ relationship with politics, we constructed a synthetic ‘politicization index’ from seven variables concerning both the level of political interest and the type of personal involvement (Table 7.1).2 Though the level of political interest is relatively high, movers’ effective involvement in political activities is relatively modest. In fact, only 16 per cent of the EIMSS sample possessed between three and seven of the attributes, 20 per cent had only two, 33 per cent one, and 32 per cent none. Thus, nearly two-thirds of respondents are scarcely involved in political life at all and their participation is virtually non-existent or extremely low. In sum, EU movers are more interested in politics, and somewhat more

160

Pioneers of European integration

Table 7.1

Participation variables for EU movers and stayers ( %)

Participation variables 1 2 3 4 5 6 7

Interest in politics*: very much Contacted politician Signed petition Took part in lawful public demonstration Trade union: participated Political party: participated Voted in last general elections

Notes:

* For EU movers: ‘Interest in politics in COO’.

Source:

ESS 2004 and EIMSS, N (minimum) 5 13 950

EU movers

Stayers (general population)

24.2 19.0 19.2 15.3 8.1 3.0 40.1

11.7 14.7 30.0 11.8 3.2 2.1 80.8

familiar with traditional modes of participation (like taking part in demonstrations or affiliating with unions and parties); but they are much less likely to vote or sign a petition. Consequently, it is not that EU movers are less ‘political’: rather, they are not consistently participative. The most politicized are also the most qualified, and those who belong to the categories at the top of the professional hierarchy. Thus, of movers with university degrees, 23 per cent have between three and seven politicization attributes, whereas this is true of only 6 per cent of those without qualifications. Likewise, 22 per cent of our respondents in the upper class have this level of politicization, but only 6 per cent in the working class. Conversely, a total absence of politicization appears more widespread among unqualified people and among movers belonging to the working class: 49 per cent and 47 per cent, respectively, have no politicization attributes. In the upper class, and among those with university degrees, about half those percentages have no politicization attributes: only 24 per cent and 23 per cent respectively. A logistic regression enables us to control for social and political factors that account for EU movers’ level of politicization (Table 7.2). Four factors are particularly influential. The first one, the most significant, is the ideological position. Leftist EU movers are more disposed to be highly politicized than the others. The second one is the country of origin. It confirms the impact of certain national origins and identities which continue to condition the EU movers’ political attitudes and behaviour. We will come back to this aspect in the next section. The third one concerns the social and educational profile of our respondents. A high level of politicization

EU movers and politics

Table 7.2

161

Logistic regression on ‘high level of politicization’ among EU movers

Left-Right position Centre Rightist Don’t answer Don’t know Ref.: Leftist Country of origin Germany Britain Italy Spain Ref.: France Country of residence Germany Britain Italy Spain Ref.: France Territorial identity No territorial identity Integrated non-European Assimilated European Assimilated non-European Separated European Separated non-European European Ref.: Integrated European Education level No qualification Secondary education completed University Ref.: University entrance qualification Class based on last occupation Bourgeoisie Petty bourgeoisie High-skilled manual Low/non-skilled manual

B

Wald

Signif.

Odds ratio

1/OR

21.011 20.900 20.914 22.180

15.478 54.052 57.333 22.549 67.759

0.000 0.000 0.000 0.000 0.000

0.364 0.407 0.401 0.113

2.749 2.458 2.494 8.843

0.092 20.419 0.147 0.568

44.534 0.400 7.770 0.976 16.277

0.000 0.527 0.005 0.323 0.000

1.097 0.658 1.158 1.764

0.912 1.521 0.863 0.567

20.379 20.264 0.271 20.433

32.590 6.376 3.497 3.983 6.836

0.000 0.012 0.061 0.046 0.009

0.685 0.768 1.312 0.648

1.460 1.302 0.762 1.542

20.998 20.476 20.285 20.561 20.095 20.461 0.302

29.153 8.279 12.017 3.200 5.557 0.341 3.914 1.557

0.000 0.004 0.001 0.074 0.018 0.560 0.048 0.212

0.369 0.621 0.752 0.571 0.909 0.631 1.352

2.712 1.609 1.329 1.753 1.100 1.585 0.739

20.351 20.471

24.489 1.896 8.740

0.000 0.169 0.003

0.704 0.624

1.420 1.602

0.270

5.918

0.015

1.311

0.763

0.316 0.098 0.105 20.507

2.197 7.694 0.409 0.281 6.254

0.000 0.006 0.522 0.596 0.012

1.371 1.103 1.111 0.602

0.729 0.907 0.900 1.660

162

Pioneers of European integration

Table 7.2

(continued)

Ref.: Routine non-manual Type of settlement Urban area Rural area Ref.: Small/Medium-sized town Constant

B

Wald

Signif.

Odds ratio

1/OR

0.262 20.122

12.665 6.169 0.963

0.002 0.013 0.326

1.299 0.885

0.770 1.130

21.010

24.610

0.000

0.364

Notes: Coefficients for age, gender, religious affiliation, marital status, length of migration experience, migration experience, EU-image are not significant. Source:

EIMSS, N 5 4 612

is more widespread among educated and qualified EU movers. Lastly, the importance of territorial identity and location has to be underlined. EU movers who feel European are more likely to be politicized. Urban areas count also more politicized people than rural ones. These results, however, are not so different from those we would usually obtain in standard studies on political participation in Europe. What does stand out from this first sketch of the EU movers is that their favourable disposition towards politics – characterized mainly by a high level of political interest – remains detached from a real politicization involving active and sustained participation. One specific feature of EU movers’ relationship to politics that deserves further analysis is where they sit on a left–right scale (on this scale, see Jaffré and Muxel 1997; Muxel 1997). In fact, we find a certain leaning of EU movers to the left: 40 per cent classify themselves as left or centreleft; 22 per cent classify themselves as right or centre-right; fewer classify themselves as centre (17 per cent), but more (21 per cent) do not classify themselves at all – either because they refuse to respond (8.4 per cent), or because they do not know how to respond (12.6 per cent).3 The comparison with stayers is striking: in the ESS sample only 33 per cent locate themselves left or centre-left, and in the European Values Study (1999) only 26 per cent. Among those EU movers who are not highly qualified, the absence of positioning between left and right is more marked: 45 per cent of those without qualifications do not position themselves, compared to 29 per cent of those with a secondary-school qualification and only 13 per cent of those with a university degree. On the other hand, the higher the qualification, the more marked the leaning to the left: among our respondents with university degrees, slightly fewer than half classify themselves

EU movers and politics

163

as leftist (49 per cent), compared to 28 per cent of those with secondaryschool qualifications and only 22 per cent of those with primary-lower educational credentials. The prevalence of leftist affiliations among the higher social classes appears to be very clear: 47 per cent of EU migrants belonging to the upper class category position themselves to the left, as well as 41 per cent of the routine white-collar group, many less among qualified blue collars (32 per cent), and only 26 per cent of the non-qualified working class. It is, therefore, within groups at the top of the social and professional hierarchy, as well as in a significant proportion of the middle classes, that leftist affiliation is most significant. In the EU general population, an increased transfer of support to the left has been observed for approximately the last 15 years. The left has lost ground with working-class levels of society but clearly gained ground with the higher and culturally privileged classes (Grunberg and Schweisguth 1997). In this, our respondents appear to be fairly representative of the far-reaching ideological recomposition of the European political landscape and of the changes in electoral and partisan power struggles that are taking place. This preference for the left is accompanied by a certain reticence towards economic liberalism: 44 per cent of EU movers express disagreement with the idea that the state should withdraw from intervention in the economy.4 They also reveal themselves to be distinctly open and permissive with regard to individual liberty and respect for private life, and the higher the level of qualification, the more this is true: 48 per cent agree that homosexuals should be able to live their lives as they please.5 In sum, EU movers are not only more mobile and more open towards Europe, but they also hold universalistic values and a fairly pronounced leftist sensibility. A certain anti-free market attitude in matters of the economy is relatively common – ironic, perhaps, given the EU’s strong libérale reputation on the continent. However the latter is not always true and, within the left, is subject to differentiated positions on economic questions. In order to explore these ideological nuances among intra-EU migrants, we further differentiated – within the 40 per cent who are on the left – between those who are hostile to economic liberalism (22 per cent), those who accept it (9 per cent), and lastly those who do not take a position on the subject (8 per cent). Leftist affiliation within our population of migrants is split, therefore, into two types of adherence. One is fairly traditionally hostile to economic liberalism (22 per cent). The other is more permeable and open to the values of a market society and unites those who declare themselves to be explicitly in favour of economic liberalism (9 per cent) and those who appear relatively indifferent, or at least not hostile to

164

Pioneers of European integration

Table 7.3

Ideological position according to sociodemographic and political characteristics (%)

Leftist against economic liberalism Leftist for economic liberalism Indifferent leftist Rightist Centre No position Entire sample

Men