**IMISCOE Research Series**

## Ilka Steiner Philippe Wanner *Editors*

# Migrants and Expats: The Swiss Migration and Mobility Nexus

**IMISCOE Research Series**

This series is the official book series of IMISCOE, the largest network of excellence on migration and diversity in the world. It comprises publications which present empirical and theoretical research on different aspects of international migration. The authors are all specialists, and the publications a rich source of information for researchers and others involved in international migration studies.

The series is published under the editorial supervision of the IMISCOE Editorial Committee which includes leading scholars from all over Europe. The series, which contains more than eighty titles already, is internationally peer reviewed which ensures that the book published in this series continue to present excellent academic standards and scholarly quality. Most of the books are available open access.

For information on how to submit a book proposal, please visit: http://www. imiscoe.org/publications/how-to-submit-a-book-proposal.

More information about this series at http://www.springer.com/series/13502

Ilka Steiner • Philippe Wanner Editors

## Migrants and Expats: The Swiss Migration and Mobility Nexus

*Editors* Ilka Steiner Institute of Demography and Socioeconomics IDESO University of Geneva Geneva, Switzerland

nccr – on the move Neuchâtel, Switzerland Philippe Wanner Institute of Demography and Socioeconomics IDESO University of Geneva Geneva, Switzerland

nccr – on the move Neuchâtel, Switzerland

ISSN 2364-4087 ISSN 2364-4095 (electronic) IMISCOE Research Series ISBN 978-3-030-05670-4 ISBN 978-3-030-05671-1 (eBook) https://doi.org/10.1007/978-3-030-05671-1

Library of Congress Control Number: 2019930146

© The Editor(s) (if applicable) and The Author(s) 2019. This book is an open access publication.

**Open Access** This book is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

The images or other third party material in this book are included in the book's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the book's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG. The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

## **Preface**

Recent decades have witnessed an increase in population flows towards highincome countries. At the same time, migration is becoming more complex because motivations, conditions and forms of migration have sharply diversified. Those changes are a challenge to the frames that shaped our understanding of migration.

The study of migration developed in American countries, which were willing to let migrants from abroad settle and populate the country. In earlier library classification, migration was catalogued under the heading demography. The mid-nineteenth century marks the establishment of the nation-states in Europe and beyond. The national frame shaped the understanding of migration, settlement was the main outcome of those geographical moves, and assimilation turned into the leading cultural narrative, which fit perfectly the nation-building agenda.

As long as the world was structured according to nation-state principles, these became so routinely assumed and "banal" that they vanished from sight altogether. Wimmer and Glick Schiller in 2002 consider the "assumption that the nation state society is the natural social and political form of the modern world" misleading. Their critique of methodological nationalism implies considering migration from a new perspective; the globalized world provides a new frame to capture individual and group experience. This shift in migration studies, they argue, is due to the epistemic move of the observer and not to the appearance of new objects of observation.

This book is based on the assumption that the globalization processes of the last 25–30 years have indeed induced migration researchers to extend the objects of observation; researchers interpret present migration patterns as the result of migrants' agency in the context of changing economic drivers, legal norms and societal factors. Migrants' position is no longer exclusively framed in the "container space" of the nation-state, delimitated by its borders; their position and agency are also contingent on a larger relational space, whereas the new residence country is still the main arena for integration strategies.

Given the globalization dynamic, it may appear paradoxical to study new migration patterns in only one country. However, Switzerland is a laboratory presenting all facets of the changing patterns of migration in developed regions, hence, the underlining hypothesis of the authors. The antagonist forces structuring the migration context appear in the small country under scrutiny in their more vivid expression. The country is one of the most integrated in the globalized economy, and it is strongly interlinked to its international immediate political context. The overarching globalization drive pushed the country to adopt the same dual migration regime prevailing in the European Union. At the same time, the peculiar, semi-direct democracy system makes the countermovement claims more vocal, pushing for welfare protection based on internal inclusion and external exclusion, two dynamics that influence both admission and integration policies.

Taking for granted the knowledge about this context, the authors focus all their analyses on recent migrant flows, whose number sharply increased in the last decade. They exploit the Migration-Mobility Survey, a new, ad hoc, migrant-specific data source they developed in order to address issues that are not covered by censuses, general population surveys and register data. This tool is peculiar because it articulates flow dynamics and their permanent and/or temporary migratory patterns with integration dynamics and their patterns of inclusion and exclusion for the foreign born. Interestingly, the integration processes of recent migrants analysed reach out not only to the labour market but also to other social spheres, such as political participation and transnational practices. Moreover, the tool exploited in this book is conceived as a first step of a longitudinal survey on first-time arrivals in line with other surveys initiated in other European countries.

The book is structured in six parts, starting with an introductory part on the Migration-Mobility Nexus and the Survey. Four empirical parts follow, and a concluding part rounds out the work. The empirical analyses cover the Migration-Mobility Nexus in the frame of inflows, cover the integration processes of recent immigrants in the labour market and in social and political spheres/domains and finally address prospective and symbolic issues involving mobility.

Studies conducted on the characteristics of recent migrations deliver insightful knowledge on differentiation dynamics within those new flows. Globalization of mobility has induced an increase in migrants counting on several previous foreign experiences. This is particularly true for the highly skilled, because international tertiary education is now a standard. However, history and geography also play a role in differentiating groups; geographically distant, Anglo-Saxon origin migrants are more often concerned, whereas migrants from neighbouring countries tend to be first-time migrants. Movements of highly skilled migrants stand out as the product of the interplay of economic drivers and political migration regimes. However, analyses show specific mediations accounting for the increased valorization of their "migratory capital"; economic actors play a crucial role in accompanying migratory trajectories, thus reinforcing the advantage enjoyed by those migrants.

Given the almost full-employment situation characterizing the Swiss labour market and the overall high level of qualification among newcomers, structural integration is not an issue for a majority of newly arrived migrants. The situation of professionals whose migration is not directly driven by labour demand is however more problematic; this point is particularly true for women. There is evidence that immigrants "downgrade" upon arrival in the host country; they have difficulty finding a job that matches their level of education and work experience. There is a significant association between the country of credential acquisition and the risk of educational mismatch. Risks of mismatch vary however according to the country of credential acquisition; they are highest for Latin American workers, followed by migrants from EU28/EFTA countries other than the neighbouring countries. Although overall structural integration of newcomers improves over time, labour disadvantage persists over time for some immigrant groups, particularly for women, regardless of their human capital characteristics and their integration in Swiss society, making segmentation and discrimination persistent features in the labour market.

Integration goes beyond the labour market domain and involves social and political spheres. Following the hypothesis that features of the reception context affect the integration process, the authors single out three dimensions of this context: perception of unfair treatment, the local political environment and political participation opportunities.

The main driver for reported discrimination is immigrants' origin. The frequency of such reports is lower for European origin groups, increases for North and South Americans and is highest for West Africans, mirroring the importance of ethnic ranking pointed at by other contributions to this book. Although the first listed groups are mostly exposed to unfair treatment in public situations, the last group indicates that work environment is the most affected. This finding echoes the greater differences observed in access to the labour market according to origin. Cantonal reception contexts do affect feeling of attachment to the new country of residence but only indirectly. Although individual length of stay remains the main driver for attachment, inclusive cantonal reception contexts and liberal cantonal integration policies in particular amplify the positive effect of years of residence on immigrants' identification with their residence country. Conversely, experiences of ethnic discrimination lower levels of identification. International mobility can constitute a reason for low political participation in the form of mainstream political organization. However, considering new forms of political engagement reveals that international mobility experiences actually become a resource for political activation of migrants. Residential stability and social connectedness stimulate political engagement, just as do feelings of discrimination.

The Migration-Mobility Nexus discussed initially in the context of inflows becomes once again crucial when considering (virtual, potential or effective) outflows. Transnational ties to the country of origin are found strong among migrants with large human and mobility capital and among migrants with minimal economic and social resources. It can be inferred that social class and migration regimes influence the specific content of transnational ties, a dimension difficult to grasp in surveys. Networks and attachment to the country of origin tend to decline with the length of stay in the residence country. Immigrants' projects for their future reflect the respondents' attitudes towards their migration experience in the host country; long-term intentions for the stay and naturalization concern well-integrated family or lifestyle migrants. Conversely, remigration intentions are found among labour migrants and international students.

The migrant-specific survey that served as a basis for the contributions embraces both admission and integration dynamics; this tool highlights the impressive continuity between conditions of entry and conditions of integration. The effect of entry conditions on forms and opportunities of integration emerges more clearly than ever. The course is set from the very beginning; the first few years of stay already contain the seed of integration processes usually analysed over much longer periods, in fact over generations. Recent migration flows are characterized by a sharp differentiation in human capital endowments, resulting from policies adopted in high-income countries in the era of globalization. The book shows that ethnic ranking and gender hierarchies interact pervasively and systematically with human capital to shape distinct mobility and integration opportunities and trajectories. The ethnic differentiations highlighted in this study relate to the different migration regimes and do not pertain to specific cultural features of the single-origin groups.

Scholars studying migration in industrialized and high-income destination countries will find an interest in this book. Scholars from a variety of disciplines such as demography, sociology, political science and economics are the target audience of the book because it addresses recent developments from a pluri-disciplinary perspective. Moreover, the book is of interest to students in migration studies and in survey methodology.

Beyond academics, the book may resonate among policymakers, in Switzerland of course but more generally in high-income countries, because most analyses examine the economic, social or political integration of immigrants. The accurate picture provided of the current situation is indeed accompanied by suggestions for approaches that can enhance immigrants' integration.

 Rosita Fibbi Swiss Forum for Migration and Population Studies SFM University of Neuchâtel Neuchâtel, Switzerland

nccr – on the move Neuchâtel, Switzerland Rosita.Fibbi@unine.ch

## **Acknowledgements**

The copublishers would like to express their deepest appreciation to all those who provided them the opportunity to complete the Migration-Mobility Survey and to realize the present book, in particular, the Swiss National Science Foundation, which finances the National Center of Competence in Research nccr – on the move since June 2014. The copublishers would also like to acknowledge with much appreciation the crucial role of the nccr – on the move Network Office, in particular, Nicole Wichmann, Gina Fiore and Ursula Gugger Suter, for all of their support during the elaboration and realization of the Survey. The Migration-Mobility Survey was organized in cooperation with the LINK Survey Institute, Lausanne, under the supervision of Dr. Massimo Sardi, who deserves our sincere thanks for his technical and scientific support during the process of data collection. The copublishers were supported for the data preparation and survey report writing by Aljoscha Landös.

All chapters in the present book were reviewed by anonymous experts whom the copublishers and all authors want to thank for their feedback and suggestions, which contributed greatly to improving its quality. In addition, American Journal Experts (AJE) proofread the present book and deserves our gratitude. The copublishers were supported for the final revision of the manuscript by Roxane Gerber.

Very special thanks finally go to the IMISCOE network and to the Springer editorial team and production for the final realization of this publication.

## **Contents**

#### **Part I Introduction**



## **Contributors**

**Daniel Auer** is a postdoctoral researcher at the WZB research unit Migration, Integration, Transnationalization. He received his PhD in Public Administration from the University of Lausanne and the National Center of Competence in Research nccr – on the move. Previously, he studied at the Universities of Vienna, Urbana-Champaign and Zurich and has been a visiting researcher at Malmoe and Oxford. His work focuses on discriminatory and policy-driven mechanisms that affect immigrants' labour market integration and on policy effects in the context of international migration. He recently published articles investigating the random placement of refugees in Switzerland across different language regions and its consequences for labour market integration (*Journal of Ethnic and Migration Studies*), on the signalling value of active labour market measures (*European Sociological Review*) and on the effect of discrimination vis-à-vis immigrants' work values on unemployment duration (*International Migration*). Personal website: daniel-auer.com.

**Salomon Bennour** is a PhD student and teaching assistant in Migration Policy Analysis at the Swiss Forum for Migration and Population Studies at the University of Neuchâtel. In addition, he collaborates with the Swiss NSF project nccr – on the move. Before starting his PhD in early 2017, he completed a bachelor's degree in Political Science at the University of Lausanne and spent a student exchange year at Uppsala University in Sweden. He obtained his master's degree in Socioeconomics at the University of Geneva in 2017. After a second experience abroad at the University of Tübingen in Germany, he wrote a thesis addressing the internal mobility of foreigners in Switzerland from a demographic perspective. His PhD thesis addresses the effects of subnational integration policies on foreigners, such as policy incentives to migrate within and outside Switzerland and naturalization behaviours. Thus, his current research interests are linked to different fields such as demography, political sociology and integration policy.

**Eric Crettaz** is professor at the University of Applied Sciences of Western Switzerland in Geneva. He received his PhD at the Swiss Graduate School of Public Administration (University of Lausanne, Switzerland) for a study of policies that alleviate working poverty in Europe and North America. Between 2006 and 2011, he was a member of the Network of Excellence Reconciling Work and Welfare in Europe, European Commission, FP6, and has since participated in several other research programmes. He is currently involved in a large-scale project dealing with the Migration-Mobility Nexus, with a focus on retirees. He has specialized in the fields of economic inequality and poverty, in particular, atypical employment and working poverty, and the quality of life of pensioners together with related comparative social policy analyses.

**Janine Dahinden** is professor of Transnational Studies and head of the Centre for the Understanding of Social Processes (http://www.unine.ch/maps) at the University of Neuchâtel, Switzerland. Her research approach represents a social science that places the understanding of the processes of migration and mobility, transnationalization and social and symbolic boundary making in the realm of ethnicity, religion and gender. She has conducted empirical research in Kosovo, Albania and Switzerland using qualitative methodologies, social network analysis and quantitative comparative surveys. From 2005 to 2006, she was a member of the board of directors of the Swiss Forum for Migration and Population Studies. She has been teaching at different Universities in Europe and has published in German, English and French in a wide variety of journals such as *Ethnic and Racial Studies*, *Journal of Ethnic and Migration Studies*, *Global Networks*, *Journal of International Migration and Integration*, *Ethnicities*, *Social Politics*, *Diversities* and *Journal of Intercultural Studies*.

**Gianni D'Amato** is professor at the University of Neuchâtel and director of the nccr – on the move, a centre of excellence in research, and of the Swiss Forum for Migration and Population Studies. His research interests are focused on citizenship, mobilities, populism and history of migration. After his MA in Sociology at the University of Zurich, D'Amato focused his PhD thesis on citizenship and migrants' integration in a comparative perspective. He is author of *Vom Ausländer zum Bürger: Der Streit um die politische Integration von Einwanderern in Deutschland*, *Frankreich und der Schweiz* (Lit Verlag third edition, 2005). His recent publications include *Critical Mobilities* (ed. together with Ola Söderström, Shalini Randeria, Didier Ruedin and Francesco Panese) (London: Routledge, 2013), *The Politicisation of Migration* (edited together with Wouter van der Brug, Didier Ruedin and Jost Berkhout) (London: Routledge) and *Multireligious Society: Dealing with Religious Diversity in Theory and Practice* (edited together with Francisco Colom) (London: Routledge, 2016).

**Rosita Fibbi** studied in Rome, Zurich and Geneva. She is a Doctor of Political Science. She taught Sociology of Migration at the Faculty of Social and Political Sciences at the University of Lausanne and works as a project manager at the Swiss

Forum for Migration and Population Studies at the University of Neuchâtel. Having received some 12 major research grants, she performed research on the sociology of migration and integration processes, acculturation processes and the life course, school and labour market trajectories, identity, discrimination, associations and migration and integration policies. Her works follow two strands: individual and social dynamics throughout the life course (with topics such as school, labour market integration and discrimination against second-generation migrants, intergenerational relationships and elderly migrants) and migration and integration policy (with topics such as national and local policies in Switzerland, political participation of immigrants and student migrations).

**Metka Hercog** works as a postdoctoral researcher at the Institute of Cultural anthropology and European Ethnology at the University of Basel. She is a part of the nccr – on the move, researching mobility of the highly skilled towards Switzerland. She received her PhD in Public Policy from Maastricht University, the Netherlands. Her PhD thesis investigated the effects of government policies on the choice of destination country by highly skilled migrants. She obtained an MSc degree in International Development Studies from Utrecht University and a university degree in Political Science from the University of Ljubljana. Prior to joining the University of Basel, she held research positions at the Swiss Institute of Technology in Lausanne (EPFL), the University of Lausanne, and at the Labour Migration Branch of the International Labour Organization (ILO).

**Aljoscha Landös** is a PhD student and teaching assistant at the Institute of Demography and Socioeconomics at the University of Geneva. In addition, she acts as a survey coordinator for the second wave of the Migration-Mobility Survey funded by the nccr – on the move. After an exchange semester at the University of Economics in Vienna and a field study in Senegal studying the impact of parental migration on the education of children left behind, she obtained a master's degree in Socioeconomics and Demography at the University of Geneva in 2016. Currently, she works on her PhD thesis analysing the impact of cantonal integration measures on the transition to adulthood of young refugees in Switzerland using exhaustive longitudinal data on educational, familial and professional pathways.

**Anita Manatschal** is assistant professor in Migration Policy Analysis at the Swiss Forum for Migration and Population Studies (SFM) at the University of Neuchâtel. This position is affiliated with the nccr – on the move. Before joining the SFM, she was a postdoctoral researcher at the University of California, Berkeley, researcher and lecturer at the University of Bern and visiting scholar at the European University Institute in Florence. She holds a master's degree in Political Science, General Linguistics and Islamic Studies from the University of Zurich and a PhD in Political Science from the University of Bern. Located at the intersection of political sociology and comparative politics, her research interests include comparative migration policy research, democratic institutions, immigrant integration, civic engagement and in-/exclusive societal attitudes. In her currently ongoing nccr – on the move subproject, she scrutinizes together with a group of international scholars regional immigrant integration policies in a global comparative manner, focusing on the USA, Canada, Germany, Italy, Spain, Great Britain, Belgium, Switzerland and Australia. Recent publications of Anita Manatschal appeared in *Policy Sciences*, *West European Politics*, the *Journal of Ethnic and Migration Studies* and various edited volumes and monographs.

**Marco Pecoraro** is a scientific researcher at the Swiss Forum for Migration and Population Studies (SFM) and the Institute of Economic Research (IRENE) at the University of Neuchâtel. He holds a PhD in Economics from the Université Catholique de Louvain in Belgium, and he has extensive experience in statistical analysis and econometrics. Dr. Pecoraro is familiar with the institutional setup of migration in Switzerland, and his work focuses on the determinants of migrant health from a labour market perspective and on the determinants of attitudes towards foreigners and immigrants. His research has also been devoted to other relevant aspects of labour market adjustments, including educational and skills mismatch. He has also acted as an international expert in the analysis of the Albanian Census in 2014. His recent publications appeared in *Applied Economics*, *International Migration Review* and the *International Journal of Manpower*.

**Didier Ruedin** (DPhil, Oxford) is a senior researcher and lecturer at the Swiss Forum for Migration and Population Studies at the University of Neuchâtel, Switzerland, and a visiting research fellow at the African Centre for Migration & Society at the University of the Witwatersrand, South Africa. His research focuses on reactions to immigration and diversity, entailing various aspects of politicization, attitudes and discrimination in the area of immigration and ethnic groups; on individual attitudes towards foreigners and immigrants in different contexts and ethnic discrimination in the labour market; and on examining how, when and why potential conflicts over migration become politicized. He also works on the political representation of women and ethnic groups in national legislatures, which appeared among others in the *European Sociological Review* and won the SNIS International Geneva Award 2010. He recently published a meta-analysis on ethnic discrimination in hiring decisions (*Journal of Ethnic and Migration Studies*) and on the role of unemployment risk and values in attitudes towards immigrants (*International Migration Review*). Personal website: druedin.com.

**Laure Sandoz** is a research fellow funded by the nccr – on the move at the Institute of Cultural Anthropology and European Ethnology at the University of Basel. Her research focuses on the institutions and agents that structure the mobility of highly skilled migrants in Switzerland. Laure Sandoz obtained her PhD at the University of Basel in 2018. She previously completed her master's degree in Social Science at the University of Neuchâtel and was awarded a prize of excellence for her MA thesis. Her MA project also won the 2016 Academic Prize of Sustainable Development at the same university. In addition to her studies, she has worked at the Centre for Migration Law, which brings together social science researchers from different Swiss universities. She is also a founding member of the Swiss Network of Young Migration Scholars, an association that aims to bridge the gap between academia and practice in the field of migration research.

**Fabian Santi** is a data scientist with wide experience in the treatment and analysis of data from various sources, ranging from refugee camp survey data and numerical simulations in neuroscience to sales and logistics data at various companies. After obtaining a master's degree at EPFL in Bioengineering and Bioinformatics, he worked for 3 years in the analytics department of a large consulting company. Currently, Fabian works as a scientific collaborator at the Swiss Federal Statistical Office (FSO) in the "Wages and Working Conditions" section. He also studies parttime for a second master's degree in Statistics at the University of Neuchâtel.

**Ilka Steiner** is a postdoc at the Institute of Demography and Socioeconomics at the University of Geneva and part of the nccr – on the move. In addition, she is a lecturer in Demography at the University of Neuchâtel in Switzerland. Her research interests lie in immigrant demography. She obtained a PhD in Demography from the University of Geneva with a focus on remigration intentions and behaviour of immigrants, under the provision of free movement of persons. In this framework, she launched a new register-based longitudinal database for migration research in Switzerland that is today used by several research teams of the nccr – on the move. In addition, she coordinated the implementation of two national surveys among immigrants living in Switzerland, thereof the first wave of the Migration-Mobility Survey. Currently, she is involved in a subproject on the impact of family formation on return migration between Switzerland and Sweden from a comparative perspective, wherefore she is currently based at the Stockholm University Demography Unit (SUDA) as a guest researcher.

**Elena Vidal-Coso** is a postdoctoral researcher in the nccr – on the move project (Swiss National Science Foundation) at the Institute of Demography and Socioeconomics (University of Geneva). In 2009, she obtained a PhD in Demography from the Centre for Demographic Studies (CED) at the Universitat Autònoma de Barcelona. She was a Juan de la Cierva fellow (2011–2014) in the DEMOSOC Research Group at the Department of Political and Social Sciences (Universitat Pompeu Fabra). She is particularly interested in understanding the effects of human capital and segmentation dynamics on the labour market performance of immigrants. Moreover, she aims to better understand couples' arrangements in terms of employment and income. Specifically, she considers the role of the relative resources of the male and female partners in shaping patterns of gender inequality. Finally, she analysed the effect of the Great Recession in increasing employment inequalities between natives and immigrants. Her work has been published in the *Journal of Ethnic and Migration Studies*, *Demographic Research*, *International Migration*, *Migration Letters*, *Revista Española de Investigaciones Sociológicas* and *Migraciones*.

**Philippe Wanner** is professor of Demography and head of the Institute of Demography and Socioeconomics at the University of Geneva. He holds a bachelor's degree in Economics and a PhD in Demography. He is also deputy director of the nccr – on the move. His research focuses on the statistical measurement of integration from a longitudinal perspective, the demographic impacts of migration, the processes of integration and the measurement of migratory flows and integration dynamics. He recently edited a book on the demography of foreigners (2012) and a book on demographic trends in Switzerland (2014). He is also author and co-author of different peer-reviewed papers on migration and integration, including the integration of economic migrants and refugees into the labour market, and on migrant health.

**Jonathan Zufferey** is a postdoctoral research fellow at the nccr – on the move. In addition, he is a lecturer at the Institute of Demography and Socioeconomics at the University of Geneva. After completing a PhD on Migrant Mortality Differentials in Switzerland, he is now working on the new forms of mobility and their socioeconomic effects. His research primarily focuses on multiple migration trajectories and on migrant spatial localization in the Swiss context.

## **Part I Introduction**

## **Chapter 1 Today's Migration-Mobility Nexus in Switzerland**

**Gianni D'Amato, Philippe Wanner, and Ilka Steiner**

#### **1.1 Introduction**

Current patterns in studying the movement of persons have been going through a process of reconsideration over the last few years. Migration studies have been increasingly superposed by more-recent studies of mobilities. When examining the linkage between mobility and migration studies, researchers must begin by questioning some of the assumptions underlying the standard definitions of international migration, located within the framework of the international state system. The linkage also raises the issue of the agency of those who are or have been literally *on the move*.

Indeed, a wide consensus exists among social researchers that migration patterns, regimes, and life-worlds have undergone a considerable transformation in contemporary Europe in recent decades. Compared with the post-war era, when migration regimes were characterized by temporary labour recruitment policies and, when immigrants settled, the paradigm of assimilation, the acceleration of globalization since the 1970s has given rise to many new types of migrants, policies, legal regulations and societal dynamics. In this recent past, logics of free movement, security and human rights discourses and anti-immigrant populist mobilization have

P. Wanner · I. Steiner Institute of Demography and Socioeconomics IDESO, University of Geneva, Geneva, Switzerland

nccr – on the move, Neuchâtel, Switzerland e-mail: Philippe.Wanner@unige.ch; Ilka.Steiner@unige.ch

© The Author(s) 2019 3 I. Steiner, P. Wanner (eds.), *Migrants and Expats: The Swiss Migration and Mobility Nexus*, IMISCOE Research Series, https://doi.org/10.1007/978-3-030-05671-1\_1

G. D'Amato (\*)

Swiss Forum for Migration and Population Studies SFM, University of Neuchâtel, Neuchâtel, Switzerland

nccr – on the move, Neuchâtel, Switzerland e-mail: Gianni.Damato@unine.ch

been at work simultaneously, questioning and re-arranging established forms of inclusion and exclusion according to class, nationality, origin, religion and gender. It appears that governments, legislators, academia, national and transnational public spheres, and migrants and their offspring lack the tools to fully grasp the situation, which is characterized by complexity and uncertainty.

The present book sheds light on how economic drivers, societal factors and legal norms shape today's migration and integration patterns in Switzerland. Thus, it investigates not only how permanent and/or temporary today's migratory patterns really are but also how Switzerland's selective regime of migration and mobility influences existent patterns of inclusion and exclusion. The notion of the Migration-Mobility Nexus, sketching a continuum between the two poles of migration and mobility, both at a theoretical and an empirical level, serves as the conceptual and overarching framework for the present book.

This introductory chapter sketches this framework and introduces the research questions, objectives and key issues addressed in this book. First and on a theoretical level, the two paradigmatic lenses of migration research and mobility studies are presented. Second, the transformation of European migration regimes since the 1970s and its effect on the patterns of migration and mobility is discussed. Third, Switzerland, being part of the European Migration Regime in transformation, can be used as a laboratory to understand the changes in and of an advanced postindustrial society. To this end, we provide a short empirical overview of the immigrant population and their living conditions in Switzerland. Fourth, the chapter provides a set of analytical questions that will be addressed throughout this volume – by means of the Migration-Mobility Survey data – and discussed in the concluding chapter.

#### **1.2 Two Theoretical Lenses: Migration and Mobility Studies1**

Classical migration research emerged at the end of the nineteenth century at the University of Chicago. This research has traditionally addressed human mobility, focussing on its national governance through the paradigm of assimilation (reframed in the last 30 years in public discourse as "integration"), opposing in remote times nativist and anti-immigrant groups that, based on eugenic arguments, wanted to prevent immigration of undeserving new migrant groups (FitzGerald and Scott 2014). The notion of "migration" highlights the capacity of a nation to define who belongs to the state and who does not. Classical migration research, therefore, operated from the perspective of the host societies and their capabilities to assimilate migrants. Since then, migration was and remains publicly debated around the concepts of integration or assimilation.

<sup>1</sup>This section presents a concise summary of a text written by Gianni D'Amato and published in a co-authored chapter (see Söderström et al. 2013).

As framed in Critical Mobilities (Söderström et al. 2013, p. 9), uncoupling migration from the nation-state framework entails conceiving it as a phenomenon embedded in a larger context, be it regional or global. Such a conceptualization was already foreshadowed in the earlier structuralist works of Piore (1979) and Castles and Kosack (1973), or in the adaptation of migration studies to international system theory by Portes and Walton (1981), who foregrounded power relations and structures of economic dependency between the "West and the Rest" (Hall and Du Gay 1996). At the same time, studies of transnationalism (Glick Schiller et al. 1992; Kleger 1997) and hybridity (Werbner and Modood 2005) emerged, opening up new perspectives on the mobility of people, objects and ideas. Such studies do not consider migration a one-way and once-and-for-all movement from a place of origin to a destination. In contrast, interactions between these two, or more, countries occur within various transnational networks in which migrants are embedded. This behaviour led to numerous empirical studies exploring cross-border kinship relationships, transnational flows of remittances, and double incorporation of migrants into two nation-states.

Connected to these trends is the emergence of a dual regime of migration and mobility. The latter is characterized in Europe, first, by the establishment of the free movement regime applicable to the nationals of the EU/EFTA member states and, second, migration being defined by the rules for controlling the entry, admission and stay of third-country nationals. This statement is particularly true for Switzerland, one of the leading Western economies, which is heavily dependent upon a migrant workforce to sustain its pace of economic growth.

Increasingly, scholars from geography and social disciplines, searching in parallel for new conceptual, theoretical and methodological approaches, share the assumption that our times are characterized by increasing mobility (Maurer 2000). Thus, "mobility has become a most suitable trope for our time, an era accelerating at what appears to be ever faster rates of speed" (Tiessen 2008, p. 112). Therefore, mobility studies highlight the dimensions of circularity and movement as constitutive elements of human societies, in particular in the scaled-up context of globalization (Cresswell 2006). Mobility studies connect the movement of people more systematically with the global circulation of ideas, goods and objects.

Several studies suggest the acceleration and diversification of the types of mobility and their increasing role in reorganizing society and the labour market (Latour 2005, for instance). With the rapid intensification of travel – whether physical (people, goods, and materials), imaginative (knowledge, ideas, and images), or virtual (money, information, practices, and e-mails) – the fact of movement, its meanings and implications must be studied in their own right and as affecting the very constitution of societies (Cresswell and Merriman 2008; Urry 2007; Watts and Urry 2008). Mobilities are also linked to the question of inequality; the capacity to move is not the same because of differences in social, financial and human resources. Indeed, mobility studies have been criticized for overlooking the political and economic structures that underlie the different possibilities to move. It has been argued that the epistemology of mobility reflects an ideology of free movement. Mobility studies should therefore rethink the extent to which migration as a phenomenon

must be defined in terms of, and derived from, the needs of the state to classify spatial mobility in a particular way. Therefore, we first must understand the power of the state to classify different types of mobile subjects and their practices (Favell 2007). In general, an international migrant is defined as a person who lives in a country other than his country of origin for a certain minimum length of time – according to the United Nations, 12 months for a long-term migrant and 3–12 months for a short-term migrant. Others who cross borders – such as tourists, business people, and international students – are not only excluded from the definition of a migrant but also invisible in the classical migration research due to the lack of registration, the short duration of their stay, the unpredictability of their settlement and the lack of interest from researchers. Moreover, the identification in migration surveys or population registers of an increasing group of mobile persons who live in two or more places – for instance retired persons, businessmen and businesswomen – is challenging and can differ from one country to another (Poulain et al. 2006). Migration studies are also a policy-driven field in which academic debate often follows political contention, for instance in terms of categorization of migrants into "wanted" and "unwanted" migrants from the perspective of the "receiving" society.

The dismantling of fixed borders, boundaries and conceptualizations underlying standard definitions of migration is a necessary move towards a critique of the fixity of categories, which the mobilities paradigm calls for. In the nexus between migration and mobility studies lies the potential of a combined approach. Instead of focussing on legal boundaries and borders, migration and mobilities must be studied as objects that are created and negotiated within the organization of labour division, state practices, transnational family making and material exchange (Rouse 1991) and that allow for different practices of spatial mobility. This structure is the paradigm change of the Migration-Mobility Nexus, in which the agency of migrants and mobile people becomes essential for its understanding.

Therefore, this book will give large space to current patterns of mobility to Switzerland, its intermediaries, intervening networks and social capital, the socioeconomic inclusion of these newly arrived persons, its habits and societal life as much as their transnational habits and future expectations, going beyond the traditional frame of migration studies to grasp new realities as experienced by mobile persons.

#### **1.3 A European Migration Regime in Transformation**

Historical research usually addresses migration as an often-neglected element of Europe's modernization that facilitated the necessary economic adjustments. A characteristic feature of early migration regimes in the late nineteenth century was that they were anchored within the emerging framework of modern nation-building. In the interwar period (1918–1939), Western states started to make a clear distinction between their own citizens and those belonging to another nation-state. This differentiation was necessary to lay the foundation for the birth of the welfare state. It is against this backdrop that the imposition of legal restrictions on immigrants entering the country or its labour market should be understood. The instruments of legal restriction took different forms, such as visa regulations to control entrance, the tying of visa issuance to labour market permits, and the decision to restrict the access of refugees and to monitor aliens within a country's borders.

Following the discontinuity caused by the two world wars, post-war Europe's immigration policies rebooted and were characterized by temporary labour and postcolonial recruitment policies. Its immigrant policies were shaped by the paradigm of (segmented) assimilation into the expanding national welfare states. However, since the oil crisis and the acceleration of globalization in the 1970s, the Fordist consensus had been transformed (see for example Castles et al. 2013). One could argue the spatial and political rearrangement of movement in Europe that has since occurred has been driven by the fluid – and often contradictory – processes of global marketization and regional securitization and by the rising influence of human and migrants' rights regimes.

Indeed, in the wake of the oil-crisis in 1973, many Western countries were struck by de-industrialization and unemployment. In reaction, many countries clamped down on the recruitment of migrant workers either by restricting the delivery of working permits (as was observed in Switzerland) or by tightening the criteria for citizenship eligibility to former colonial subjects (i.e., UK). In reaction to the strengthening of populist anti-immigrant voices, possibilities for legal entry for migrant workers, particularly for those from non-European countries, were restricted. However, some innovations of the post-war era had changed the conditions of nationally sovereign governance of migration. The Universal Declaration of Human Rights (1948), the International Refugee Convention (1951), and the emergence of the national welfare state – as the legitimate institutionalized form of organization of the political system that promotes chances for inclusion based on external closure – expanded the realm of those protected by substantial equality and solidarity (Jacobson 1996; Soysal 1994). Since then, the political semantics of solidarity and equality have been expanded to all those who contribute to the welfare of the nation, but at the same time reinforcing social differentiation and individualization. The nation state found itself caught in the "liberal paradox"; because it aimed to maintain control over its borders, it needed to resort to illiberal means, such as legal and political closure. In this sense, international human rights law and the protection of migrants by the welfare state impeded governments from deporting important numbers of unwanted temporary migrants even during economic downturns. In contrast, the state has found itself forced to allow family reunification and grant immigrants equal rights in a number of domains (Joppke 1998). In contrast to the earlier intention of labour rotation, migrants settled permanently in the Western countries as a sort of set of new ethnic minorities. Although provisions for the legal and social inclusion of established minorities were implemented and multicultural policies of recognition were slowly introduced in some countries, minorities often experienced discrimination in the labour market, in education and in housing for the above-mentioned reasons. Consequently, they often remained in the lower strata of the labour market, or they were channelled into welfare assistance. Hence, one secondary effect of the liberal paradox became palpable; immigrants, the former guestworkers and colonial subjects, even when included in the social security systems of receiving societies, faced important difficulties in joining the political realm that is characterized by equality of full citizens.

As a reaction to the global economic crisis of the 1970s, many Western industries outsourced production to former colonies in the frame of an international division of labour. Thus, new service economies were established in Europe, from where the global and flexible chains of production were managed (Castells 1996; Harvey 1990). Such outsourcing was increased after the fall of the communist regimes in Eastern Europe and the resulting increase in the commercial links between Western European countries and emerging countries in Central-Eastern Europe. The transformation of the European economy, giving space to more flexible forms of specialization, occurred at the same time as the Single Market Project was realized in Europe, which led to the abolition of borders in the Schengen area and the solidification of external borders through the Dublin Convention. Within the EU, a dual migration regime was installed that was believed to give rise to the more efficient allocation of labour within the member states. Similarly, this new international division of labour led to a curtailment of channels of legal entry into labour markets for so-called third country nationals, with the exception of highly qualified personnel.

In reaction to this debate, Swiss immigration policy was fundamentally reformed. The new immigration model formulated in early administrative papers established distinctions based on country of origin. The "inner circle" included the EU and EFTA states, for which free movement was provided. The express objective was to enable Switzerland to adopt a more European orientation because it was assumed that European labour and highly qualified persons from the "second circle" (USA, Canada, Australia, New Zealand, Israel and several small European states) would meet with less resentment than would those from the "third circle" (rest of the World), thus minimizing the continuous politicization of migration. Moreover, a statutory definition of integration was laid down for the first time, with corresponding measures that also considered the desire to ease naturalization.

In this context, a dual migration regime evolved that gave rise to important reconfigurations of extant patterns of inclusion and exclusion. Although migration by low-skilled working migrants was impeded or redirected to the asylum channel, highly skilled professionals were attracted as staff for the headquarter economies in the "global cities" (Sassen 1991). As an integral part of this dual migration regime and freedom of movement within the EU, the internal and external borders of the EU have been more strongly controlled and regulated by legal, technological and political means (Feldman 2012).

In Switzerland, this change was accompanied by the opening of the labour market through the Agreement on the Free Movement of Persons within the European Union, which entered into force in 2002 and constituted a watershed change in Swiss migration policy. The quotas and priority for Swiss nationals – the hallmark of corporatist immigration policy catering to sectoral and regional political interests since the 1950s – hence appeared to be outdated. This arrangement imposed a new

dynamic on the Swiss labour market, with its protectionist-corporatist contours, which had formerly been geared to the domestic economy. Concerning immigration into Switzerland by non-Europeans, by comparison, the new Immigration Act in effect since January 1, 2008, still contains restrictions prioritising labour urgently needed by the economy due to special qualifications. The referendum held on February 9, 2014, however, in which 50.3% of the eligible population voted in favour of stemming "mass immigration", has cast doubt on the freedom of movement negotiated with the EU in bilateral treaties. Although the voting called for quotas, to appease the European Union, in its implementation law, the Swiss Parliament ordered employers to prioritize Swiss residents rather than foreign workers in locations in which the unemployment rate is above average.

Summarizing, since the oil crisis and the acceleration of global trade in the 1970s, the Fordist consensus has been transformed as much as migration as a condition to let this regime work. One could argue that the spatial and political rearrangement of movement in Europe that has occurred since then has been driven by the fluid – and often contradictory – processes of global marketization and regional securitization and by the rising influence of human and migrants' rights regimes. In particular, this transformation must be viewed against the backdrop of the rise of the European service economy and a "new international division of labour". Connected to this larger process, one can observe the emergence of a dual regime of migration and mobility that, in the Swiss case, has been characterized by the establishment of the free movement regime applicable to the nationals of the EU/EFTA member states and, second, by restrictive rules for controlling the entry, admission and stay of third-country nationals.

#### **1.4 New Migration and Mobility Dynamics in Switzerland**

As mentioned above, the Migration-Mobility Nexus refers to the continuum between two types of movement; at one end, migration can be defined as a one-off, long-term or permanent movement from one place to another. At the other end, mobility can be defined as (a series of) multiple, temporary movements between different places. This nexus englobes the differentiation of inclusionary and exclusionary logics in the regimes, and the realities of migration and mobility have given rise to new dynamics in different domains. Those emerging in contemporary Switzerland are the following:

• Demographically, new patterns and types of migration and mobility are observed, i.e., the increase in highly skilled professionals and the rise of circular migration. However, the lack of the statistical infrastructure in Switzerland to identify those new types made necessary the provision of new instruments to grasp these new realities through the launching of the Migration-Mobility Survey (see Chap. 2 of this book) and the realization of this book.


Those different dynamics affect the transformation of a society in which the patterns of immigration are evolving quickly but are also diverse and in which the inclusion of migrants in the labour market and the society more generally differ according to their sociodemographic characteristics, in particular in terms of gender and origin. Due to the availability of new survey data and the development of statistics based on population registers that provide new tools to analyse migration (Steiner and Wanner 2015), Switzerland can be considered a laboratory in which to test how migratory patterns and trajectories have been shaped by the interplay of legal norms, economic drivers and societal factors. All three influences have been strongly pronounced in Switzerland, a country whose population today consists of 25% foreigners and almost one-third foreign-born individuals, shares that are even higher when the labour market population is considered or within the main agglomerations of Switzerland (Basel, Geneva, Zurich).

Since the late nineteenth century, high demand for foreign labour has attracted migrants to Switzerland, where they found one of the most liberalized labour market policies in Europe. In the last two decades, the implementation of the dual regime of migration and mobility led not only to a shift in the characteristics and the diversity of the flows but also to a significant increase in the demographic effect of migration. Currently, migration flows contribute directly to the relatively important demographic growth observed in Switzerland compared with its neighbouring countries (Fig. 1.1). However, they also affect the so-called natural increase by contributing to the natality of the population living in Switzerland. The growth rate of the Swiss population since the beginning of the twenty-first century has been three times higher than the one recorded in the EU-28/EFTA and has been significantly higher than that of all bordering countries. This situation, together with the increase in the proportion of migrants and their children in the population, is often referred to in political discourses, in particular prior to migration-related referendums or initiatives.

Although impressive, the Swiss case is representative of the situation observed in most industrialized countries. According to the OECD, all member states of the organization with accurate statistics on migratory flows recorded a positive net

**Fig. 1.1** Growth rates by components and of the total population, EU-28/EFTA countries, Switzerland and some of its neighbouring countries, 2000–2015 Source: Eurostat and Destatis, https://www.destatis.de/EN/PressServices/Press/pr/2013/05/ PE13\_188\_121.html. Accessed 29 September 2016

migration in 2015.2 However, small countries benefitted more from immigrants, relative to their population size, than did larger countries. It is therefore not surprising that in 2016, the immigration rate was the highest in Luxembourg, with 39 new immigrants per 1000 inhabitants, followed by Malta with 37‰, Switzerland being ranked sixth with a rate of 18‰.

The general increase in immigration flows to Switzerland but also to other EU-28/EFTA countries is, inter alia, also due to an increase in intra-European migration flows. In 2016, according to Eurostat, 6 of 10 migrants moving to Switzerland came from an EU-28 country, a proportion that is only higher in Slovakia, Iceland and Luxembourg (Fig. 1.2). With the disappearance of borders between nations in Western Europe, international migration progressively merges with internal mobility, and one can observe in some places, such as in Geneva, an increase in (residential) international migration, even among persons keeping their job in the city. Other forms of migration, some of them having lost importance since the 1970s, also re-emerged, such as temporary or seasonal migration. Finally, other mobility behaviours appear, for instance circular migration, supported by a better knowledge of job opportunities in connected labour markets.

In addition, the dual admission regime supplies Switzerland's service economy with the required highly skilled labour. The relationship between the labour market and migration being so important, the demands of the labour market profoundly affect the socioeconomic composition of the flows. In the last decade, 60% of adult

<sup>2</sup>OECD Migration Database, http://www.oecd-ilibrary.org/social-issues-migration-health/data/ statistiques-de-l-ocde-sur-les-migrations-internationales\_mig-data-fr. Accessed 28 March 2018.

**Fig. 1.2** Nationality of immigrants in EU-28/EFTA countries, 2016 (in %) Source: Eurostat Database, http://ec.europa.eu/eurostat/data/database. Accessed 5 May 2018

immigrants who have entered Switzerland held a tertiary level of education. In 2010, and in terms of the highly skilled migrant stock, Switzerland was ranked sixth out of all OECD countries, after the US, Canada, the United Kingdom, Australia and Israel.3 Both the share and the number of tertiary educated migrants have significantly increased in the country since the end of the twentieth century. Today, approximately 70,000 highly skilled adults enter Switzerland every year, primarily for job reasons (Fig. 1.3). In fact, due to the specialization and tertiarization of the Swiss economy since the beginning of this century, the native newcomers entering the labour market have not been able to fulfil the demand of the economy, increasing its demand for foreign highly qualified labour.

Nevertheless, the majority of migrants that arrived in the last 10 years in Switzerland did not have strong links with the country beforehand. In fact, fewer than 10% of the sample in the Migration-Mobility Survey were cross-border

<sup>3</sup>DIOC database 2000 and 2010, DIOC-E database 2000, http://www.oxfordscholarship.com/ view/10.1093/oso/9780198815273.001.0001/oso-9780198815273-chapter-2. Accessed 18 May 2018.

1 Today's Migration-Mobility Nexus in Switzerland

**Fig. 1.3** Level of education of immigrants according to the year of arrival, Switzerland, 1991–2014

Source: Own estimation based on the Swiss Structural Survey

commuters before migrating to Switzerland, and more than 90% had never lived in Switzerland before (Table 1.1). Moreover, less than one-third of migrants had relatives in Switzerland when they arrived. However, and compared with less-educated migrants, tertiary educated migrants are more frequently in their second (or more) stay in Switzerland but are less concerned by the presence of relatives in Switzerland prior to migration. These last results indicate that the context of immigration differs according to the educational characteristics of the migrants. The less educated are more frequently involved in family migration (including family reunification, but also migration driven by the family networks, as suggested by the new economy of migration (Stark and Bloom 1985)). However, the highest educated are more likely concerned with a mobility referring to the labour market, without the support of the family (and by extension without the support of for example a social network or friends in the destination country), but eventually relying on external reallocation services.

Unsurprisingly, spatial proximity improves professional links with Switzerland prior to the migration, as can be observed with French immigrants; more than 20% of the migrants were already cross-border commuters before their migration to Switzerland. The actual immigration to Switzerland represents for those migrants


**Table 1.1** Connections with Switzerland at the time of migration, 2016 (in %)

an extension of their prior working experience. Portuguese migrants, in contrast, generally arrive without a professional link but frequently with relatives already living in Switzerland (close to 60%), demonstrating the capacity to be supported in the migration trajectory by the family or the network.

In total, approximately 60% of the migrants surveyed mentioned professional reasons to move to Switzerland, whereas only 8% mentioned education or study reasons. Professional reasons are more frequently mentioned by men than by women, a result that confirms the gender dimension of migration. Gaining new experience, lifestyle or starting and accompany families (particularly among women migrants) are also reasons that are relatively frequently mentioned to explain the migration to Switzerland (Fig. 1.4). Although Switzerland mostly attracts professional migrants, the diversity of reasons mentioned suggests the complexity of the logics explaining today's patterns of mobility and migration. This complexity, which is one of the characteristics of the Migration-Mobility Nexus, will be explored throughout this book as a factor of inclusion and exclusion or of integration and discrimination.

**Fig. 1.4** Reasons mentioned to explain the move to Switzerland, 2016 (in %) Note: More than one response could be given, which is why the total does not equal 100% Source: Migration-Mobility Survey 2016. Weighted results

#### **1.5 Objectives and Key Issues of the Book**

The volume seeks to clarify *how economic drivers, societal factors and legal norms shape today's migration and mobility patterns in Switzerland.* Thus, it investigates not only how permanent and/or temporary today's migratory patterns really are but also how Switzerland's selective regime of migration and mobility influences existent patterns of social and professional inclusion and exclusion. The title of the book *Realities of Migration and Mobility* thus reflects the panoply of migratory patterns and realities from the perspective of immigrants. The notion of the Migration-Mobility Nexus, sketching a continuum between the two poles of migration and mobility at both a theoretical and an empirical level, serves as the conceptual and overarching framework for the present book.

To date, the consequences of the new forms of migration and mobility for the migrants themselves (in terms of social and structural integration) are poorly researched. This lack is largely due to the absence of adequate data. In fact, the adaptation of the statistical infrastructure to a new social phenomenon, such as these new emerging forms of migration and mobility, requires time. More specifically, the shift from (low skilled) labour migration to more diversified migratory flows took many national statistical systems by surprise. Information as diverse as the level of education and/or qualification of migrants, the duration of stay and the probability of return, and the social and structural integrations are often poorly documented in European countries.

Despite being one of the most important immigration countries worldwide in proportion to its population, Switzerland is also ill equipped in terms of monitoring systems and databases for tracking individual migration and structural and cultural integration. Focussing on recently arrived immigrants, the recently conducted Migration-Mobility Survey (see Chap. 2), on which the analyses of all chapters in this book are primarily based, closes this gap and thus allows for new insights into these new migration and mobility patterns, not only in Switzerland but also in other high-income countries. Therefore, the Migration-Mobility Survey has created a new instrument that not only informs about the mobility evolution in Switzerland but also is able to support future cooperative European research in this emerging field.

The mixed-mode survey (online and telephone) was conducted at the end of 2016. In total, almost 6000 foreign-born migrants, aged 18 years or older at the time of immigration, between 24 and 64 at the time of the survey, and who arrived in Switzerland within the last 10 years participated. Individuals holding the nationality of one of the following 11 countries/regions of origin were surveyed: Germany, Austria, France, Italy, United Kingdom, Spain, Portugal, North America, India, South America, and West Africa.

To address the interconnections between contemporary demographic patterns, legal regulations, political orientations and economic interests as much as the societal dynamics unleashed by a new mobility regime, the book covers different aspects of life in the host country covered by the Migration-Mobility Survey, including the family dimension, the labour market and political participation and social integration. Therefore, the collection is written from a pluridisciplinary perspective (anthropology, demography, economics, political sciences, and sociology), aiming to provide in-depth analyses of migration causes and consequences. It also considers the chronological dimension of migration by structuring the book according to the following four key issues: the migrants' arrival, their stay, and their expectations concerning return and naturalization:

#### *Part I: Migratory process and arrival in the host country*

As argued, the arrival in the host country is determined by a state-driven migration policy and an economically driven mobility regime. Nevertheless, these structural conditions are experienced as constraints or facilitators according to the migrants' agency, that is, their socio-demographic characteristics, such as level of education and origin. The first two chapters add to the Migration-Mobility Nexus by disentangling these processes at the time of immigration and by describing how they create differentiated pathways of inclusion and exclusion between different immigrant groups.

Within the context of a demand-driven admission policy in which employers play a central role for selecting candidates to migrate, Sandoz and Santi investigate the relocation support that employers provide to different groups of migrants (see Chap. 3). Using a mixed-method approach, combining the quantitative analysis of the Migration-Mobility Survey and an ethnographic study, their article thus examines who has more power to negotiate advantageous relocation conditions and, in this sense, represents a more "wanted" migrant for profit-oriented actors.

Zufferey investigates the underlying factors of both the migration trajectories and why frequent international movers have arrived in Switzerland (see Chap. 4). His chapter thus examines the heterogeneity in serial migration practices considering individual characteristics, such as the level of education, origin, and institutional dimensions.

#### *Part II: Labour force participation of migrants*

Not only the legal but also the economic context within which migration occurs contribute to the rapid integration of immigrants into the labour market. Due to the relatively favourable economic situation of the Swiss economy, compared with other EU countries (OECD 2015), the integration into the labour market of newcomers during the last decade was mostly rapid. However, the quality of this integration, in terms of use of skills or the level of satisfaction, is occasionally questionable, inter alia due to the acceleration and the diversification of the types of mobility. The second part of the book thus investigates how Switzerland's selective regime of migration and mobility influences patterns of professional inclusion and exclusion.

Wanner studies the extent to which the self-declared reason for migration affects the level of integration in the labour market (see Chap. 5). A special focus lies on the integration of secondary migrants (accompanying spouses) into the labour market. He thus investigates how the motive for migration and gender influence the position migrants have in the labour force and how this position triggers a need for groupspecific integration policies.

Analysing immigrants' labour market trajectory throughout their settlement in Switzerland, Vidal considers their employment situation in the country of origin before migrating and the characteristics of the family migration process (see Chap. 6). She also focusses on the gendered dimensions of the Migration-Mobility Nexus because she examines integration patterns of women and men in comparison.

Because evidence shows that immigrants experience difficulties in finding a job that matches their human capital endowment, Pecoraro and Wanner measure the incidence of educational and skill mismatches among different groups of the immigrant population according to their origin and the status of foreign credential recognition (see Chap. 7). They thus examine whether the recognition of foreign diplomas is a factor reducing the risk of skill or educational mismatch.

#### *Part III: Social life and political participation*

As for the labour market dimension, lines of exclusion and inclusion are being readjusted inter alia according to gender, class, and origin. The third part of the book thus investigates the immigrants' social lives and their political participation.

Based on several data sources, Bennour and Mantaschal investigate how cantonal norms of in- or exclusion, as expressed by cantonal integration policies and attitudes towards immigrants (xenophobia and right-wing voting), affect immigrants' national identity in terms of their feeling of attachment to Switzerland (see Chap. 8). By focussing on the subnational policy level, this study circumvents the narrow focus on the nation-state, which is predominant in classical migration studies.

Shedding light on the mechanisms of perceived discrimination (see Chap. 9), Auer and Ruedin investigate who, among recent immigrants, is more likely to feel discriminated against and report it when asked in a survey. They examine not only

the exclusionary logic of the societal sphere but also the inclusionary logic of the economic sphere and the opposition between the dual regime of migration and mobility and individual contexts.

Expanding the outlook on what counts as a political activity, Hercog investigates how various resources and forms of capital influence different types and aims of migrant engagement in Switzerland (see Chap. 10). In fact, although almost onefourth of the Swiss population has no political rights because they hold foreign citizenship, political participation can also involve other forms such as donations, petitions, or demonstrations.

#### *Part IV: Transnational life and future migratory expectations*

Since the early 1990s, the Anglo-American social science literature on transnationalism has grown steadily. Transnationalism refers to the construction of social fields in which migrants create a link – imaginary or real – between their home country and their host society (Fibbi and D'Amato 2008; Glick Schiller et al. 1992). In these transnational spaces, migrants develop social and economic relationships, activities and political identities that transcend traditional boundaries and benefit from global economic processes in a world divided into nation-states. In the field of migration, the approach of transnationalism has been posed as a third approach, between the option of assimilation and that of return and between a perspective in which individual action is overdetermined by macro-social constraints and a perspective that sees it as the result of the will and preferences of social actors.

In this context, Dahinden and Crettaz study how transnational today's migrants in Switzerland really are by considering the three dimensions transnational mobility, network transnationality and transnational belonging (see Chap. 11). The authors investigate the effect of migration regimes and social class on transnationality, mobility and migration.

Finally, and because immigrants' intentions summarize the respondent's attitude towards the migration experience in the host country, Steiner investigates how demographic characteristics, the feasibility of a further move, transnational ties and embeddedness in Switzerland determine different types of immigrants' intentions, that is, remigration, or settlement and/or naturalization (see Chap. 12).

#### *Part IV: Conclusions*

The fifth and concluding part reviews and summarizes the contributions of all chapters in light of the Migration-Mobility Nexus. Moreover, it provides information referring to the future challenges of the Migration-Mobility Nexus in European and other high-income country settings. Overall, the book thus demonstrates how the implementation of a nationwide survey, such as the Migration-Mobility Survey, leads to new knowledge about current migration processes and helps identify approaches and solutions that can enhance immigrants' integration when necessary. It is therefore relevant to both the scientific community and migration policy.

#### **References**

Castells, M. (1996). *The rise of the network society* (Vol. 12). Cambridge, MA: Blackwell.


Hall, S., & Du Gay, P. (1996). *Questions of cultural identity*. London: Sage.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Chapter 2 Surveying Migrants in Europe. Experiences of the Swiss Migration-Mobility Survey**

**Ilka Steiner and Aljoscha Landös**

#### **2.1 An Increasing Need for Data on International Migrants**

The need for information on immigrants and their migration trajectories has risen in recent decades due to an increase in their numbers, in their diversity with respect to their countries of origin and in their reasons for immigration. The need has also risen due to the increasing social, economic and political importance of the foreign population in all European countries, particularly in Switzerland (Font and Méndez 2013, p. 20).

As an alternative to censuses, register data and traditional population surveys, migrant surveys can satisfy this demand for several reasons. They allow for an indepth analysis of migration flows and migration-related topics such as transnational ties and attachment to the host country, lived discrimination or migratory or settlement intentions – topics that are not covered in censuses or traditional population surveys. Moreover, another shortcoming of traditional population surveys presents low number of migrants in the sample, inter alia because they are a "hard-to-survey" population (Tourangeau et al. 2014).

However, when organizing a survey among migrants, challenges concerning the survey setup are accentuated compared to traditional surveys; the difficulties were emphasized by Font and Méndez (2013), an edited volume reviewing the challenges of migrant surveys performed in different countries and thus contexts. Namely, underrepresentation of certain subpopulations due to under-coverage and nonresponse poses a serious challenge and thus requires thorough considerations and anticipation.

nccr – on the move, Neuchâtel, Switzerland e-mail: Ilka.Steiner@unige.ch; Aljoscha.Landoes@unige.ch

I. Steiner (\*) · A. Landös

Institute of Demography and Socioeconomics IDESO, University of Geneva, Geneva, Switzerland

<sup>©</sup> The Author(s) 2019 21

I. Steiner, P. Wanner (eds.), *Migrants and Expats: The Swiss Migration and Mobility Nexus*, IMISCOE Research Series, https://doi.org/10.1007/978-3-030-05671-1\_2

Despite the methodological challenges, several surveys were launched in recent years in European countries, inter alia the Swiss Migration-Mobility Survey.

Thus, this chapter reviews in Sect. 2.2 the methodological challenges when organizing a survey targeting migrants by discussing the two main components of underrepresentation, which are under-coverage and non-response. Section 2.3 provides an overview of specific migrant surveys that were undertaken in Europe and how they have addressed some of the methodological challenges, while Sect. 2.4 reviews the survey setup of and the coverage and non-response bias in the Swiss Migration-Mobility Survey. The chapter closes with a short overview of the main sociodemographic characteristics of the participants (Sect. 2.5), serving thus as a basis for the following chapters of this book, and a short preview of the planned second wave of the Migration-Mobility Survey (Sect. 2.6).

#### **2.2 Under-Representation, the Main Challenge When Surveying Migrant Populations**

A major difficulty in the production of data on migrants and migration concerns the differing definitions, concepts and measurement parameters used, as emphasized in a study based on data from the migration module of the European Social Survey (ESS) (Davidov et al. 2018). Foremost, the definition of the "target population" varies sensibly from one study to another. Depending on the research goals and financial constraints as well as the legal concepts of migrants in the host countries (Font and Méndez 2013), "migrants" can be defined based on, for example, their nationality, foreign-birth status, recent moves, ethnicity, duration of stay, or reason for immigration. Even when similar or the same definitions are applied, the migration and/or naturalization policies must be considered when interpreting the constructed categories. Thus, the access to citizenship varying from one country to another, defining "migrants" through the nationality may not have the same signification and yield comparable results. This particular challenge is accentuated in multi-site surveys, that is, when several places in sending and receiving countries are involved, to study migration as a dynamic phenomenon (Beauchemin and González-Ferrer 2011).

Additionally, migrant populations share several characteristics of "hard-tosurvey" populations (Tourangeau et al. 2014): sampling difficulties, higher mobility, multiple languages, vulnerability and occasionally also reluctance to participate. All of these characteristics can, depending upon the chosen definition, lead to under-representation of certain groups due to under-coverage (at the sampling stage) and/or unit non-response (at the participation stage). For example, whereas some migrants will be difficult to contact due to their high mobility (i.e., highly skilled migrants or expats), others will be difficult to interview due to cultural and/ or language differences.

In any case, the survey methods must be adapted to these heterogeneous populations and contexts, not only causing important methodological challenges but also ultimately rendering the production of comparable data very challenging.

#### *2.2.1 Under-Coverage*

Coverage biases refer to a non-observational gap between the target population and the sampling frame1 (Groves et al. 2009). Under-coverage occurs when members of the target population are missing in the sampling frame. And according to Groves et al. (2009, p. 70), "perfect frames do not exist; there are always problems that disrupt the desired one-to-one mapping" between the frame and the target population.

Moreover, in some cases missing information hinders performing the desired survey mode or properly specifying the migratory status. Concerning the former, population registers rarely include email addresses, preventing the invitation of participants or the administration of a survey by email. Additionally, in many countries for which a sampling frame is available, the landline telephone coverage is rather low or insufficient. According to Lipps et al. (2013, p. 3), since the 2000s, Swiss survey institutes have faced those challenges due to "mobile-only" households, the increasing replacement of landline service by cell phones and the abolition of the mandatory registration of phone numbers in the public directory in 1998. Concerning missing information on the migratory status, data on ethnicity are very rare in official statistics and the migratory status is mostly defined by the country of origin or the nationality. In some countries, migrants are only registered when intending to reside for at least 1 year in the country (e.g., Sweden).

In the absence of an adequate exhaustive sampling frame, alternative sampling methods can be used such as randomly dialled telephone numbers or random routes.2 Nevertheless, because these methods are not very cost effective, particularly when targeting "rare elements" (such as migrants), other strategies, such as snowball sampling, are proposed to identify the target population. For a comprehensive list of alternative sampling strategies, refer to Reichel and Morales (2017).

However, alternative sampling strategies can introduce a representativeness bias. For example, and many others, snowball sampling must be treated with caution with respect to the representativeness of the target population due to differing probabilities of being selected (Reichel and Morales 2017; McKenzie and Mistiaen 2009). Additionally, sampling strategies using phone directories can introduce a selection bias when people with published phone numbers differ from those whose phone number is not available. In fact, landline phone numbers are more often available than are cell phone numbers. In addition, the level of coverage of landline phone numbers depends on the municipality size (the more urban, the fewer phone numbers listed) and the matrimonial status (singles less often have a number listed than married persons) (Lipps and Kissau 2011). Due to their over-representation among young, single, urban dwellers, migrants are thus more exposed to the problem of under-coverage. Camarota and Capizzano (2004), cited by Lipps et al. (2013), also

<sup>1</sup>A sampling frame is the source material from which a sample is drawn. It is a list of all those within a population who can be sampled (elements) and can include individuals, households or institutions. A typical sampling frame is the population register.

<sup>2</sup> Interviewers are assigned with a starting location and provided with instructions regarding the direction or the side of the street as well as the selection procedure of the households.

mention the instability of foreign minorities' living arrangements and residential situations as a cause of their under-representation (Blohm and Diehl 2001; Morales and Ros 2013). According to Lipps and Kissau (2011), "national minorities usually have a lower likelihood of owning a landline telephone and – even if they do – being listed in the telephone book".

One possibility of reducing under-coverage consists of proposing either a survey mode that includes all target population elements – for example, a pencil-paper survey when postal addresses are accessible –, a mixed-mode approach, or a multiple frame design, where missing phone numbers in the population registers are added from telephone directories. Based on a sample drawn from the Swiss population register, Lipps et al. (2013) showed that by matching information from commercial telephone directories and external sources (e.g., from the internet), not only was the proportion of missing phone numbers for the frame population reduced from 24% to 14% but the socio-demographic representation of the population was also improved. In particular, they managed to limit the under-representation of the younger age categories, singles, divorced and residents of larger cities.

#### *2.2.2 Unit Non-response*

A unit non-response bias occurs in case of a non-observational gap between the sample and the respondents and is particularly problematic when it is correlated with the variables of interest of the study (Deding et al. 2013, p. 173). Non-response can be explained by different reasons: for example, refusal,3 inability to participate due to disability or illness, or non-contact.

Although social and political surveys face increasing non-response rates, the problem is aggravated when surveying migrants due to their lower accessibility (Font and Méndez 2013; Martin et al. 2016). Little is known thus far because few studies have focussed on unit non-response among immigrants, and not all of them have separately analysed non-contact and refusal (Deding et al. 2013, p. 173). In fact, due to the high mobility of several migrant groups, availability depends upon the time spent at home and is reduced by professional, educational and leisure activities (Abraham et al. 2006). For these reasons, young individuals, singles and urban dwellers are more difficult to contact (Stoop 2004; Stoop et al. 2010). Among the explanations of lower response rates among migrants are the increased prevalence of irregular work schedules (Feskens et al. 2006; Deding et al. 2008), longer absences due to visits to the home country (Blohm and Diehl 2001), a younger age structure (Ette et al. 2015) and more-urbanized locations (Morales and Ros 2013). Locating individuals, particularly highly mobile persons, at the address indicated in the sampling frame therefore also depends upon the time elapsed between the drawing of the sample and the fieldwork.

<sup>3</sup>For a list of factors influencing willingness, see p. 170 of Groves et al. (2009).

But, inciting individuals to participate depends, particularly for migrants, upon language competences and thus the proposed languages of the survey (Deding et al. 2013), trust in interviewers and the survey institute (Martin et al. 2016), and topic interest (Groves et al. 2009). Concerning certain host country-related topics, migrants might not feel sufficiently concerned to participate, or they might be inhibited from speaking openly about the topics. Moreover, migrants living alone or being socially isolated (Deding et al. 2013; Stoop 2005) are less disposed to participate in a survey. Hence, duration of stay, and eventually naturalization, might increase participation because of better integration (Deding et al. 2013, p. 177).

Another solution to increase willingness can present unconditional (not dependent upon participation) or conditional (only once the questionnaire is filled out entirely) incentives. Nevertheless, the results concerning the effectiveness of incentives are mixed (Bosnjak and Tuten 2003; Singer and Ye 2013; Porter and Whitcomb 2003; Göritz and Wolff 2007). Several studies have shown that the amount, the type, and the timing of the incentives can be decisive for participation in web surveys targeting the general population (Gajraj et al. 1990; Sánchez-Fernández et al. 2010; Scherpenzeel and Toepoel 2012). However, according to Morales and Ros (2013, p. 153), "it is not clear that monetary incentives necessarily increase cooperation rates for immigrants, even if they appear to be successful with native respondents". A recent feasibility study in Austria showed that incentive strategies are particularly important for panel surveys because of the necessity to keep the respondents as long as possible in the panel (Hosner and Schlechter 2015). A German study on the reaction of ethnic minorities to different incentives has shown that investment in a double incentive strategy (conditional and unconditional) reinforces reciprocity and trust in the usefulness of participation (Fick and Diehl 2013). Furthermore, a survey from the United Kingdom showed that respondents who received an incentive (in the form of a stamp book) were more likely to cooperate than were those who did not. Finally, concerning the influence of incentives on participation, they found no significant differences between sub-groups (regional or ethnic minorities) (Department For Communities And Local Government 2012).

The participation can also vary or depend on the survey mode that is proposed. According to Czaja et al. (2014), telephone surveys remain among the most widely used methods of data collection. Among other reasons, they permit reaching people who lack access to the internet or to a computer or who have difficulties with writing and/or reading. In addition, the interviewer can convince some reluctant individuals to take part in the survey. The same is also true for face-to-face interviews. In contrast, self-administered surveys, for example, online surveys, increase the participation of difficult-to reach or unreachable individuals and offer a greater flexibility concerning place and time of participation.

The main challenge of online surveys refers to the digital divide, in which access to and the use of the internet vary substantially between groups (Dillman et al. 2014). Highly educated, computer-literate individuals with more up-to-date equipment (Czaja et al. 2014), younger age categories (Ette et al. 2015) and men compared with women are more likely to complete an online questionnaire. According to Martin et al. (2016), little is known concerning access to and the use of the new technologies by people with an immigrant background. However, one could argue that migrants present a higher affinity with modern communication tools and the internet because these technologies facilitated not only planning the move from one country to another but also maintaining contact with family and friends in the origin country (Ette et al. 2015; Mau and Mewes 2007).

A solution to limit unit non-response is to propose a mixed-mode design. Because financial constraints are among the most important limitations concerning mode choice, one possibility is to begin with a cheaper mode of data collection and then to use a more expensive mode for the non-respondents from the first mode (sequential mixed-mode). Another possibility consists in proposing two modes simultaneously and from the beginning (concurrent mixed-mode). When the combination of computer-assisted telephone interviews and a face-to-face or mail survey exceed the budget, an online survey appears to be the most cost-effective alternative, not least because of important cost-saving potential for the data-entry procedure (Dillman et al. 2014).

Nevertheless, when proposing a combination of self- and researcher-administered questionnaires, measurement bias can occur when, due to differing degrees of interviewer involvement, one group is better (resp. less) informed than is the other or participants are inhibited from answering more-sensitive questions when talking to a researcher (Groves et al. 2009, p. 141). Moreover, every mode has its own tradition of question formats, which can affect response distribution. Thus, measurement equivalence needs to be ensured by questionnaire equivalence, that is, question design, layout and instructions used in the main mode needs to be adapted to the auxiliary mode (see Dillman et al. 2014, pp. 232–240).

#### **2.3 European Migration Survey Landscape**

Thus far, long-term integration of immigrants has primarily been addressed by qualitative non-representative surveys at the local level (Jacobs 2010). However, in recent years, quantitative surveys have been launched in European countries (Mendez and Font 2013) to cover the scientific and political need for adequate data among migrants or foreigners. What follows is an overview of the such surveys. This inventory is the result of a rigorous search. On the one hand, we consulted PROMINSTAT,4 a compilation of meta-information on statistical datasets on migration, integration and discrimination in 29 European countries, which was financed by the European Union. In addition, we consulted the Swiss Centre of Expertise in the Social Sciences (FORS) Database,5 which refers to international surveys that are conducted for Switzerland. On the other hand, we performed an online search

<sup>4</sup>Prominstat database, http://www.prominstat.eu/drupal/node/64. Accessed 31 October 2017.

<sup>5</sup>FORS-Database, http://forscenter.ch/de/our-surveys/. Accessed 15 December 2017.

including a keyword6 search on Google and Google scholar and an analysis of comprehensive overviews of survey methods (Mendez and Font 2013; Bonifazi et al. 2008; Reichel and Morales 2017).

We only considered quantitative surveys that were performed after the year 2000.7 In addition, we excluded surveys targeting children of immigrants and undocumented migrants, refugees and asylum seekers because the survey methodology that must be deployed to reach them varies significantly from other migrant categories.

The final list of surveys described in this chapter (Table 2.6 in the Appendix) includes a majority of surveys among migrants in Europe. Nevertheless, it is not exhaustive, particularly because small surveys at the regional level are missing.

Depending upon the general research goal, the specific demographic and thematic foci and financial and time constraints, varying survey methods were deployed. First, we can distinguish between *migration modules added to existing surveys and specific migration surveys*. Modules allow for an extension of the original questionnaire and therefore a better coverage of migration-related topics, a strategy that also has a positive effect on the response rate because migrants may feel more concerned with the questionnaire and thus the survey in general. Examples include the migration and minority module of the European Social Survey (ESS), the module on the labour market situation of migrants and their immediate descendants of the European Labour Force Survey (EU-LFS), and the Migration Module of the German Socio-Economic Panel (IAB-SOEP). Among the immigrant surveys that were identified, we find several multi-national surveys, such as the European Union Minorities and Discrimination Survey, Immigrants and Ethnic Minorities in European Cities: Life-courses and Quality of Life in a World of Limitations (LIMITS), the Six-Country Immigrant Integration Comparative Survey (SCICS), and other specific national surveys.

One of the major challenges when surveying a specific population is to define a *sampling strategy that guarantees the representativeness of the target population*. In a few countries, such as Switzerland or Sweden, population registers allow for national coverage and if necessary also the oversampling of specific groups (e.g., Longitudinal Internet Studies for the Social Sciences (LISS) Panel and LISS Immigrant Panel). The oversampling strategy is also an often-applied strategy in already-existing surveys, even more so when a migration-related module is added (e.g., IAB-SOEP or SLFS) (Kraler and Reichel 2010).

In cases in which exhaustive sampling frames are missing (or are out-dated), other strategies must be developed. Several countries base their sampling on census data. Nevertheless, due to an absence of addresses and telephone numbers in the census data, this strategy implies, in a second stage, a time- and cost-intensive

<sup>6</sup>Key words used and combined in the online search for surveys: Survey, immigrants, ethnic minorities, integration, labour market, highly skilled migrants, living conditions, migration, foreigners, migrant sample, migrant modules, hard-to-reach, hard to survey, longitudinal or cross-sectional survey.

<sup>7</sup>The only exception is the NIDI/Eurostat Push and Pull Factors of International Migration, which occurred in 1996/1997.

household screening (e.g., Spanish National Immigrant Survey) or randomly dialled telephone numbers (e.g., France in the Immigrant Citizen Survey). Other projects randomly selected migrant groups in centres of aggregation8 (for instance, in Milan in the LOCALMULTIDEM project) or selected individuals based on their names from the telephone directory (Six Country Immigrant Integration Comparative Survey or Lyon in the LOCALMULTIDEM project) (Lynn et al. 2018). The NIDI/ Eurostat Push and Pull Factors of International Migration project combined different sampling strategies, first, by using a disproportionate sampling in order to oversample sample areas with a high prevalence of migrant households and second, by adding a snowball sampling procedure in cases in which few migrants were found in the screened areas (Groenewold and Bilsborrow 2008). The latter allowed boosting the response rate of migrants and thus to complete strata characterized by a high-nonresponse (Reichel and Morales 2017). Also, targeting a specific population, as was for example done in Europe in the Pew Global Attitudes Project regarding Muslims in Europe (Pew Research Center 2006) where they oversampled Muslims, can reduce the coverage problem. This approach allows focussing on a highly relevant topic for the surveyed population and therefore reach a higher response rate. However and more generally, when the target population is too narrowly defined, an absence of specific characteristics in the sampling frames might prevent accurate sampling (e.g., religion, occupation and/or education).

Even though international migration is a *transnational phenomenon,* most surveys are performed in the destination country, while some investigate the origincountry perspective. A very few actually collect information on both sites to obtain a global picture of migration (Beauchemin et al. 2013). The latter is particularly subject to sampling problems because it requires sampling in both geographical places. In addition, the sampling frame must be representative of the host and sending country, and thus reaching a sufficient number of individuals in the final sample is even more challenging.

The MAFE project (Migration between Africa and Europe) is an example of a multi-site survey, covering three countries of origin in Africa (Senegal, DR Congo and Ghana) and six destination countries in Europe (France, Spain, Italy, Belgium, Great Britain and the Netherlands). Due to missing sampling frames in most of the countries (except for Spain) and to obtain a representative sample of the target population, several strategies were applied. Although in the countries of origin individuals were selected using Census Data or the 2007 DHS Survey Data, in the destination countries, quota sampling methods (age, gender and target areas) and complementary methods such as origin-based snowballing or time-location sampling methods9 were applied (Reichel and Morales 2017; Schoumaker et al. 2013).

Transnational snowball sampling in the context of African-European migration has an important representativeness bias because migrants with closer links with their migrant community are over-represented (Beauchemin and González-Ferrer

<sup>8</sup> Identifying a list of places (centre of aggregations) where the target population concentrates. The final selection of the respondents at the identified place is random.

<sup>9</sup>Method used to sample hard-to-reach populations by choosing locations where the target population concentrates (at different times) (e.g., Centres of aggregations).

2011). Thus, results might be biased such that the effect of migration on transnational financial transfers, poverty in households at the origin, and the influence of previous migration experiences on the individuals' propensity for out-migration are all overestimated.

Another cross-sectional multi-country survey is the already mentioned NIDI/ Eurostat Push and Pull Factors of International Migration. This survey was conducted in Italy and Spain (both countries of destination), Morocco, Egypt, Turkey, Ghana and Senegal (countries of origin). The survey was also subject to the same three constraints: geographical dispersion of migrants, migrants being rare elements, and the lack of sampling frames. Thus, the researchers identified and oversampled regions with a high density of migrants (Schoorl et al. 2000; Groenewold and Bilsborrow 2008).

Despite an increasing *effort to launch longitudinal surveys*, most of the immigrant surveys are cross-sectional. However, to measure social and structural integration, longitudinal surveys appear more suitable (Jacobs 2010). Moreover, cross-sectional surveys often have a validity problem due to method variance biases (variances that are due to the measurement method) and the impossibility of causal insights. Both methodological biases could be reduced by using longitudinal data because of the separation by time of the outcome and predictor, which guarantees a causal inference and minimizes the bias due to respondent moods or response styles and the survey context (Rindfleisch et al. 2008; Podsakoff et al. 2003).

Although, due to regular follow-ups, longitudinal surveys require higher financial and time investment compared with cross-sectional surveys, a handful of migration-related longitudinal surveys have been launched in Europe, such as the Longitudinal Survey of the Integration of First-Time Arrivals (ELIPA, France) and the Dutch LISS Panel.

The Dutch LISS Panel started in 2008. To ensure adequate coverage, the survey used a register-based sample and selected individuals were initially contacted in person or by phone for a 10-min interview. Then, participants could switch modes and continue online. To include participants without internet connection, access equipment was loaned (Scherpenzeel 2011). In 2008, no particular attention was given to the foreign population (no oversampling strategy and Dutch as survey language), which led to their lower participation, in particular among non-Western migrants. The LISS Immigrant Panel, which followed-up in 2010, stratified the population by country of origin, including first and second generations of four non-European immigrant groups (Moroccan, Turkish, Surinamese and Antillean) (Das 2012; De Vos 2010).

The French ELIPA study started in 2010 and was repeated in 2011 and 2013. This large-scale survey aimed at gathering information on recently arrived migrants from a longitudinal perspective. The survey was translated into numerous languages – Arabic, Turkish, Chinese, Russian, English, Serbian, Tamil, Bengali, Spanish, Vietnamese, Thai, Albanian, Soninke and French – a strategy that allowed covering 95% of the migrant population in France. Like the LISS Panel, the longitudinal approach of the survey required long-term adherence of the participants,

which was supported by telephone and personal pre-recruitment and a retention strategy through information letters between waves (Panel Care).

Another challenge of longitudinal surveys refers to the risk of attrition (dropout). As discussed, dropouts due to non-response in case of high mobility behaviour can lead to biased right-censuring (Rindfleisch et al. 2008). An intensive Panel Care and a tracking strategy to re-contact individuals who have moved is needed to guarantee a satisfying response rate for the follow-ups (Lee 2003). The ELIPA survey reached for instance a response rate of 78% for the second wave. However, other nonmigrant-specific surveys have shown that over a long period, the attrition increases significantly; in the case of the German Socio-Economic Panel, for example, the response rate reached in the first wave in 1984 was 70% but dropped to 25–45% in 2004.

Time-and-money-consuming longitudinal surveys are occasionally also replaced by retrospective questionnaires in cross-sectional surveys – for example, in the surveys Parcours et profil des migrants (PPM), Trajectoires et Origines (TeO), and NIDI/Eurostat Push and Pull Factors of International Migration of, or the Immigrants and Ethnic Minorities in European Cities: Life-courses and Quality of Life in a World of Limitations (LIMITS). The last is a longitudinal and comparative survey launched in 2005/2006 in six European cities: Amsterdam, Bielefeld, Lisbon, Rotterdam, Stockholm and Vienna. It aims to describe migrants' life trajectories and includes migrants from the cities' four main sending countries (Turkey, Morocco, Serbia, and Cape Verde) who lived for at least 15 years in the receiving country. To ensure comparability of the data, experts of all countries proposed a cross-national sampling design but with country-specific sampling strategies.10 LIMITS includes a retrospective questionnaire based on a life-history calendar that identifies specific events in a visual and user-friendly form. Nevertheless, this approach can be problematic due to the whitewashing, or downgrading, of subjective impressions (i.e., the satisfaction of the move to the host country) and to the participant's incapacity to remember past events (Blossfeld and Rohwer 2002; Glasner and van der Vaart 2009; Latcheva et al. 2006).

A further challenge concerns divergent sensibilities with respect to the adequate survey method due to the diversity of the migrant population in terms of nationalities, age, education and socioeconomic position. The *choice of the survey mode, incentives and proposed language(s)* might have a decisive effect on the participation of migrants (Hosner and Schlechter 2015) and thus on the representativeness of the survey.

Although the design varies considerably from one survey to another, most of the surveys chose computer assisted face-to-face interviews (e.g., Immigrant Citizen Survey, German Socioeconomic Panel, Longitudinal Survey of the Integration of First-Time Arrivals in France, and Spanish National Immigrant Survey) or a mixedmode approach (e.g., LISS Panel (online using particular designs e.g., Smartphone, GPS and telephone) and Diskriminierungserfahrung in Deutschland (face-to-face

<sup>10</sup>Random sampling through registers (Amsterdam, Rotterdam and Bielefeld) or snowball sampling, imposing different entrance points.

and online)). The Italian Survey on Social Conditions and Integration of Foreign Citizens of 2012 (SCIF) added a self-administered web survey to the paper-pencil mode to reach young participants. A strategy that not only allowed interviewing several persons simultaneously but also reducing the duration of the interview (Conti et al. 2017).

Few surveys use incentive strategies (e.g., the Generation and Gender Survey (GGP) or the LISS Panel). For instance, the LISS Panel used a prepaid incentive strategy to increase participation rates, adding to the first invitation letter ten Euros and a promise of an additional ten Euros when completing the questionnaire. According to the researchers, a higher amount of money did not significantly increase the response rate (Scherpenzeel 2011). In the end, the response rate for the first wave was 75%, and for the second wave 48% (Scherpenzeel 2011).

Finally, the language(s) proposed in a survey might influence participation willingness. As done in most migration surveys and modules, the Migration Module of the German Socio-Economic Panel in 2013 and 2015 (IAB-SOEP) proposed for example a questionnaire in the predominant languages of the target population (English, Polish, Turkish, Rumanian and Russian).

A last major and increasing challenge in a globalizing world relates to the *production of comparable data at the European level*. As already mentioned, definitions of migrants can vary across Europe, which hinders or at least impedes such conjoint data collections.

A major effort gathering data to evaluate the European integration policy was made by the cross-sectional and transnational Immigrant Citizen Survey (Huddleston and Dag Tjaden 2012). The survey occurred in 15 cities in seven EU states (Belgium, France, Germany, Hungary, Italy, Portugal and Spain) and included migrants who have lived for at least 1 year in the destination country and who held a residence permit. In addition, the sampling strategy, which was inspired by the European Social Survey (ESS), was based on either the country of birth or the nationality, using the best national sources available (register data or censuses).

Another comparative survey, the European Union minorities and discrimination survey (EU-MIDIS), collected data on discrimination faced by ethnic minorities in the EU. Due to the varying proportion of the immigrant population in the different European countries, the geographical coverage varies from one country to another (cities, nationwide or metropolitan areas). To be included in the survey, two main criteria had to be fulfilled: high vulnerability and membership in the largest immigrant groups. The sampling strategy, to survey 500 individuals per group, was adapted to the particularities of each member state, varying from register data, random route sampling, and focussed enumeration11 to network sampling (European Union Minorities and Discrimination Survey (EU-MIDIS) 2009).

As shown by this short review, all particularities and challenges when surveying migrants were handled in different ways. Standard survey procedures must be

<sup>11</sup>To render the random route procedure more efficient, identified persons are asked to "map" the neighbours that correspond to the target population and therefore to identify easily areas where migrants live.

adapted to the population under study and the specific conditions of the country and project-related budget and time constraints. Section 2.4 of this chapter explains how these issues were addressed when setting up the Migration-Mobility Survey.

#### **2.4 The Migration-Mobility Survey**

Due to the economic and social importance of recently arrived and highly skilled migrants in Switzerland and to the lack of systematic research on their behalf, the nccr – on the move performed the Migration-Mobility Survey in 2016.12

The survey covers the main origin groups of recently arrived migrants (up to 10 years). It gathers information on the migratory history, citizenship intentions, education and employment history and current situation, family configuration and household situation, integration (language skills, personal network and transnational ties, leisure activities, and civic engagement), and life in Switzerland (see Table 2.7 in the Appendix for an overview of the questionnaire). The questions were adapted from international and/or Swiss surveys, allowing the researchers to compare their results with other survey data. The survey focussed on the most numerous immigrant groups, including German-, French-, Italian-, English-, Spanish- and Portuguese-speaking nationals. To reach those groups adequately and to guarantee the representativeness of each group, a mixed-mode approach of an online survey and telephone interviews proposed to occur in all six languages was selected. The survey was conducted by the LINK Survey Institute between October 7, 2016, and January 9, 2017. Ultimately, the goal set was largely outreached, with 5973 completed interviews, of which 97% were conducted online.

What follows is a short description of the survey setup (target population, sampling, and design), the data collection and response rate analysis, and a brief presentation of the participant's socio-demographic characteristics.

#### *2.4.1 Target Population*

It was decided to propose a questionnaire in six languages to cover the most numerically important foreign-born groups of Switzerland. Table 2.1 provides an overview of the nationalities, the attributed languages and their share in the total foreign-born population. First, considering German, French and Italian – which are also the three main languages spoken in Switzerland – allowed us to include the most important immigrant groups from Switzerland's neighbouring countries (Germany, Austria, France and Italy). Second, by including English, we were able to reach North

<sup>12</sup>Further information can be obtained on the webpage of the nccr – on the move (http://nccronthemove.ch/research/migration-mobility-survey/) or on FORSbase, a Swiss data-access portal (https://forsbase.unil.ch/project/study-public-overview/14592/0/). Both accessed 14 June 2018.


**Table 2.1** Geopolitical region, origin groups, attributed language and share in the total foreignborn population

Note: a The share in the total foreign-born population considers other selection criteria (age at immigration and the time of the survey, specific resident permit and max. ten years of residence) Source: Migration-Mobility Survey 2016

American (US/Canada) and Indian citizens. Third, Spanish and Portuguese, the mother tongues of further numerically important immigrant groups originating from Spain and Portugal, were included. Based on these six languages, it was further decided to include for comparison two groups from non-industrialized and non-EU/EFTA countries: West Africa (French-, Portuguese- or English-speaking countries) and South America (Spanish-, English- or Portuguese-speaking countries).

The survey focussed on immigrants who had arrived in the last 10 years (after June 2006) and held a resident permit (B), settlement permit (C), short-term permit (L), were diplomats/international civil servants or the latters' family members (Ci) at the time of immigration. Due to the survey's aim and methodology, only "voluntary" migrants were considered and thus asylum seekers (N), provisionally admitted persons (F), and persons without a legal residence permit were excluded. Not only was the questionnaire not adapted to their specific migratory trajectory and life situation, but the methodology (for example, with respect to languages and sampling frame) would also have required alteration to reach and survey these specific populations.

Finally, only immigrants aged 18 years or more at the time of immigration – presuming that minors do not independently decide on their migration – and aged between 24 and 64 years at the time of the survey were included. Based on these criteria, the 11 selected groups represent 68% of the total foreign-born population.

#### *2.4.2 Sampling*

The Swiss Federal Statistical Office's (SFSO) sample register (SRPH),13 which is drawn from the harmonized registers of persons of the Swiss Confederation, the cantons and the municipalities, served as the sampling frame. The frame is an exhaustive list of persons living in Switzerland, providing the above-mentioned inclusion criteria for the target population. Finally, register-based sampling allows the calculation of post-stratification weights.

To reach representative samples for the 11 origin-groups, a stratified random sampling strategy was applied. Concerning simple random sampling, each element in the sample has a known and nonzero probability of being chosen. However, key subgroups are represented, guaranteeing greater precision and improving therefore the potential for the units to be more evenly spread over the population. Concerning recently arrived immigrants (less than 2 years of duration of stay) and to anticipate their lower participation rates and higher attrition rates when conducting a second wave (see Sect. 2.7), we decided to oversample individuals who had recently arrived in Switzerland. Therefore, the representation of the length of stay (up to 2 years and 2–10 years) and both genders were ensured by a disproportionate stratification. The West African group presents an exception because the size of this population does not allow a stratification based on the duration of stay. In total, we defined 42 strata.

The targeted sample size for every group was set between 352 (West Africa) and 384 (Germany) interviews, setting 4130 desired interviews. Based on a recent survey with German migrants in Switzerland (Steiner, Ilka. 2017. Immigration and settlement? German Migration Flows to and from Switzerland under the Provision of Free Movement of Persons. Unpublished PhD-Thesis, University of Geneva), we drew 13,660 addresses, that is n × 3.5 for all groups, except for the Portuguese (n × 6) because studies in Switzerland have shown lower response rates for this group (Fibbi et al. 2010).

Moreover, reserve samples for all 11 sub-groups (n = 6476) with a similar structure as the main sample were drawn. These additional addresses aimed to guarantee the achievement of the survey's minimum sample size and at ensuring a sufficient number of interviews for each stratum in the event of too-low response rates. The

<sup>13</sup>Statistikerhebungsverordnung https://www.admin.ch/opc/de/classified-compilation/19930224/ index.html. Accessed 15 March 2017.

samples were drawn by the SFSO in early September 2016 based on the list of address available in the register at the end of June 2016.14

#### *2.4.3 Survey Design*

Due to the low coverage of phone numbers for our target population in the sampling frame (16%), we opted for a mixed-mode approach in which the collection of information was primarily centred on an online questionnaire (CAWI) that people could complete using different electronic devices. This approach was complemented with telephone interviews (CATI) on request from the participants.

As a first step, all selected persons received an invitation letter asking them to participate in the online survey. The letter provided a link and a personalized access code, allowing persons to access the online questionnaire directly. Although favouring the online survey, the letter also mentioned the possibility to participate by calling a toll-free hotline, either to convene an appointment for an interview or to participate directly in a telephone interview. In addition, throughout the fieldwork, people contacting the hotline could still participate by CAWI. In this case, the person was asked to indicate his/her email-address and then received an email containing the online access link immediately after having called the hotline. Finally, when people only partially completed the online survey, their username and personal code allowed them to return to the questionnaire and to resume participation from the last question before quitting. However, once the questionnaire was completed, the personal access was closed.

To stimulate participation, two additional letters were sent to non-respondents (see Table 2.2): a first reminder approximately two and a half weeks after the invitation letter and a second reminder approximately two and a half weeks after the first reminder.

Each selected person received the invitation letter in two languages: the main language of their country of nationality and, if different from the latter, the main language spoken in the municipality of residence.

Concerning the incentives strategy, three tablets were randomly attributed to participants agreeing to participate in the draw. Moreover, a flyer describing the research purpose accompanied the first invitation letter. Moreover, additional information was made available on the survey's webpage (http://nccr-onthemove.ch/ research/migration-mobility-survey/). Finally, a newsletter to keep participants, researchers and other interested people informed was set up. The information material (invitation letters, flyers, and webpage) was also available in all six survey languages.

<sup>14</sup>Moreover, an additional sample of 600 addresses was drawn in July 2016 for a pilot. The pilot's aim was to test the general reception of the survey, the programming and filters, and the length of the questionnaire, rather than question wording and comprehension, although the last two were also considered.


**Table 2.2** Number and date of letters sent

Source: Migration-Mobility Survey 2016

#### **2.5 Participation and Non-response Rate**

At the end of the fieldwork, 5973 interviews were completed, surpassing the goal of 4120 interviews (see Table 2.3). In total, 4702 people of the main sample participated. Moreover, the use of the reserve sample – of which 1271 people participated – helped not only to increase the number of interviews but also to achieve the target set for each of the strata, with the sole exception of German men with a duration of residence of less than 2 years. In general, the participation rates of recently immigrated men were among the most deficient strata.

In total, more than one-third of the sampled individuals with valid addresses participated, with total response rates ranging from 26% for the Portuguese subsample to 45% for the Indian sub-sample (see Fig. 2.1).

This participation rate can be considered satisfactory given the characteristics of the population under study (mobility and low rate of landline numbers for the reminders). Moreover, the age and gender structure of the sample and the respondents being very similar, the results guarantee a certain degree of representativeness within each group.

The mean duration of the online interviews was 35 min and 20 s. The correlation between the assigned language (based on the nationality) and the interview language was high, thus validating our strategy; only 1.1% of individuals who were assigned German chose another language, 1.2% for French, 4.6% for English, 5.1% for Italian, 14.8% for Spanish and 16.3% for Portuguese.

To identify a possible non-response bias, logistic regression analyses were applied to identify the most influential variables for non-response (see Table 2.4) Variables tested included nationality, age, gender, residence permit and civil status. All variables had a significant effect on the response rate, although to varying degrees. In fact, women showed a significantly higher response rate than did men. Short-term permit holders and international diplomats presented a higher risk of non-response than did other permit holders. This can be explained by their higher mobility and shorter duration of stay. Additionally, specific nationalities, such as Spain, Portugal and South America, had a significantly higher risk of non-response. Compared with the other nationalities, these groups were also more frequently employed in low-skilled jobs and held a tertiary degree less often. These results are confirmed by several studies that conclude that an underrepresentation of national


**Table 2.3** Status of sampled addresses, for the main sample, reserve sample and the total (N)

Source: Migration-Mobility Survey 2016.

Note: response rate calculated according to AAPOR (American Association for Public Opinion Research, AAPOR response rates, https://www.aapor.org/Education-Resources/For-Researchers/ Poll-Survey-FAQ/Response-Rates-An-Overview.aspx. Accessed 15 June 2018) Source: Migration-Mobility Survey 2016


Note: \*\*\* p < 0.001, \*\* < 0.01, \*p < 0.05, + p < 0.1 Source: Migration-Mobility Survey, 2016. Weighted results (design weights)

minorities results in bias, even after controlling for socioeconomic status (Elcheroth et al. 2011; Lipps and Pollien 2011; Laganà et al. 2013).

We also observe a weak (35–44 years compared to 24–34 years) or no difference (44–54 years and 55–65 years compared to 24–34 years) between the age categories and a slightly lower likelihood of participation among divorced or widowed participants compared with married individuals.

Duration of residence was not included in the regression model because it was highly correlated with the resident permit. Nevertheless, a descriptive analysis of the response rate by duration of residence shows a U-shape relationship. The proportion of respondents is 32% among individuals who stayed up to 1 year in Switzerland, 43% among migrants with five to 6 years of residence, and 36% for the participants that stayed between 9 and 10 years.

Finally, based on the analyses presented in this section, a weight variable was calculated following a three-stage process: considering the probability to be selected in the sample (design weight), correcting for non-response (non-response rates) and

**Table 2.4** Logistic regression results for unit non-response (odds ratios) adjusting for the size of the reference population (calibration). For more-detailed information on the weighting procedure or other aspects of the methodology, please consult the survey report, which is available online.15

#### **2.6 Socio-demographic Characteristics of Participants**

Table 2.5 presents the socio-demographic characteristics of participants by origin. First, although slightly more men participated than did women, the distribution varies considerably from one origin group to the other. For Germany and France, the men's share is as high as or even surpasses 60%. North Americans and West Africans present rather balanced gender ratios. The only origin group in which the men's share is lower than the women's share is migrants from South America, with 28%. This share nevertheless represents the ratio observed in these populations, indicating a highly feminized migration flow. In fact, the largest differences between participation and population structure are found for German men, who participated more often (61% vs. 56%) and West African men, who participated less often (51% vs. 56%).

Considering that only migrants aged between 24 and 64 years at the time of the survey were included, the mean age of all participants is 39 and ranges from 35.5 years for Indians to 42 years for US/Canadian-citizens. No differences in the age structure of British respondents and the British population are observable, with a mean age of respectively 43 years.

Concerning civil status, 28% of all participants are single, of whom 18% are in a relationship, whereas more than half are married (63%). The highest shares of married participants are observed among extra-European country nationals (between 75% for West Africans and 89% for Indians) and the lowest among citizens from Switzerland's neighbouring countries (between 41% in Australia and 49% in Italy). Nevertheless, the latter often live in a partnership without being married; in fact, they present above-average shares for singles who are in a partnership (between 26% for Italy and 33% for France). Once weighted (considering that civil status was considered when computing the weight variable), the share for the married participants is ten percentage points lower in the total population (53%), but no large differences are observed when separately considering the origins.

In total, 68% of participants hold a tertiary degree. High above-average shares are found for the English-speaking origin groups, that is, India (97%), North America (94%) and Great Britain (91%), followed by France (85%) and Germany 72%). The lowest shares are observed for the Portuguese citizens (24%) and West Africans (44%). Except for the total share – that is, 62% after weighting – we find the same educational distribution by origin once the data are weighted.

<sup>15</sup>FORSbase, Migration-Mobility Survey, https://forsbase.unil.ch/project/study-public-overview/14592/0/. Accessed 14 June 2018.


**Table 2.5** Socio-demographic characteristics of participants, by origin (in % if not otherwise indicated)


Note: \*Rounded values to 0.5 Source: Migration-Mobility Survey, 2016. Unweighted results

Although a bit less than one-third of all participants holds a settlement permit (C), two-thirds hold a residence permit (B). Germans present the highest share of settlement permit holders (53%), which is also reflected by their higher mean duration of residence – in fact, EU/EFTA citizens obtain a settlement permit "almost" automatically after a residence duration of 5 years. However, once weighted, this share is much lower at 46%, whereas the one for the residence permit holders increases from 45% for the participants to 51% for the weighted results. Although the Spanish citizens present the highest shares of residence permit holders compared with all EU/EFTA nationals (71%, a share that does not change after weighting), 77% of all West African participants hold a residence permit (slightly lower after weighting at 73%). A last result concerns the short-term permit holders (L). In fact, one of five Indians and one of ten Portuguese hold such a permit, indicating their involvement in more short-term employment engagements. Nevertheless and when considering the educational distribution, the Indians are instead working in highly qualified jobs, whereas the Portuguese are in less-qualified jobs.

Finally, concerning the duration of stay, Indians (31%), North Americans and Italians (each 26%) present the highest shares of participants that have thus far stayed for less than 2 years in Switzerland, whereas the Germans' (13%) and Portuguese's (11%) shares are among the lowest. Although once weighted, the shares are slightly higher (due to the higher non-response of individuals with a short(er) duration of stay, the ranking does not change – e.g., Great Britain 34% and Portugal 12%).

#### **2.7 Conclusion**

The Migration-Mobility Survey joins other surveys initiated in other European countries, providing innovative data on recently arrived migrants to Switzerland with respect to their migration trajectory and integration process in Switzerland. As done in other surveys, standard survey procedures were adapted to the migrant population, the specific conditions of Switzerland and the research objectives of the nccr – on the move.

In the future, the Migration-Mobility Survey will be extended. A second wave will occur in autumn 2018. Its aim is threefold. Participants of wave 1 who agreed to be interviewed again (62.5% of all participants) and remained living in Switzerland will be asked questions on their current labour market situation and social integration, to measure the pace of integration. Also, participants not living in Switzerland anymore will be asked about their emigration motives. Finally, a new sample of migrants that is representative of the whole immigrant population will be drawn, whereas the 11 origin groups are kept for comparative reasons with wave 1. A questionnaire similar to that used in wave 1 will be applied. Indeed, attrition is expected to be relatively high in the first wave because of re-emigration. To obtain – in the future – longitudinal results consistent with our research objectives, we must ensure the ability to include in 2020 and thereafter a sufficient number of immigrants in the longitudinal Migration-Mobility Survey.

**Appendix**




**Table 2.6** (continued)


Date of

Target

Longitudinal

Sampling


**Table 2.6**(continued)


Date of

Target

Longitudinal

Sampling


**Table 2.6**(continued)


Name

Abbreviation

survey

Country

population

Date of

Target

Longitudinal

or Transversal

Frequency

Sample size

Thematic focus Module

Sampling

strategy

Survey mode Incentive

Languages

Note: \* Although this survey occurred before the year 2000, it nonetheless seemed important to include it in the table Source: Online research by the authors. See Sect. 2.3 for further information on the research methodology


**Table 2.7** Questionnaire structure, sections and topics

Source: Migration-Mobility Survey 2016. Questionnaire

#### **References**


*of limitations. Final Report* (Research Report). http://irihs.ihs.ac.at/2728/. Accessed 11 Mar 2017.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Part II Migratory Process and Arrival in Switzerland**

## **Chapter 3 Who Receives More Help? The Role of Employer Support in Migration Processes**

**Laure Sandoz and Fabian Santi**

#### **3.1 Introduction**

Why are white people expats when the rest of us are immigrants?

This question, raised by the African social activist Mawuna Remarque Koutonin in the Guardian in March 2015, critically addresses the problematic dimension of the categories that we commonly use when we talk about mobile people. Although "immigrants" are typically represented as poor, ethnically marked low-skilled people, "expats" tend to be imagined as white, wealthy, highly skilled individuals who easily travel from one country to another according to their job (Favell et al. 2006; Al Ariss and Crowley-Henry 2013; Cranston 2017). These stereotypical representations are problematic in many respects, but they reflect nonetheless the global inequalities and power relations that currently structure migration processes.

This article highlights some of these inequalities by focussing on the support that different groups of migrants receive when they move to Switzerland. Drawing on both an ethnographic study and the nccr – on the move Migration-Mobility Survey, we want to understand how different actors structure different forms of migration for different categories of people. We focus in particular on the relocation support that labour migrants receive from their employer when they come to Switzerland. We contribute to the theoretical framework of the Migration-Mobility Nexus by examining how the mobility practices of companies intersect with the Swiss

L. Sandoz (\*)

F. Santi University of Neuchâtel, Neuchâtel, Switzerland e-mail: Fabian.Santi@unine.ch

© The Author(s) 2019 57

I. Steiner, P. Wanner (eds.), *Migrants and Expats: The Swiss Migration and Mobility Nexus*, IMISCOE Research Series, https://doi.org/10.1007/978-3-030-05671-1\_3

nccr – on the move, Neuchâtel, Switzerland e-mail: Laure.Sandoz@unine.ch

migration system to create differentiated pathways of inclusion and exclusion according to social characteristics such as gender and nationality.

Relocation support not only facilitates adjustment to a new environment (Ravasi et al. 2015) but also motivates and enables a move (Van den Broek et al. 2015; Harvey et al. 2017; Groutsis et al. 2015). Support is particularly important for the third-country nationals who want to migrate to Switzerland because the restrictive admission system limits access to residence permits for this category of people. Labour migrants from third-countries can only be admitted if they are supported by an employer. In addition, candidates for family reunification need support from a family member with a right to stay in Switzerland (Amarelle and Nguyen 2010). Analysing the role of support agents for recent migrants in Switzerland thus enables us to better understand how opportunities and obstacles to mobility and social inclusion are constructed. Observing which categories of people receive more support from employers informs us about who has more power to negotiate advantageous relocation conditions and, in this sense, represents a more "wanted" migrant for profit-oriented actors.

We start this article by discussing the construction of migrant categories in the academic literature. We then present our methodology, which draws on both ethnographic and survey data. We introduce in Sect. 3.4 our main hypotheses by combining findings from the literature, interview quotes from the ethnographic study and the descriptive analysis of the survey data. We finally present in Sect. 3.5 a model based on a logistic regression to test these hypotheses, and we discuss their relevance with respect to our main research questions. We conclude this article by arguing that definitions of "wanted" migrants must consider the practices of various actors, including the employers who select and attract people according to their immediate needs.

#### *3.1.1 Defining the "Wanted" Migrants*

Most of the categories that researchers use to define people on the move arise from the historical construction of nation-states (Favell 2008). Several authors in recent years have argued that the representation of citizens as sedentary subjects that differ from migrants based on the rights and privileges granted to them by the state is not sufficiently reflected upon in the academic literature (Dahinden 2016; Fassin 2011; Favell 2008). These authors acknowledge the central influence of political categorizations in migration processes, but they encourage researchers to maintain a critical stance to allow research to remain an independent field of knowledge production rather than a tool in the service of states (Hercog and Sandoz 2018a). They remind us that although political constructions such as nationality shape the possibilities of individuals to move and stay in different places (Wagner and Reau 2015; Kaufmann et al. 2004), these categories are not neutral because they rely on political, economic and social dynamics that normatively construct migrants according to culturally situated values. In a global world shaped by selective borders, the recent popularity of the opposition between "highly skilled" and "low-skilled" migrants relies for instance on the emergence of skill-focussed immigration programmes in the 1960s whose objective was to attract "the best and the brightest" individuals and prevent the immigration of "unwanted" people (Parsons et al. 2014; De Haas et al. 2016). These new approaches to migration management have given rise to what Shachar (2006) has called the "Race for Talent", and they have contributed to new categorizations of migrants as either "highly skilled" or "low skilled".

Although approaches that critically analyse migrant categories remain rare in the social sciences (Hercog and Sandoz 2018a; Favell 2008), several authors have contributed to this field by reflecting on the problematic dimensions of some commonly used categorization processes. They have shown that, although policies that select migrants based on their skill levels are often perceived as less discriminatory than other selection tools because of their merit-based approach, they nonetheless lead states to discriminate between migrants according to criteria such as their race (Tannock 2011), cultural closeness (Yeung 2016) and social desirability (Hercog and Sandoz 2018b; Simon-Kumar 2015). Moreover, oppositions between "highly skilled" and "low skilled" migrants raise important issues concerning the value given to different forms of knowledge in a world in which power is unequally distributed across places and social groups (Wagner 2007). The gendered bias of skillselective policies and the disadvantages they entail for women have also been highlighted by several researchers (Kofman and Raghuram 2005; Boucher 2007; Kofman 2014). Finally, some authors have argued that the value and rights attributed to different groups of migrants rely more on decision makers' perception of a given situation than on the migrants' actual characteristics (Hercog and Sandoz 2018a, b; Sandoz 2018a). These different studies show that keeping a critical stance towards categories is important to avoid reproducing or legitimizing discriminatory systems of representation.

Critical approaches towards migrant categories usually focus on the role of states in managing, naming and counting moves (Favell 2008). However, researchers have highlighted the growing influence of non-state actors in the process of selecting, supporting and employing migrants (Groutsis et al. 2015; Cranston et al. 2017). Many countries, including Switzerland, currently use demand-driven systems and partly delegate the task of selecting migrants to employers (Chaloff and Lemaitre 2009; Parsons et al. 2014). In these systems, the employers are responsible for applying for the admission of the candidates they want to hire, and they thus contribute to defining the "wanted" migrants who can obtain access to the national territory and labour market (Gelatt 2017).

Examining the type of support that different groups of migrants receive from their employer is an interesting approach to assessing who obtains more-privileged access to migration in the context of a demand-driven system such as that of Switzerland. It indicates whom the employers are most willing to attract in spite of the administrative hurdles of the admission process. Moreover, such an examination hints at who obtains smoother access to Swiss territory because the relocation support provided by employers also aims to facilitate transitions between places and to ease adjustment processes (Tissot 2018; Ravasi et al. 2015).

This article contributes to the literature on the construction of migrant categories by analysing the specific role of employers in supporting certain categories of migrants and thus constructing them as "expats" who differ from other migrants due to their professional situation and the other advantages that derive from such a status. In light of Mawuna Remarque Koutonin's quote presented at the beginning of this article, we argue that how various social actors perceive the social characteristics of migrants is central to defining them as wanted or unwanted, welcome or unwelcome.

#### *3.1.2 Methodology*

We use a mixed-method approach, combining the quantitative analysis of the Migration-Mobility Survey with an ethnographic study conducted by Laure Sandoz as part of the nccr – on the move project "The Mobility of the Highly Skilled towards Switzerland" (Sandoz 2018b). This combination enables us not only to highlight the main variables that structure access to employers' support for our population but also to understand some of the social and cultural logics that lead employers to give preference to some categories of migrants.

The ethnographic study was conducted between 2014 and 2018 in the two Swiss regions of Basel and the Geneva Lake Area. It focussed on the strategies of institutional actors from the public and private sector to select, attract and retain mobile individuals who represent value to them. It included 16 semi-directed interviews with professionals who work for institutions that have a stake in attracting, selecting or retaining migrants in Switzerland, 13 biographical interviews with people who moved to Switzerland under the auspices of such institutions, and numerous observations at events that aim to inform and support highly skilled migrants in Switzerland. This article primarily draws on eight semi-directed interviews of 1–2 h conducted as part of this research with human resources staff, professional recruiters and relocation agents about their daily practice of recruiting and facilitating the relocation of foreign employees in the two regions under study. We selected them for this article because of their relevance to the discussed topic. Biographic interviews with migrants who relocated to Switzerland with the support of their employer and ethnographic observations at events targeted at highly skilled migrants and their families further inform this article by providing additional data on how employers' support is perceived and experienced by the people who benefit from it.

The survey analysis focussed on the relocation support that respondents to the survey reported. The population under study consists of people from German-, French-, Italian-, English-, Spanish-, and Portuguese-speaking countries who arrived in Switzerland during the past 10 years (for more information about the survey design, see Chap. 2 of this book). The survey is weighted to represent the migrant population in Switzerland within the scope of the survey. Throughout this chapter, the computation of numbers considers these weights. The numbers or ratios reported are thus the expected values based on the whole population within the scope of the survey.

The survey participants were initially asked the following question:

When moving to Switzerland, did you receive any support in one of the following areas? (Multiple answers are possible)

Respondents had the choice between "yes", "no" and "not applicable" for each support category. However, we noticed that the respondents understood the difference between "no" and "not applicable" in different ways. For instance, respondents without children occasionally replied "no" and occasionally replied "not applicable" to the question of whether they received support for school or childcare. We thus decided to recode the "not applicable" answers as "no" to ensure homogeneous treatment of the data.

This first question was asked of all of the survey participants (n = 5973), but only the respondents who replied "yes" to at least one support category (60.1%) were asked the second, following question:

From whom did you receive support? (Multiple answers are possible)

We thus recoded the second variable about sources of support to include the whole survey population in both questions (we considered the 39.9% of filtered respondents from the first question as having replied "no" in the second question). Our analysis specifically focussed on the people who reported support from their employer.

When presenting numbers for specific subgroups of the population, based on factors such as gender, qualification level and nationality, we highlight differences between these groups and determine which properties define those differences best. To confirm the relevance of the differences observed between groups, we apply a Chi-square test of independence after re-weighting the data using normalized weights. P-values are reported at the 5%, 1% or 0.1% levels. Because the results of the statistical tests using this method are approximate, a conservative approach should be taken in evaluating the power of the statistical tests.

We then use a weighted logistic regression in Sect. 3.5 to evaluate the effect of different variables while controlling for the other factors. This approach allows us to verify our hypotheses, which are developed in the presentation of the ethnographic study and the data. The regression presents and summarizes our findings in a single coherent model.

The dependent binary variable in the logistic regression is whether the respondents received support from their employer when they moved to Switzerland. The independent variables are qualification level1 (highly qualified (reference category), less qualified), gender (men (ref), women), relationship status (married or in a rela-

<sup>1</sup>We define highly qualified people as people with either advanced technical and professional training or an academic education. Less qualified people are people with a high-school education, vocational training or less.

tionship when coming to Switzerland (ref), not married or in a relationship when coming to Switzerland), nationality2 and professional sector.3 Moreover, we use as control variables possession of a job or a job offer in Switzerland before the migration (or not), age,4 presence of children (or not) and occupational status.5

To choose our categories of reference, we imagined a person who corresponds to the usual stereotype of an "expat" receiving particularly high degrees of support. We refer directly to Mawuna Remarque Koutonin's critique about the construction of differences between "immigrants" and "expats" based on race, class, nationality and gender, and we want to check to what extent employers reproduce these stereotypes when attributing relocation support. In light of the literature on discrimination and social inequalities, we expect the most "wanted" migrants to be highly qualified married men from Anglo-Saxon countries working in socially valued sectors and having managerial responsibilities. We are particularly interested in observing how these different parameters influence access to employers' support in a model that controls for the effect of all of the other variables involved.

Hence, the statistical analysis enables us to discuss the role of employers' support on the relocation to Switzerland of recent migrants from the main immigration countries, whereas the ethnographic study enables us to discuss these results in the light of direct experiences from the field.

#### **3.2 Access to Support: Data, Variables and Hypothesis**

Access to support from an employer, family member, friend or organization can constitute an important motivation to migrate (Piguet 2013; Groutsis et al. 2015). In Switzerland, a large share of the economy relies on the expertise of foreign professionals (Haug and Müller-Jentsch 2008; Mach et al. 2011), and companies play a major role in attracting migrants (Davoine et al. 2015). To this end, they have

<sup>2</sup>The origin includes, based on the nationality Germany, France, Italy, Austria, Spain, Portugal, United Kingdom, United States, Canada, India, Brazil, West Africa (Benin, Ivory Coast, Guinea, Mali, Niger, Burkina Faso, Senegal, Togo, Gambia, Ghana, Liberia, Nigeria, Sierra Leone, Saint Helena, Guinea-Bissau, Cape Verde, Sao Tome and Principe) and other South American countries (Argentina, Bolivia, Chile, Ecuador, Colombia, Paraguay, Peru, Uruguay, Venezuela and Guyana).

<sup>3</sup>The professional sector includes the following categories: agriculture, forestry and fishing; construction; wholesale and retail trade, transportation, hotels and restaurants; public administration, defence, education, human health and social action; information and communication; professional, scientific, technical, administration and support service activities; financial and insurance activities; manufacturing, mining and quarrying and other industry; and other activities and services.

<sup>4</sup>The age is taken as 2018 minus the year of birth, as given in the survey, and as a whole number.

<sup>5</sup>Occupational status includes the following categories: self-employed workers; company owners; relatives employed in a family business; directors or board members and/or with managerial responsibility; people employed without managerial responsibility; people employed in a protected workshop (except support staff); apprentices; and PhD students.

developed various strategies to access the employees they need the most (Ravasi et al. 2015; Van den Broek et al. 2015).

Support in the form of relocation services provided directly by the employing companies or through relocation agencies mandated by the employers constitutes one of these strategies (Davoine and Salamin 2012; Sandoz 2018b; Tissot 2018). Before the migration, the promise of a generous relocation package and positive information about the new environment can help convince a promising candidate to accept a new position. During the relocation, services such as the payment of moving costs and support for administrative issues can help alleviate the constraints associated with migration. After the relocation, services such as spouse employment support and payment for language courses can facilitate adaptation to the new environment. These services enable employers to attract and keep the employees they want most by constructing migration as a smooth and easy process (Tissot 2018). Moreover, they contribute to positioning the people who benefit from such services in a privileged social situation in comparison to other migrants. This quote from an interview with a human resources manager about relocation practices in the banking sector nicely illustrates this situation:

If you want to be professional, you have to mandate a [relocation] agency. But agencies are expensive! And then you give them [the relocating employees] the feeling to be expats. And in the head of expats it's: I want an expat contract and the school for the children, the private school. […] So, if he has a rare skill, he wants 100,000 for the school, he wants the accommodation … it's a bit deterring for the employer (Human resources staff, 11 April 2015).

Employers usually use economic arguments to explain how they attribute relocation support to their employees; although relocation packages are expensive, they become worthwhile if they enable a company to access employees who can bring more value than what attracting them cost (Davoine and Salamin 2012). Under this logic, the relocation support received by employees can be considered a measure of their skills' value for the company.

However, the negotiation of relocation packages also involves other mechanisms (Sandoz 2018b). For instance, the employees must be aware of their value to negotiate advantageous relocation conditions at the moment of their employment. Relocation packages thus also give indications about the position and negotiation power of relocating employees with respect to their employer. Moreover, the value of an employee's skills is very context-dependent because it relies on the company situation, the skills' rarity and the difficulty for the employer to obtain them. The following quote illustrates this complex situation:

If he's someone with a commercial approach, you know how much he costs, you know how much profit he can make, you can calculate, you can objectivize. But it is not always easy to estimate. But if he's a super professional with very specialized skills, he is the only one in Europe who can do the thing, he lays down his conditions. He lays down the market conditions (Human resources staff, 11 April 2015).

Of course, not all migrants have access to the support of an employer. In our survey, although a majority of the respondents declared that they had moved to Switzerland

**Fig. 3.1** Main sources and areas of support received by migrants, Switzerland Source: Migration-Mobility Survey 2016. Weighted results

for professional reasons (61.6%), and more than half of them had obtained a job or a job offer in Switzerland before coming (52.2%), only one-third reported support from an employer. Moreover, nearly 40% of them reported no support at all. Although the employers nonetheless clearly constitute a major source of support for recent migrants in Switzerland, Fig. 3.1 shows that the respondents to our survey could also rely on the help of other actors for organizing their move to Switzerland, in particular their personal contacts such as friends, relatives and business colleagues.

To better understand the effect of the employers on relocation support, we compared the situation of people who had received some type of support from their employer with that of people who had not. Table 3.1 shows that people who could rely on their employer had, in general, access to more services than did those who relied on other sources of support, in particular with respect to financial support for organizing their move, support for finding accommodation and addressing administrative issues, allowances for language courses and access to schooling or childcare. This descriptive analysis shows that support from employers provides access to specific resources and services that are less available to migrants who rely on other forms of support.

Although these findings highlight the important role played by employers in providing relocation support, they also show that such support is not available to every migrant. It is therefore important to understand whom this support prioritizes and what other resources are available to people on the move.


**Table 3.1** Effect of employer support: comparison between migrants who declared support from their employer and migrants who did not declare any support from their employer but who received support from (an)other source(s)

A comparison between these findings and those of a study by Ravasi et al. (2015) – who focussed exclusively on the relocation services that 12 multinational companies in the French-speaking part of Switzerland offer to their highly qualified employees – suggests differences in terms of qualification level, professional sector and types of companies. Support availability is much higher in their study than in ours (financial support for moving costs was for instance available to 94.1% of the respondents in their study, against 17.6% in ours). These differences point to a significantly higher degree of relocation support in multinational companies, which encourages examining the specificities of professional sectors. Moreover, both our ethnographic observations and the literature in this field point to the effect of nationality and gender on selection processes within companies.

We thus propose to analyse differences in access to relocation support for various categories of migrants, in particular according to their qualification level, gender, nationality and professional sector. We want to understand who obtains access to more relocation support in the context of the Swiss immigration system and thus constitutes a more "wanted" category of migrant in the eyes of the employers. In the following paragraphs, we discuss the potential effect of different variables in the light of the literature and based on our ethnographic observations and descriptive statistical analyses. This discussion introduces the model that we use in Sect. 3.5 to test our hypothesis.

#### *3.2.1 Qualifications*

The descriptive analysis of our survey data suggests that access to relocation support is closely connected to the level of education. As shown in Fig. 3.2, highly qualified migrants are more often supported by their employer than are less qualified people, whereas less qualified people receive more support from relatives in Switzerland. These data indicate a stronger dependency of less qualified migrants on kinship networks, whereas highly qualified people more often rely on the support of their employer. Support from friends and colleagues, however, is more evenly distributed between the different categories and indicates no clear correlation with the qualification level.

Although this observation is interesting, we hypothesize that it only reveals part of the story. As many other authors have shown, qualifications are only one of the signals that employers use to evaluate candidates for employment (Findlay et al. 2013; Zschirnt and Ruedin 2016). Perceptions and stereotypes associated with the nationality, ethnicity, gender, family situation and age, among other variables, contribute to defining the value of candidates during recruitment processes. We thus hypothesize that they also influence access to relocation support.

In the logistic regression presented in Sect. 3.5, we distinguish between highly qualified and less qualified migrants, and we expect this variable to have a significant effect on the support offered by employers. However, we do not expect this variable to have more effect than other variables less directly linked to skills, such as nationality. We assume that, in contrast to economistic discourses that conceive

**Fig. 3.2** Support received by migrants from relatives and employers by level of qualification, Switzerland

skills as purely marketable products obeying clear economic logics (Urciuoli 2008), representations of "wanted" migrants involve more than a simple evaluation of a person's competences. Beyond what a person can do, what a person is or appears and the social, political and economic contexts in which this person moves influence the types of resource and privilege that s/he can access (Hercog and Sandoz 2018a).

#### *3.2.2 Gender and Relationship Status*

Gender is without any doubt indispensable for analysing the issue of access to employers' support. However, it is important to distinguish between differences that rely on structural factors and differences that rely on direct discrimination by the employers.

Although many authors have noticed an increase in the number of female-led relocations and dual-career couples over the past decades (Salamin and Davoine 2015; Crompton and Lyonette 2006; Harvey et al. 2009; Brookfield 2016), our survey data clearly show that the dominant model in Switzerland remains that of the woman following her partner. Although both men and women respondents reported professional reasons to be their main motivation for migrating to Switzerland, we observe that men more often migrated for professional reasons (70.4% vs 50.4%), whereas women more often migrated to accompany family (29.3% vs 7.5%). Moreover, 62.9% of the men had already found a job in Switzerland before migrating, against only 38.5% of the women.

Of course, this situation has implications for the type of support received by men and women. In fact, the data show that men received more support than women did in all of the surveyed categories except for "spouse/partner employment support", "school/childcare" and "other forms of support". Not surprisingly given their main migration reasons, women more often reported having been supported by relatives in Switzerland than did men (13.7% vs 10.2%), whereas men more often reported having been supported by their employer than did women (37% vs 26.4%).

These data clearly indicate the existence of structural gendered norms within couples that lead women to give priority to the career of their male partner over their own. They thus point to the importance of the intersection between gender and relationship status for structuring access to relocation support. However, they say nothing about processes of direct discrimination by employers. In fact, our descriptive analysis rather points to the absence of discrimination towards single women because the comparison between men and women who migrated as single people indicates no significant difference in terms of access to employers' support (Table 3.2).

Nevertheless, the important literature on gendered recruitment and skill valuation processes encourages us to dig further into the case of potential discrimination towards women by employers. A recent study in Sweden showed for instance that employers are reluctant to hire women who live a long distance from the workplace, whereas men in the same situation encounter no such obstacles (Brandén et al. 2018). Moreover, many studies have highlighted the tendency of employers and


**Table 3.2** Support received by migrants from the employer according to gender and relationship status when moving to Switzerland

other actors to evaluate skills differently depending upon whether they are associated with a man or a woman (Steinberg 1990; Phillips and Taylor 1980; Boucher 2007; Kofman 2014; Jenson 1991; Daune-Richard 2003). If the value of relocation packages derives from the value that employers attribute to their employees, we can expect in the light of these studies to observe a disadvantage for women, even in cases in which we control for the effect of other important variables.

At the same time, our interviews with relocated couples and employers point to other dynamics. They suggest on the one hand that the gender norm according to which men must be professionally active remains strong even in couples in which the woman initiated the relocation. This situation leads some women to firmly negotiate their relocation conditions at the moment of their employment to ensure that their partner gets sufficient support to remain economically active despite his position as a trailing spouse (Sandoz 2018b). Although these ethnographic observations do not enable generalizations at the level of a population, they can contribute to explaining situations in which women obtain access to more relocation support than their male counterparts do.

On the other hand, employers are aware that many relocations fail because the spouse is not satisfied with his or her new situation (Salamin and Hanappi 2014). We see this phenomenon for instance in this quote from the human resources manager of a bank:

The man who works for us, if his wife is not happy, the children are a bit lost at school and everything, it's not going to work. So you know that if you want to keep them, you have to organize a complete relocation of the family with all the proper rules. It costs you an arm and a leg, you're not sure that it will succeed (Human resources staff, 11 April 2015).

The interviewee explicitly refers to a situation in which the man is the lead and the woman follows. He mentions the difficulty of keeping the new employee if the partner remains unemployed. He also explains that the costs of moving both partners are particularly high. Between the lines, he suggests another idea that came up in many other interviews, which is that when employees leave shortly after the relocation, the company loses the money they invested for attracting them. In situations in which the main candidate for relocation is a woman and the trailing partner is a man, we might expect employers to be even more reluctant to take the risk of investing in an expensive relocation because they might consider that the risks for a male unemployed partner to be unhappy and want to leave are even higher because of the effect of gender norms (Harvey and Wiese 1998; Mancini-Vonlanthen 2016). They might thus decide either to invest more resources to support the professional integration of the male trailing spouse or to reduce the costs of the relocation to limit the losses for the company in the event of an unexpected departure of the female employee.

We are thus particularly interested in observing the effect of gender on employers' relocation support in the logistic regression in Sect. 3.5 that controls for other important variables. We hypothesize a difference that is more strongly connected to the interaction between the relationship status and gender than with gender alone because a large part of the observed differences between men and women appears to rely on the interaction between couple dynamics and structural gender norms. However, in light of the literature, we would not be surprised to observe differences that suggest more-direct forms of discrimination towards women from the employers.

#### **3.2.2.1 Nationality**

Opportunities to migrate, to be recruited by an employer and to receive relocation support greatly vary amongst nationalities. One main reason for this variation is the existence of national border regimes that discriminate between countries and regions (Hercog and Sandoz 2018b). In Switzerland, only "qualified workers from third countries who are absolutely needed" and who have a job contract in Switzerland are in principle allowed to enter the Swiss labour market (Swiss Federal Council 2002). In contrast, citizens from the European Union (EU) and from countries belonging to the European Free Trade Association (EFTA) can freely stay and work in Switzerland, as long as their personal resources allow doing so (Sandoz 2016). This dual admission system contributes to structuring migration flows to Switzerland. On the one hand, people from third-countries need special assistance from their employer if they want to migrate to Switzerland as labour migrants; on the other hand, this system encourages the employers to prioritize recruitment from within the EU/EFTA and only to attract specialists that they need the most from thirdcountries. Given this specific legal regime, we can expect that labour migrants from third-countries who come to Switzerland with a job will receive more support from their employer than will EU/EFTA migrants because they constitute a category of particularly "wanted" migrants from the perspective of the employers.

However, beyond immigration policies, we also must consider the effect of social and economic processes on the position of different nationalities. The transferability of skills is never neutral because it reflects power relations at a global level that enable the characteristics associated with certain regions and countries to be perceived as internationally more valuable than others are (Wagner and Reau 2015; Sommer 2016). For instance, the privileged position of the English language in international environments reflects the leading economic and cultural influence of the United States in the world. As the sociologist Anne-Catherine Wagner (1998) says, "the relations between nationalities present similarities with the relations

between social classes within a country. Those who can assert the 'international' value of their national attributes oppose those who rather tend to repress them in order to adapt to dominant norms" (own translation by the authors).

Consistent with this observation, the descriptive analysis of our survey data presented in Fig. 3.3 clearly shows that both nationals from Anglo-Saxon countries and people from Switzerland's neighbouring countries receive on average more support from their employer than people from less economically powerful countries do. It does not mean that people from this second group receive no support at all, but they tend to receive it more from their relatives in Switzerland than from an employer. At the same time, we observe that the recent emergence of certain developing countries as major global economic players and workforce providers in specific fields – for instance, India in the field of science and information technologies (Swiss Federal Department of Foreign Affairs 2017; Xiang 2007; Hercog 2014) – has produced new groups of well-supported professionals who challenge the classical representations of the "expats" as white individuals coming from western countries (Kunz 2016; Pavic 2015). In this sense, the strict Swiss admission system for third-country nationals interacts with economic dynamics and global power relations to create different possibilities for different nationalities to migrate and to be supported. We see this variation clearly in this quote from the interview with a professional recruiter working for the life sciences business:

After a while, I stopped looking outside of Europe. […] We always looked at nationality. […] If there was a country outside of Europe where we searched more, it was the United States. But it's true that it was complicated. Then, if it was the right person, it was possible. But it really had to be the perfect profile (Professional recruiter, 5 February 2015).

The logistic regression presented in Sect. 3.5 enables us to go one step further and test the influence of nationality on access to employers' support while controlling for the effect of other important variables. Consistent with the previous discussion, we expect people from the United States and other Anglo-Saxon countries to be particularly well supported by their employer in comparison with people from less economically powerful regions such as West Africa and Latin America. However, we would also expect people from neighbouring countries to receive less relocation support than would people from more-distant, non-EU countries because Switzerland is more accessible for them due to both geographical proximity and the free movement of persons system.

#### *3.2.3 Professional Sector*

A last important aspect that must be considered when analysing access to relocation support concerns the recruitment practices of employers in specific professional sectors. As the migration researcher Robyn Iredale (2001) notes, "The type and level of regulatory mechanisms, the level of internationalization and the relative influence of the market, the state and the profession, and the global labour market demand/supply situation are all very significant factors in explaining migration".

These factors also contribute to explaining differences in access to relocation support. In particular, how economic actors define needs and shortages in certain sectors influences the amount of resources that they are willing to invest to attract workers (Findlay et al. 2013; Ruhs and Anderson 2010). Hence, relocation support is granted not so much according to an individual's level of skill as according to the skills' perceived rarity and necessity in a specific economic context and according to the specific position and negotiation power of the employee. As one of our interviewees working in the human resources department of a chemical company said:

I think that the definition of highly skilled depends basically on the skills that are on the local labour market […]. So you might consider that the skills of a CEO are relatively low if you can easily find somebody like a CEO on your local labour market (Human resources staff, 18 February 2015).

In addition, access to support relies on the companies' internal relocation policies, which often depend upon their internationalization degree and dependence upon a foreign workforce (Iredale 2001). We have seen for instance that multinational companies tend to more systematically offer relocation services than other companies do (Ravasi et al. 2015). However, the current tendency among multinational companies consists of reducing the costs associated with such services (Le Temps 2015; Davoine and Salamin 2012; Cartus 2014). We can thus expect to observe differences between professional sectors that have more to do with internal management decisions and priorities than with clearly identifiable economic factors.

To illustrate these aspects, Fig. 3.4 suggests that some differences in support rely on the specificities of the professional sectors, although we also observe a relationship between the level of qualification and the support received from either the employer or relatives in the different sectors. For instance, the respondents working in "information and communication" reported less support from their employer than did the respondents in "manufacturing, mining and quarrying and other industry", although 84.1% of the people in the first sector have a higher education against 74.3% of the people in the second sector. Moreover, a further analysis highlights specific recruitment channels within some activity sectors. For instance, nearly onethird of all of the Indians work in the "information and communication" sector and report very high degrees of support from their employer. We also noticed that a surprisingly high share of low-qualified migrants from Portugal had reported support from an employer in the construction and agriculture sectors, indicating that not only highly qualified migrants are being actively recruited towards Switzerland.

Nevertheless, the logistic regression in Sect. 3.5 that analyses differences in support between professional sectors also must consider other variables because relocation support partly depends upon a person's hierarchical position, level of responsibility and type of occupation within a company (Davoine and Salamin 2012). We can expect some categories of workers, for instance self-employed workers and company owners, to receive significantly less employer support compared with other categories of workers. We thus control inter alia for the effect of occupational status in our model, to observe whether differences in support between professional sectors persist and, if true, which sectors offer more support to their employees.

#### **3.3 Determinants of the Employer Support**

The results of the logistic regression on employer support are presented in Table 3.3.

The effect of the qualification level is, as expected, highly significant. With all of the other variables being equal, the probability of receiving the support of an

**Table 3.3** Effect of different variables on the employer support received by migrants, Switzerland. Results of a logistic regression


Note: +p < .1. \*p < .05. \*\*p < .01. \*\*\*p < .001. Model controlled for possession of a job or a job offer in Switzerland before the migration, age, presence of children and occupational status. The accuracy of the model is 71.3%, for 4347 observations. The Nagelkerke R2 is 0.259 Source: Migration-Mobility Survey 2016. Weighted results

employer is approximately halved for less qualified people compared with highly qualified people. However, the model also shows that the qualification level is by far not the only relevant variable to explain access to employers' support. In particular, the effect of most nationalities is higher than the effect of the qualification level, indicating that a low qualified American has for instance a higher probability of receiving support than does a highly qualified West African.

The effect of gender is significant and indicates a clear disadvantage for women compared with men concerning access to employers' support. According to the model, women have an approximately 35% lower probability of receiving support than men do. However, we observe that this disadvantage is specific to the women relocating with their partner because the fact of being a single woman cancels out the disadvantage of being a woman. In contrast, men in a relationship experience no significant disadvantage. We thus observe that, even when all of the other variables are equal, women with a partner receive less support from their employer than single women do or than men do, whether single or in a relationship.

Nationality strongly affects access to employers' support. As expected, US nationals occupy a particularly favourable position because they receive more support than do any of the other nationalities except for people from the United Kingdom, Canada and India, for whom the results are not significantly different. The more disadvantaged nationalities with respect to employers' support are the West Africans and the Portuguese, who appear to have a greater than six times lower probability of receiving support from their employers compared with the Americans, even when they have a similar level of qualifications, work in a similar field and have a similar occupational status. It is interesting however that the model does not clearly differentiate between EU/EFTA and non-EU nationals. As expected, people from neighbouring countries receive relatively little support, but this point is also true for people from West Africa and Latin America. In contrast, English-speaking people from both EU/EFTA and third-countries (except for West Africans) have a clear advantage in accessing employers' support.

Differences between professional sectors are less pronounced than for the other variables. However, we observe that people working in the financial sector occupy a particularly favourable position for accessing employers' support. In contrast, people with state-related professions (public administration, defence, education, human health and social action) and with jobs that necessitate specific local skills (information and communication) have approximately a 50% lower probability of being supported by their employer. The sectors that most notably rely on low qualified migrant workers (e.g., hotels, restaurants, and construction) also appear to invest less in relocation support.

#### **3.4 Discussion**

The statistical analysis investigated the role of employers in providing support to immigrants. To the extent that relocation support is a tool that enables employers to attract economically profitable employees, observing which categories of people receive more support gives us information about who has more power to negotiate advantageous relocation conditions and, in this sense, represents a more "wanted" category for profit-oriented actors.

A first important finding is that employers grant less relocation support to nonsingle women compared with men and single women. Traditional representations of gender roles thus interact with corporate systems and mobile ways of life, influencing the type of resource that individuals can access depending upon both their gender and family situation. This finding reinforces the argument already introduced by Brandén et al. (2018) that employer recruitment choices contribute to the trailing spouse phenomenon by favouring the relocation of men over that of women. Our analysis adds to this argument by pointing to the crucial importance of the interaction between gender and relationship status. It suggests the existence of a dynamic that specifically disadvantages non-single women; if employers are reluctant to supporting non-single female employees because they fear that the relocation will fail due to the partner's bad adjustment (Mancini-Vonlanthen 2016; Salamin and Hanappi 2014), then they also increase the risk that the relocation will fail. As other authors have shown, corporate support has a positive effect on accompanying partners' adjustment and well-being (Grant-Vallone and Ensher 2001; Mancini-Vonlanthen 2016). By providing less support to accompanied women, the employers contribute to reproducing a self-fulfilling prophecy because they create a lower probability for their relocation to succeed compared with male-led relocations. This finding calls for more awareness about this phenomenon and for more research on the situation of couples in which the woman's job is the main driver of the relocation (Cangià 2017; Punnett et al. 1992).

Then, the data clearly show that employers do not treat all of the different nationalities similarly. This differentiation is not so surprising at first sight because the Swiss migration system creates different admission requirements for EU/EFTA and third-country nationals. Nevertheless, the differences are only loosely connected to immigration regulations. For instance, although both North America and West Africa are considered third-regions in Switzerland, the probability of receiving support from an employer is approximately eight times lower for West Africans than for Americans, even when they have a similar level of qualification, work in a similar professional sector and have a similar occupational status. Moreover, it is surprising that workers from West African and Latin American third-countries have on average a similar likelihood of receiving support from their employer compared with most EU/EFTA nationals because the former category of people is supposed to be more strictly selected by employers than the latter. The absence of clear differences thus suggests a disadvantage for the third-country nationals from these regions because the support of an employer is much more crucial for them than for Europeans due to the restrictive dual immigration system.

Finally, the analysis points to the importance of considering differences between activity sectors. It shows that people with similar characteristics receive different forms of support depending upon their job. This variation is currently not addressed as a political concern given the demand-driven focus of the Swiss immigration policy and the importance granted to immediate economic needs. Nevertheless, in a country that largely benefits from the expertise and knowledge of migrants, understanding the specific needs of different groups of migrants, developing programmes to address them and taking action to promote access to equal opportunities appear as crucial issues for the future.

#### **3.5 Conclusions**

This chapter focussed on the situation of recent migrants coming to Switzerland from a selected set of countries. We wanted to understand how access to employers' support is structured and who benefits more from it. We also wanted to critically address the construction of migrant categories by showing that employers create specific migration conditions for some people to whom they attribute an economic and social value that is partly independent of their actual skill level.

We used a combination of statistical and ethnographic methods to investigate differences in access to employers' support by recent immigrants. We showed on the one hand that employers actively attract some people to Switzerland, whereas on the other hand, different categories of migrants have unequal access to their support. The analysis revealed that, beyond the qualifications, social categories such as nationality and gender structure access to employers' support. Moreover, observing relationship status and professional sector is important for understanding differences in access to support.

A main finding of this analysis is that economic power relations between countries, gendered norms and recruitment practices within professional sectors all interact with immigration policies to create systems of inclusion and exclusion that enable highly qualified men from economically powerful countries to access significant corporate relocation support. Although this finding corresponds to the stereotype of the "expat" as a highly qualified man from a rich Anglo-Saxon country, it also shows that highly qualified men from rich Anglo-Saxon countries are actively given the possibility to become "expats", whereas people with similar levels of qualification and experience but with a different gender, nationality or background have fewer opportunities to access support and migrate. In this sense, the very notion of "expat" is a construction that reflects power relations at a global level. However, we also observe that categories and power relations can change over time. For instance, highly qualified Indians currently appear to have a probability of obtaining support from their employer similar to the Americans and the British. This new phenomenon can in turn transform our perception of them into "expats"

instead of "immigrants". This finding adds to the Migration-Mobility Nexus theoretical framework by showing that regimes of economically driven mobility and state-driven migration interact with each other dynamically, reproducing in some cases older forms of inclusion/exclusion but also creating new ones.

These findings bring us back to the discussion of the notion of "wanted" migrants at the beginning of this chapter. In the academic literature, a clear distinction is often made between high-skilled and low-skilled migrants. We tried to show, however, that skills are not all that matters. The possibility of using one's skills and being acknowledged for one's skills is linked not only to a person's characteristics but also to the social, economic and political context in which they are embedded. In that sense, how actors involved in supporting and controlling mobility view migrants and differentiate between categories has direct consequences for the personal experiences and biographies of individuals moving across borders because it contributes to creating more- or less-supportive resource environments for them. As Mawuna Remarque Koutonin reminded us, categories are not neutral because they can transform someone's life.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Chapter 4 Who Are the Serial Movers? Sociodemographic Profiles and Reasons to Migrate to Switzerland Among Multiple International Migrants**

**Jonathan Zufferey**

#### **4.1 Introduction**

In recent years, more and more individuals are experiencing mobility, which leads to an increase in international migration (Castles 2013). In parallel with this increase in population movement, migration is becoming more complex and can no longer be analysed as a unique and definitive settlement in a new country (King 2002). Although previous research has extensively deepened our understanding of why individuals leave a place to settle in another country, little is known about individuals who have gone through several international movements.

Using the Migration-Mobility Survey (see Chap. 2), this research aims at deepening the understanding of individuals who have undertaken one or more international migrations before settling in Switzerland. As a country of migration – in 2017, 29% of the population is foreign born – Switzerland is a good laboratory to study new mobilities in the European context. Since the beginning of the twentyfirst century, there have been important changes in the migration forms (Piguet 2009 or see Chap. 1). Between World War II and the beginning of the 1990s, most migratory flows were of workers who responded to the needs of the Swiss economy for a low-skilled workforce. Due to trends in the labour market structure and the entry into force of the free movement of persons between the European Union and Switzerland in 2002, the long-standing demand for a low-qualified workforce persists but is being overtaken by highly educated migration flows (Müller-Jentsch 2008; Wanner et al. 2016). However, little is known about the evolution of these migration flows and the resulting settlement processes. European highly qualified

J. Zufferey (\*)

Institute of Demography and Socioeconomics, University of Geneva, Geneva, Switzerland

© The Author(s) 2019 83

nccr – on the move, Neuchâtel, Switzerland e-mail: Jonathan.Zufferey@unige.ch

I. Steiner, P. Wanner (eds.), *Migrants and Expats: The Swiss Migration and Mobility Nexus*, IMISCOE Research Series, https://doi.org/10.1007/978-3-030-05671-1\_4

migrants are presumably more mobile (Favell 2011; Recchi 2015), but more empirical evidence is needed to understand the forms of migration currently occurring.

This chapter aims at providing a better overview of multiple migration phenomena in the context of the free movement of persons between the European Union and Switzerland. It contributes to the understanding of the Migration-Mobility Nexus in determining the magnitude and the main sociodemographic characteristics of multiple movers. It also presents why migrants have arrived in Switzerland and how the reasons vary according to the number of multiple migrations. Finally, it explores the last country of residence of multiple migrants. Before presenting empirical results, the first section proposes an overview on multiple migrations in the scientific literature.

#### **4.2 Multiple Migrations in Previous Research**

#### *4.2.1 On Multiple Migrations*

Migration was long considered a unique event leading to a definitive settlement in the host country. Before the beginning of the twenty-first century, there was almost no research on multiple migrants because the topic did not arise in this early conceptualization of migration. However, because it appeared that not every migrant settles in the long term in the host country, empirical studies initially developed on return or temporary migrants (Borjas and Bratsberg 1994; Cassarino 2004; Lindstrom 1996). Then, research explored other forms of multiple migrations and tried to conceptualize and define them through the notions of *circular* (Constant and Zimmermann 2011; Hugo 2003; Schneider and Parusel 2015; Vadean and Piracha 2010), *repeated* (Constant and Zimmermann 2011), *secondary* (Takenaka 2007), *twice* (Agrawal 2016), *onward* (King and Newbold 2007; Lindley and Van Hear 2007; Toma and Castagnone 2015; Kelly and Hedman 2016), *stepwise* (Paul 2011), or *serial* (Ossman 2004) migrations. There are of course slight differences between these definitions, but all of them refer to a trajectory of migration that includes stops in multiple countries. The concept of multiple migration has been defined by Oishi (2014) as "extensive geographical movements of individuals across multiple national borders within one generation, which could be circular, linear, or both". The non-linearity of the trajectory and the multiple borders crossed make this phenomenon complex.

Due to a methodological nationalism in the data production (Kalir 2013), quantitative research had difficulties in approaching extensive sequences of multiple migration (Toma and Castagnone 2015) or could only map those occurring within national borders. Therefore, multiple movers have usually been studied in the country of destination using qualitative interviews, biographical questionnaires or census data (King and Newbold 2007; Nekby 2006; Ossman 2004; Takenaka 2007).

#### *4.2.2 Literature Review*

As stated, research on multiple international migrations has grown in recent years. In one of the first studies on this topic, Greenwood and Trabka (1991) were surprised to find that more than 85,000 migrants have lived in a third country before settling in the United States. They explained this large number of multiple migrants with the indirect trajectories of Southeast Asian refugees in the 1980s who stopped in several countries before reaching the USA.

The high level of multiple migrations among refugees has been confirmed in other research (Lindley and Van Hear 2007). War, population displacements, and economic crises in *failed states* led to significant flows of refugees or illegal migrants that, due to the absence of a clear international response, manage their way to a new destination. These turbulent trajectories from southern country migrants, as described by Schapendonk (2012), are diverse, unpredictable, and conducted according to opportunities and constraints.

Nevertheless, in Western societies, the majority of migrants who have arrived through a secondary destination are characterised by a high level of education and highly qualified occupations (Agrawal 2016; Nekby 2006; Takenaka 2007). At the individual level, multiple migrations are a means of developing social and human capital in different countries, languages, and contexts and should allow developing a professional career and attaining highly qualified occupations (Kelly and Hedman 2016). At the macro level, the mobility of the highly skilled is promoted by states that compete to attract the highly qualified professionals needed by their economies (Shachar 2006). In the European Union (EU), the principle of circulation between country members of a highly skilled work force has even been fostered by the EU commission as a political strategy to avoid highly qualified labour shortages and prevent negative public attitudes to mass immigration (Agunias 2006; Hugo 2003). Multiple migrations are however not only driven by highly qualified individuals; stepwise migration trajectories are also used as a strategy by low-qualified migrants from southern countries who accumulate experiences in several neighbouring countries that can offer the opportunity to move to a desired destination, that is, North America or Europe (Paul 2011).

Therefore, education – or skills – can be considered the most important factor in understanding the multiple-migration phenomenon, but its effects and the strategies applied differ according to the origin country. In an early work on remigration from the United States, Borjas and Bratsberg (1994) demonstrated that selection on remigration depends not only upon the origin but also upon the skills and the reasons for remigration. For the authors, remigration occurs either after accumulating sufficient financial resources or because of a failure of the migration project, i.e., not being able to realize the expected economic opportunities in the host country. Additional research has shown that remigration tends to occur when facing difficult situations; unemployment, sickness or lack of integration foster individuals to seek more success in an onward destination or to return to the home country (Kelly and Hedman 2016; Toma and Castagnone 2015). In a study on Sweden, Nekby (2006) showed that selection processes of the highly qualified are stronger among onward migrants compared with return migrants. However, the research of Nekby also showed that there were important differences in onward migrations by country of origin. Other factors being equal, North and South Americans and Africans were more prone to undergo an onward movement than were West Europeans. In contrast, Nordic neighbouring country immigrants were less prone to move to an onward destination. The trajectories of multiple migrations are therefore highly associated with the origin and interact with the distance, a North-South divide, historical links between countries, and the education selectivity of the flows.

Previous research showed that individual factors are of great importance in the decision to migrate. A study on Senegalese migrants moving stepwise within Europe highlighted that having social or family networks in a third European country fosters the decision to re-migrate (Toma and Castagnone 2015). Family configurations are also important because migration involves a family decision process and should match the expectations of all family members (Mincer 1977). Having a partner and/or children can therefore be considered a barrier in undertaking a migration (Clark and Davies Withers 2007). Conversely, family formation is a reason for migrating when partners do not live in the same country, particularly for South-North family migration (Kofman 2004). The position in the life cycle is decisive because mobility behaviours are known to increase among young adults (25–35 years). In the European context, this phenomenon is exacerbated by international student programmes such as Erasmus, which supports student migration experiences in another country (Wächter 2014). In a cross-sectional study, age reveals not only the position in the life course but also period and generation effects. Thus, older individuals clearly had more time to have experienced multiple migrations (Schwartz 1976). This chapter does not intend to analyse age effects in detail but does control for position in the life course.

If aspirations, social position, and networks shape the planned – or unplanned – individual migration trajectories, institutional factors, such as nationality and legal status, can definitely eliminate the opportunities to migrate (Paul 2011). In the EU, built as a fortress open on the inside but closed to the outside, there is a divide between those who have extended access to any European country (EU citizens and highly qualified migrants) and those, in less favourable situations, who cannot move to their desired destination (King 2002; Van Mol and de Valk 2016). The labour market rules the possibility to settle because employment is the main entry gate to any European country. Finding a job is often a prerequisite for citizens of non-EU countries who aspire to settle in the European Union. The legal alternatives open to non-EU citizens involve migrating through family reunification, for studying, or as asylum seekers.

#### *4.2.3 Research Hypotheses*

This chapter describes the main characteristics of frequent movers who have recently arrived in Switzerland.

It focusses on the articulation between origin, level of education, and family structure. In particular, we aim at testing the three following hypotheses:

	- (a) Citizens from EU countries tend to be more mobile and use the facilitated access to the union labour market to move and work elsewhere.
	- (b) Neighbouring countries and countries with a long history of migration with Switzerland (Portugal and Spain) however come more directly to Switzerland and experience less mobility.
	- (c) Low-qualified southern country migrants tend to be more mobile, and follow a stepwise migration strategy.

H3. Family (partner and children) restrains highly mobile trajectories.

This research also aims at determining in an explorative setting the main factors that have made individuals undertake one more move and arrive in Switzerland. In Sect. 4.6, we also explore the trajectories of multiple migrants in analysing the last country of residence.

#### **4.3 Data and Methods**

This research is based on the Migration-Mobility Survey, a survey on recent immigrants in 11 selected groups of migrants living in Switzerland by the end of 2016. For the analyses, we exclude 43 individuals who had another nationality than the selected ones in the sampling procedure. Our sample size reaches 5930 individuals representing 455,733 migrants; for more details on the survey, see Chap. 2.

The main dependent variable is the number of countries lived in by the migrant before arriving in Switzerland. The question is as follows: "Except Switzerland and your country of birth, in how many different countries have you lived for three or more months?" The answer ranges from 0 to 19, but less than 9% of migrants lived in three or more other countries.

In the next section, we display the weighted distribution of the number of countries lived in for the three categories, "0", "1", and "2 or more" by nationality, level of education and family composition. We then examine whether the reasons for migration can be associated with multiple migrations. In Sect. 4.5, we run multinomial logistic regression models to analyse, ceteris paribus, the contribution of individual characteristics in explaining multiple migration trajectories. In Sect. 4.6, we examine the trajectories of multiple migrants and, more precisely, the last country lived in. In the final Sect. 4.7, we discuss the main finding, and to conclude, we check whether multiple movers expect to be mobile in forthcoming years.

The independent variables used in this research are as follows: level of education (secondary or less, technical and professional training, Bachelor, and Master/PhD), nationality (Germany, France, Italy, Austria, Portugal, Spain, United Kingdom (UK), North America, India, West Africa, and South America), family configuration at arrival (built using the partner status at arrival and the date of birth of the first child), self-declared reasons for migration (multiple choices between professional, study, start a family, accompany a family, lifestyle reasons, new experience, and other), and age and gender as control variables.

#### **4.4 Socio-demographic Profiles of Multiple Migrants**

In the population under study, almost one-half of immigrants have lived in a country other than their country of birth before coming to Switzerland: 27% had a unique experience in another country, 11% in two countries, and 9% in three countries or more.

#### *4.4.1 Education*

Recent migration flows to Switzerland are characterized by a general high level of education. In the population under study, 53% of the immigrants hold an academic degree. Compared with the one-time movers, multiple migrants are better educated (Table 4.1): the higher educated, the more mobile. The share of academics among the one-time movers is approximately 40%; it increases to 61% for migrants who have lived in one other country and to 76% for those who have lived in two or more different countries.

Previous research highlighted that the highly skilled are more mobile, a point particularly true in Switzerland. The Swiss labour market faces a shortage of highly qualified workers and tries to attract the foreign highly skilled. Scholars have explained that multiple migrations by the highly educated was a strategy to



Note: Cramer's V = 0.217

strengthen professional skills (Kelly and Hedman 2016; Takenaka 2007). Those who have been mobile appear to have ease of access to the job market in Switzerland. In parallel, institutional factors foster the mobility of those who are going to be highly qualified. In the Bologna context, there has been an internationalization of university studies. Numerous students experience a mobility for a few months in a programme such as Erasmus or Socrates, or for their whole Bachelor or Master study (Wächter 2014) because studying abroad has been shown to have a positive effect on labour market integration (Parey and Waldinger 2011). Thus, individuals holding an academic degree have been more mobile during their courses.

#### *4.4.2 Nationality*

Table 4.2 shows the large heterogeneity in the migration trajectories of international migrants by nationality. Approximately 70% of recent migrants from the United Kingdom and North America have lived in third country prior to their arrival in Switzerland – 42% have even lived in two or more other countries. Among French, Austrian and West African migrants, the share of multiple migrants rises to between 52% and 59%. For Spain, India, and South America, the proportion of multiple migrants lies between 48% and 50%. In contrast, the most numerous origin groups in Switzerland, Italian, German, and particularly Portuguese are less prone to multiple migrations with, respectively, 45%, 44%, and 32% being multiple migrants.

These differences cannot be explained only by univariate factors, i.e., the ease of entry in Switzerland (permit), geographical proximity, or a North-South divide as observed in the United States by Takenaka (2007). It appears that the phenomenon of multiple migrations simultaneously includes several factors covered by national-


**Table 4.2** Row distribution of the number of other countries lived in before coming to Switzerland by nationality (in %)

Note: Cramer's V = 0.167

ity. The main entry door to Switzerland is the labour market, which is profoundly segmented among migrants (Pecoraro 2005). This segmentation relies on the labour market structure and the need for both highly qualified and low-qualified workforces.

The less mobile Portuguese have for instance a long tradition of low-qualified migration to Switzerland that started in the 1980s (Piguet 2009). Since then, historical links and family and social networks have fostered a continuum of direct migrations. In the population under study, Portuguese are the least qualified group, with only 17% holding an academic degree. The share of academics stands at 61% for all of the other nationalities.

Italians and Germans are the two largest foreign communities in Switzerland. Both have developed strong labour immigration links to Switzerland (Piguet 2009). Those links and their cultural (language) and geographical proximity most likely explain why the share of multiple migrants is relatively low among them. Surprisingly, this point is not true for French and Austrian migrants, who despite coming from a neighbouring country have more frequently experienced multiple migrations before arriving in Switzerland. UK, North American and Indian migrants are characterized by a high level of education and highly qualified occupations in Switzerland, but only the first two are characterized by a high degree of international mobility. West Africans and South Americans are more diverse and less qualified, but they are over-represented among those who have lived in one other country before arriving in Switzerland.

#### *4.4.3 Family Composition*

Family composition at arrival does not have a clear influence on the number of previous migrations (Table 4.3). Only single parents have been less numerous in experiencing multiple migrations prior their arrival in Switzerland. The share of multiple migrants is approximately the same among those who were single or in a relationship at migration, having or not having a child. In other words, partnership and family are not obstacles to having experienced multiple migrations. Thus, the multiple migrations belong to a family decision process. Therefore, we can expect that


**Table 4.3** Row distribution of the number of other countries lived in before coming to Switzerland by family situation (in %)

Note: Cramer's V = 0.027

multiple migrations refer not only to an individual strategy of gaining individual migrant capital but also at the collective, family level, strategy as theorized by Mincer (1977).

#### *4.4.4 Reasons to Migrate to Switzerland and Multiple Migrations*

Multiple migrations have often been interpreted as a strategy to gain migration capital, skills, and experience that foster professional careers. In this explanatory section, we would like to understand whether the reasons for moving to Switzerland changed according to the past trajectory, i.e., whether the decision to move is the same for those individuals who experienced their very first migration and for those who have previously lived in one or more countries. The Migration-Mobility Survey asked about the reasons for immigrating to Switzerland; the respondents could choose one or more responses from among 10 propositions.

Almost 62% of the new immigrants indicated that they migrated for professional reasons. Family reasons are also important; 28% of the weighted sample indicated migrating either to accompany a family or to start a family. Then, lifestyle reasons and envy to gain new experiences were cited by, respectively, 19% and 18% of the respondents. Except coming for studying (8%), other reasons to migrate only collected a few percentages.

Table 4.4 shows how the cited reasons vary according to the number of previous countries lived in. It is not surprising that the share of individuals who mention professional reasons to migrate increases with the number of multiple migrations. Professional reasons are indeed more often cited by highly qualified individuals. Thus, 56% of the individuals who have only lived in their birth country prior coming to Switzerland cited professional reasons. The percentage rises to 65% for those who have lived in one other country, and it increases to 73% for the more mobile.


**Table 4.4** Reasons to immigrate by the number of other countries lived in before coming to Switzerland (in %)

Note: N = 5930

In contrast, the share of those indicating family reasons tends to decrease with the number of countries lived in. This point is particularly true for those who want to start a family. Family reasons were cited by 14% of the direct migrants, by 11% of those who have lived in another country, and by 8% among the more mobile; for those who indicated accompanying their family, the gradient is slighter at, respectively, 18%, 16%, and 15%.

Lifestyle factors and envy to gain new experiences in Switzerland were also less cited by the multiple movers than by those who came without a secondary migration. A similar pattern can be observed for those who cited other reasons to migrate.

#### **4.5 Explaining Multiple Migrations: A Multivariate Analysis**

In the previous sections, the diversity of the multiple migrants has been highlighted. Some factors appear to play a determinant role in the migration trajectories and the decisions to migrate to Switzerland. There are of course interdependences between these factors that a multivariate analysis will be able to disentangle. A multinomial logistic regression has been run to identify the most important drivers of multiple migrations, all other factors being equal. The reference category of the dependent variable involves the individuals who have not lived in any country other than their birth country and Switzerland. The variables presented in the previous section are included in the model; age at migration and gender are also included as control variables. The results are presented in Table 4.5.

The coefficients in the table indicate the change from the reference category of the log odds of having experienced multiple migrations. For example, for the first coefficient, French individuals are, all other factors being equal, exp.(0.408) = 1.5 more likely to experience one migration compared with the likelihood that Germans will.

First, despite the control for education levels and reasons for migrating, there remain important differences in the migration experiences between nationalities. Except for Indian immigrants, the larger communities – German, Portuguese, and Italian – have been less prone to have experienced multiple migrations. In contrast, migrants from the United Kingdom and North America appear to have experienced the highest mobility before arriving to Switzerland; they have, respectively, 4 and 3.2 times higher likelihood to have experienced two or more multiple migrations than have the German migrants. For most nationalities, compared with the Germans, there is an increase – or stagnation – in the likelihood of having done an increasing number of multiple migrations. Italians are the exception with a decrease; they are more likely than Germans to have moved once but equally likely to have moved twice or more.

In the regression models, the education level appears the most important driver of multiple migrations1 : a clear gradient can be observed between an increase in the education level and an increase in the number of migration experiences. Thus, edu-

<sup>1</sup>This variable has the highest effect on the model quality in an AIC stepwise model selection (forward and backward).


**Table 4.5** Results of the multinomial logistic regression on the number of previous countries lived in, reference no other country lived in

Note: \*p < 0.05; \*\*p < 0.01; \*\*\*p < 0.001. N = 5930

Source: Migration-Mobility Survey 2016. Weighted results

cational attainment level is the key dimension of the recent rise in multiple migrations in the European context.

Despite the control for education, migrants who have come for professional reasons are more prone to move internationally before arriving in Switzerland. Accompanying a family or coming for study reasons has no significant effects on explaining multiple migrations; those who have cited these reasons are neither more nor less mobile. However, migrants who intend to start a family have experienced less mobility.

Concerning the control variables, no gender effect could be observed. Fewer individuals aged less than 30 years have lived in two or more countries, most likely because they had less time to do so.

#### **4.6 Arrival Trajectories of Multiple Migrants**

Section 4.5 highlighted an important variability in the migration practices by origin, all other factors being equal. However, little is known about the trajectories that led multiple migrants to Switzerland. This question has not been widely studied mostly because of the lack of good data. It can however bring new insights on the strategies developed by international migrants. Indeed, migration from southern countries has often been interpreted as a stepwise strategy that brings step-by-step migrants to their desired destination (Paul 2011; Takenaka 2007). What about migrants from other origins? Analysing the last country of residence allows understanding the main pathways used to arrive in Switzerland.

Table 4.6 shows the distribution of the last country of residence of multiple migrants. The first column "nationality" indicates migrants who came directly from their country of nationality and who have lived in one or more other countries in the past. The other columns are divided by continent except for Germany, Spain and the UK, which gather large flows of onward migrants to Switzerland.

First, more than one-half of multiple migrants have come directly from their country of nationality. The share of *direct* migrants is higher for the Germans and French, above 60%. As expected, the lowest percentages are observed for southern country migrants – 27% for West Africans, 28% for Indians, and 37% for South Americans; all of those countries have approximatively the same share of multiple migrants (see Table 4.2). Interestingly, a majority of stepwise southern migrants have come from a small number of European countries. The trajectories of West African multiple movers have mostly passed through Italy, France, Spain and the United Kingdom (40%), those from India through Germany, North America, and the United Kingdom (35%), and those from South America through Spain, the UK, and Germany (32%). However, 20% of the Indian multiple migrants have lived in another Asian or Oceanian country before coming to Switzerland. For West Africans, 19% have previously lived in another African country, but only 6% of South American multiple migrants have come to Switzerland through another South American country. Thus, multiple South American migrants differ in their trajectory by having a very low mobility in the sub-continent before moving to Switzerland because their preferred destination is Spain due to their region's long-standing cultural and economic relations with Spain.

For European and North American immigrants, the last country of residence of a majority of multiple movers was the country of nationality. This means that the migration episodes were interrupted by a return to the home country. For the minority who has come through stepwise movements, some previous countries of residence emerge as dominant: Germany for Austrians (15%), the UK for Italians



Note: N = 3014

(12%), and Spain and French for Portuguese (11% and 10%). Furthermore, North America appears an important hub of European migrants who thereafter moved to Switzerland. Between 7% and 9% of multiple migrants from Italy, Germany, France and Austria who have arrived in Switzerland have lived previously in North America, almost exclusively in the United States.

These exploratory results have shown very different pathways to arrive in Switzerland among multiple migrants. If some evident pattern of stepwise strategies appears for West Africans who immigrated to southern Europe (Italy, France and Spain) before arriving to Switzerland, or for South Americans who initially arrived in Spain before being able to move to Switzerland, more research is needed to understand and explain the longitudinal trajectories of multiple international migrants.

#### **4.7 Discussion and Conclusions**

This chapter has shown that multiple migrations are very common among recent migrants. Almost one-half (47%) of the recent migrants surveyed through this study have lived in one or more countries before their arrival in Switzerland. The origins of this phenomenon are diverse, but it clearly demonstrates the globalization of mobility among migrants.

In the relatively rare academic literature on the topic, a sub-population is known to be very mobile – the highly skilled. This research confirmed the predominance of the highly skilled among multiple migrants (our first hypothesis); among those who have lived in two or more countries, 76% have a university degree. The highly skilled benefit from multiple experiences in various countries that improve in particular their professional skills, languages competences, and social and professional networks. After gaining experiences in one or more other countries, they are more valued in the labour market. Because Switzerland faces a highly skilled worker shortage, those who have been mobile could most likely gain facilitated access to a job. Furthermore, with the internationalization of university studies, an increasing number of students chose a mobility during their studies. Thus, some highly educated individuals are inherently multiple migrants by means of a mobility during their studies.

This research has shown an important heterogeneity in multiple migration experiences between countries. In the context of free movement of persons between the European Union and Switzerland, it was expected that European nationals, who have a facilitated settlement to every EU country and Switzerland, experienced more mobility before their arrival in Switzerland (Hypothesis 2a). Although migrants from the UK confirm this trend, it cannot be observed for most European nationals. Furthermore, the second-most-mobile group comes from North America. In fact, as shown by Nekby (2006) in the Swedish context, migrants coming from neighbouring countries tend to experience less mobility (Hypothesis 2b). This research showed that not only geographic but also cultural distances reduced the complexity of the previous international migrations. Therefore, the lowest mobility is observed in the largest communities – the neighbouring German and Italian, and the Portuguese, who have a long history of migration with Switzerland. Their lower propensity for multiple migrations demonstrates the dominant role of historical links and of social and family networks in fostering trajectories that are more direct. However, despite their geographical proximities, migrants coming from France and Austria are surprisingly more mobile than are those from the other neighbouring countries. For Austria, the analysis in Sect. 4.6 showed that many multiple migrants have had a previous experience in Germany. Their higher mobility can thus be explained by a type of stepwise trajectory to Switzerland.

Spain and Switzerland also have a long history of migration; Spaniards are today the fifth-largest foreign group in Switzerland. The apparent high mobility among recent migrants in the descriptive and multivariate analyses is actually overestimated and resembles the behaviours of the Portuguese. This similarity is due to the large proportion of Spanish migrants who were born outside Spain and obtained the Spanish citizenship. Indeed, almost 20% of the holders of a Spanish nationality are born abroad, and most of them come from South America. Latin American migration to Spain has experienced a dramatic increase in the last two decades. Because of a facilitated naturalization process – 2 years of continuous legal residence in Spain (Hierro 2016) – in a time of economic crisis and as their socio-economic condition deteriorates in Spain, naturalized immigrants from South America had the opportunity to move within the Schengen area to seek better job opportunities and to fulfil a multiple migration strategy (Mas Giralt 2017). Many South Americans who lived in Spain but did not naturalize most likely also followed this pathway (Sect. 4.5), seeking better life conditions in Switzerland.

It was hypothesized that southern country migrants tend to use a step-by-step migration strategy to reach a destination that improves their socio-economic conditions (Hypothesis 2c). This point is partially true for South American migrants who massively immigrated to Spain and moved on thereafter; Mas Giralt (2017) showed that their onward movement was not planned but can be interpreted as a coping strategy after the rise of the economic crisis. Among West African migrants, multiple movers mostly lived in only one previous country; France, Spain and Italy are the entry doors to Europe for African migrants who try to move onward when facing vulnerable situations (Toma and Castagnone 2015). Indian migrants are an exception because they appear in the multivariate analysis to have experienced a low level of multiple migrations despite their high average level of education. India is known to export highly skilled workers in the information technology and engineering sectors (Chanda and Sreenivasan 2006). Thus, Indian migrants immigrate more easily, due to a high demand of these sectors in the Swiss labour market.

The analyses in this chapter have also shown that professional reasons are the most important driver of recent migrations to Switzerland (62%). Professionals have been more mobile in their course regardless of their levels of education. Contrary to the hypothesis on family migration (see our third hypothesis), the family configuration (having a partner or children), or having a reason such as accompanying a family, does not reduce the number of previous international migrations.

Family is therefore a component of multiple migrations that will be analysed in more depth in Chaps. 5 and 6 of this book.

This chapter has given an overview of the current multiple migration phenomena in Switzerland. When asked about their future intentions, the respondents of the survey answered, in a majority (58%), that they do not yet know if they want to settle in or remigrate from Switzerland. Among the remaining 42%, 80% aim at staying forever in Switzerland. The share rises slightly among the one-time migrants to 84%. Indeed, the country's quality of life is found among the highest in the world in many studies, as is its purchasing power. Is Switzerland the last stop for multiple migrants? For now, it is the last stop for those who have decided to stay, but for many individuals, the future of their migration trajectory remains unknown.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Part III Labour Market Participation**

## **Chapter 5 Integration of Recently Arrived Migrants in the Swiss Labour Market – Do the Reasons for Migration Matter?**

**Philippe Wanner**

#### **5.1 Introduction**

Over the last three decades, Switzerland has observed huge changes in the structure of migratory flows (see Chap. 1). Although the country was already characterized by the arrival of groups of immigrants with various characteristics and levels of technical skills, a new trend in migration flows has emerged, with an increased proportion of highly skilled migrants being registered in the last 10 years (Wanner et al. 2016). In the framework of the so-called Migration-Mobility Nexus (see Chap. 1), this migration is largely explained by demand-led immigration provoked by a rapid transformation towards a more specialized economy and by a shortage of human resources in some high value-added sectors, such as IT services, health and specific financial sectors. Other factors, such as taxation, quality of work and business expansion in some branches, can also play an important role in the decision of highly skilled migrants to relocate (Mahroum 2000); these factors have also been observed in Switzerland. Finally, changes in the legislation concerning labour workers in relation to the ratification of bilateral agreements between Switzerland and the EU have accelerated the trend towards highly skilled immigration because it is now easier to hire EU workers.

This transformation of the flows has led to an increase in the average age of migrants, who increasingly frequently move in their thirties or even forties, often after the constitution of their families, resulting in a progressive increase in the proportion of family migrants since the 1970s.1 However, the preliminary results of the

© The Author(s) 2019 103

<sup>1</sup>According to the Swiss State Secretariat for Migration, more than 30% of the migration to P. Wanner (\*)

Institute of Demography and Socioeconomics IDESO, University of Geneva, Geneva, Switzerland

nccr – on the move, Neuchâtel, Switzerland e-mail: Philippe.Wanner@unige.ch

I. Steiner, P. Wanner (eds.), *Migrants and Expats: The Swiss Migration and Mobility Nexus*, IMISCOE Research Series, https://doi.org/10.1007/978-3-030-05671-1\_5

Migration-Mobility Survey 2016 demonstrate that professional factors remain the main reason for migration, concerning almost two-thirds of migrants. The characteristics of immigration towards Switzerland are very complex, and the motives to immigrate are varied. For instance, almost one-third of those surveyed provide more than two reasons for their migration.

With the development of international migration, a considerable amount of research has been performed on the issue of structural integration, i.e., at school and in the labour market (for Switzerland, see for instance Liebig et al. (2012) and Fibbi et al. (2006)). Beyond measuring the level and occasionally the pace of integration, most of the research also identifies the individual factors, if any, that influence migrants' integration: for instance, educational background, country of origin, age, and family structure; or contextual factors such as integration policies, discrimination or stereotypes concerning some groups of migrants. One aspect that has thus far been poorly explored is how the reasons behind migration affect integration (Kofman 2004; Kofman and Raghuram 2005). The lack of studies on this topic is notably explained by the limited availability of appropriate data (Bevelander and Pendakur 2014). Surveys and population registers rarely contain information on both the reasons for migration and the status in the labour market. In general, researchers who consider this topic focus on one specific migrant group, most frequently asylum seekers or refugees (for instance, Rashid (2009) for Sweden and DeVoretz et al. (2004) for Canada), who are the most concerned with integration issues. The researchers do not adopt a comparative approach among migrants categorized according to their reasons for migration.

However, the level of integration and satisfaction in the labour market can differ according to the context in which migration has occurred. In particular, it is expected that the so-called "secondary migrants", "tied migrants" or "family migrants" have lower levels of integration into the labour market (Meares 2010) due to the time needed to find a job corresponding to their skills, a situation that is conducive to poor usage of the skills available (brain waste). A "tied" or "secondary migrant" is generally defined as an individual in a family that decided to move due to the employment status of his partner but who, if single, would have stayed (Cooke 2013). Tied migrants are usually women following their partners to a foreign country, although a recent study in the United States demonstrated that this phenomenon is not limited to women (Cooke 2013).

For the host country, it is important that secondary migrants can also access the labour market not only to realize their economic potential but also to contribute to their effective integration into the host society. It is also expected that their level of satisfaction with the migration is lower in the case of difficulties in accessing the labour market, which can also lead to different health-related issues.

It is in this context, using the Migration-Mobility Survey, that this paper aims at measuring the extent to which the reason for the migration affects the level of

Switzerland is today explained by family reunification. Although no data are available in the long term, one can state that this proportion is far higher than in the period of immigration of guest workers, between the end of World War II to 1970 approximately.

integration into the labour market. We use four indicators to infer the position of migrants in the labour market and their level of satisfaction concerning the working conditions. We aim at assessing the effect of the reasons for migration on those indicators. We also aim at identifying which other individual migrant characteristics influence the level of integration into the labour market. The hypothesis tested in this chapter is that, after considering other confounding variables, the reason to migrate is significantly correlated with our indictors.

The one question in the survey documenting the reason for migration, which was self-declared by the respondents, makes it possible to test such a hypothesis. One can however observe that the decision to migrate is also increasingly diverse and complex to identify precisely, which can make it difficult to link the declared motive with the level of structural integration (Luthra et al. 2014).

#### **5.2 State of the Art**

As mentioned below, the situation of family migrants is rarely documented, in particular because of the focus on migration by men and its economical outcome (Webb 2015). In contrast, the level of integration in the labour market is not only one of the objectives of any immigration policy2 but also the topic of numerous studies in industrialized countries. The labour force participation of migrants, particularly those coming from southern countries, refers to different dimensions addressing national legislation (for instance referring to the recognition of foreign qualifications), the individual characteristics of migrants, or attitudes of employers, for instance in terms of discrimination practices (OECD and European Union 2015). Based on the literature, one can observe that immigrants, whatever the motive for migration, need time to integrate into the host labour market in a position that corresponds to their skills. Based on an analysis of immigrant earnings in different host countries, (Adsera and Chiswick 2007) show that approximately 18 years are necessary for immigrants to close the gap with natives in terms of wages, after considering confounding variables.

Poor skills in the local language are frequently mentioned as being among the individual factors hampering this integration into the labour market (Dustmann and Fabbri 2003). Conversely, social networks in the host country, social contacts with the host population, and a high level of education (Amuedo-Dorantes and De la Rica 2007) can accelerate the pace of integration into the labour market. Concerning education, one acquired in the host country is generally considered more useful for integration into the labour market than is the one acquired in the country of origin (Akresh 2008). Country of origin, age and household composition are other factors influencing the level of structural integration. Gender is also mentioned in studies as

<sup>2</sup>See for instance the action plan of the European Commission, https://ec.europa.eu/home-affairs/ what-we-do/policies/legal-migration/integration/action-plan-integration-third-country-nationals\_ en. Accessed 29 Mai 2018.

a factor influencing integration (Blau et al. 2011). The influence of the motives behind the decision to migrate is rarely considered in studies measuring the factors of integration, although such motives provide an understanding of the factors triggering the migration and therefore must be explored.

Migration theory is progressively addressing the increasing complexity of migration. In fact, rational theories based on the neoclassical approach, which links the mobility of individuals to wage differentials, has progressively diminished in importance and relevance as other, more sophisticated theories have emerged, for instance the new economics theory of migration (Harbison 1981; Stark 1991). In contrast to earlier theories, which stated that migration is an individual decision, the new approach suggests that the decision to migrate is made after considering different dimensions or potential effects that the choice can provoke not only for the migrant but also for his/her family. In our case, the decision to migrate in Switzerland is rarely an individual decision but increasingly a collective decision made within the family (Nauck and Settles 2001).

In the context of highly skilled migration, the personal motivations of the primary migrant (in particular, the one who has secured employment in the Swiss labour market) might collide with the personal expectations of the partner (or secondary migrant). This point is particularly true when – generally highly educated – partners give up their current employment to accompany the primary migrant in the host country without having secured a job for the time of arrival. Occasionally, the partner is not ready to lose his/her situation and stays in the country of origin ("tied stayer", see Cooke (2013). However, the partner, a "tied migrant", often accepts being confronted with issues concerning his integration into the Swiss labour market.

Family migrants are usually confronted with greater difficulties than are primary migrants in integrating professionally (see for instance in Canada Man (2004). The difficulties are amplified by the fact that highly skilled migration generally occurs when one is older than is true of low-skilled migration, at a period in life when education has been completed, professional skills are precisely defined and professional flexibility is relatively low compared with the situation of younger people. This situation can contribute to making it more difficult for partners of primary migrants to reintegrate into the labour market in a position similar to the one occupied before the move (Webb 2015).

However, due to the dearth of research on the topic, the literature does not provide a clear understanding of the relationship between the motives for migration and the level of structural integration and satisfaction in the labour market. One of the publications addressing the topic is that of Aydemir (2011). Working on the characteristics of immigration towards Canada and using the Longitudinal Survey of Immigrants to Canada – LSIC, the author confirms that the increase in the share of higher-skilled migrants has affected the spouses' skill distribution. Due to the homogamy between both partners, spouses are more qualified than in the past. The immigration of highly skilled workers is associated with the arrival of highly qualified partners. Moreover, the study observes that the overrepresentation of higherskilled immigrants does not necessarily lead to better overall performance in the labour market due to the difficulty those migrants have with transferring Canadian human capital acquired abroad. In Australia, Webb (2015) stated, "the migration of highly skilled tertiary-educated people, particularly female partner migrants, is often risky, resulting in disrupted careers, intensification of domestic responsibilities and a re-feminization of health and human service work". According to the author, female migrants face downward occupational mobility and re-orientation. Secondary applicants appear to be more frequently unemployed or partially employed in Australia. A negative effect on job retention among wives (tied migrants) was also observed in the United Kingdom and the United States (Boyle et al. 2001).

Another study using longitudinal data from Canada specifically analysed the link between the motive for migration and the level of integration (CIC 1998). It measured the economic performance of migrants (based on five indicators – level of employment, earnings, incidence of employment earnings, incidence of unemployment insurance benefits and incidence of social assistance payments) for four categories of migrants: principal economic applicants, spouses of economic migrants, immigrants linked to family reunification3 and refugees. The authors demonstrate that a good performance in the labour market is observed for the first category. The last two categories show low performance. The spouses and dependents are characterized not only by low income but also by a high risk of unemployment and high rates of social assistance payment.

This relatively poor performance of spouses or family migrants in the labour market is not observed everywhere. For instance, Nekby (2002), studying the employment convergence of immigrants and natives in Sweden using a longitudinal database for the last decade of the twentieth Century and covering 200,000 individuals of whom 19,000 were born abroad, observed "no notable gender differences in employment convergence". These results are in line with those observed in the United States by Blau et al. (2002), who find that "both husbands and wives work and earn less upon arrival than comparable natives, with similar shortfalls for men and women". Furthermore, "both immigrant husbands and wives have similar, positive assimilation profiles in wages and labour supply and eventually overtake both the wages and the labour supply of comparable natives."

The lack of convergence in the literature concerning the position of family migrants in the labour market, particularly in a context of highly skilled migration, justifies further studies. However, an absence of data on the motives for migration hinders such studies. In general, only the administrative reason for migration is available in European statistical data. However, the recorded reason for providing the permit does not always correspond to the actual (subjective) reason. Only through specific surveys can information on the actual motives behind the decision to migrate be obtained and thereby help to clarify the link between the motive(s) and the level of structural integration.

<sup>3</sup>This category includes migrants arriving after a family member is already living in Canada.

#### **5.3 Data and Methods**

The Migration-Mobility Survey 2016, which is described in detail in Chap. 2, was conducted on a population of persons born abroad and with foreign citizenship who have moved to Switzerland in the last 10 years. To be included in the sample population, respondents had to be between 24 and 64 years of age at the time of the survey and at least 18 years of age on their arrival in Switzerland. A total of 5973 people replied to the full questionnaire, either online or by telephone. The survey provides information on the individual's subjective reasons for migration. It also provides information on the existence of an employment contract (job and job offer) in Switzerland at the time of immigration. Based on this information, we construct a typology of reasons for migration, distinguishing between labour-related, familyrelated and other reasons detailed in the next paragraph. Structural integration is assessed using answers to the four questions described below.

#### *5.3.1 Definition of the Motives to Migrate*

The survey asked for the reasons for migrating to Switzerland. The offered choices included the following: professional reasons, educational and/or study reasons, to start a family, to accompany family member(s), lifestyle reasons, to gain new experience, presence of a social network in Switzerland, tax reasons, political reasons, and other reasons. Respondents were invited to indicate multiple answers, but the survey did not request respondents to rank the answers according to their importance. Professional reasons are the most frequently given reasons to migrate because 61.6% of the respondents mentioned them. Among the other reasons mentioned, we find lifestyle (19.6%), gaining new experience (18.8%), accompanying family (17.1%) and starting a family (12.0%). Compared with the administrative reasons for migration available from the statistics of foreigners,4 the survey underestimates family reasons (31.3%) and overestimates professional motives (46.9%). This divergence is explained by the fact that partners of "professional" migrants are often considered "family-related" migrants by the administration, although their main and self-declared motivation to move to Switzerland is to work. It is understandable that some of the migrants arriving with a family-related administrative permit come to Switzerland with work as the primary motive.

After different attempts at classification, the variable "Reason to immigrate" is defined using five categories:

• Professional reasons mentioned (including study) but no family reasons, with job or job offer (n = 2263, 69.5%5 men, 30.5% women)

<sup>4</sup>SEM, Statistique sur l'immigration, les étrangers en Suisse. https://www.sem.admin.ch/dam/data/ sem/publiservice/statistik/auslaenderstatistik/monitor/2016/statistik-zuwanderung-2016-12-jahrf.pdf. Accessed 18 June 2016.

<sup>5</sup>Weighted proportions.


#### *5.3.2 Labour Market Integration*

In the survey, labour market integration was measured according to four indicators that were considered separately: (1) declaring a substantial or slight improvement in the current professional situation, compared with the situation before the move6 ; (2) declaring himself currently seeking a job (unemployed); (3) declaring himself currently working under short-term contract or without contract; and (4) not making use of all skills in the current job?.7 The choice of indicators is determined by their availability in the survey. The last three indicators are also integrated in the OECD range of indicators of immigrant integration (OECD and European Union 2015).

#### *5.3.3 Methods*

Descriptive results are displayed to describe the characteristics of the sample according to the reason for migration and gender. In a second part, logistic regressions (binomial and multinomial) are applied on the indicators to assess the effect of the reason for migration on the level of structural integration. Logistic regressions aim at explaining the probability (p) of answering one question (for instance, whether seeking a job) positively in relation to different explanatory variables. The formula is as follows:

$$\log \text{fit}\left(p\right) = \ln \left(\frac{p}{1-p}\right) = \beta\_0 + \beta\_1 \mathbf{x}\_{l,1} + \beta\_2 \mathbf{x}\_{l,2} + \dots$$

<sup>6</sup>The question was formulated as follows: "Concerning your professional situation, what would you say overall when comparing your situation today with your situation before moving to Switzerland?"

<sup>7</sup>The question was formulated as follows: "On a scale from 0 (not at all) to 7 (to a very high extent), to what extent are your knowledge and overall skills utilized in your current work? By knowledge and overall skills, we mean your formal education and the skills you obtained while working (onthe-job training)". After considering the distribution of the answers, we considered scores 6 and 7 to indicate a good use of skills, and scores between 0 and 5 to indicate a poor (or no) use of skills.

with *β*0 being a constant and *β*1,…*n* being the coefficients of the explanatory variables x\_(1,… n). The exponential value of *β*1,…*n* represents the odds ratios. A value greater than one indicates an increased risk for a group (for instance women) of being integrated in comparison to the reference group (for instance men, (cf. Cox and Snell (1989)). Confidence intervals at the 95% limit and significance levels (\* p < 0.05; \*\* p < 0.01; \*\*\* p < 0.001) are also presented.

To control for possible confounding factors intervening in the relationship between the reason for migration and the level of integration, the models also include different variables suspected of affecting integration, according to the literature mentioned in the previous part of this chapter. The variables introduced in the model are the following:


For indicators referring specifically to the professional outcome (type of contract, use of skills and satisfaction), we retain only occupied persons. We deliberately excluded non-active persons because the indicators do not refer to them. However, we are aware that the sub-group of inactive persons includes migrants who have given up on the idea of entering the labour market due to failure in their quest for a job. The indicator of unemployment refers to active migrant, whether or not occupied. Models are also run on the sub-sample of those arriving for family reasons to identify the sociodemographic factors intervening in the integration of family migrants.

#### **5.4 Results**

#### *5.4.1 Reasons for Migration and Context in Which Migration Occurs*

Table 5.1 presents the distribution of the reasons for immigrating among the currently active population (whether or not occupied) according to some personal characteristics. Active men are more concerned with professional reasons than are women (70% versus 57%), and more than 40% of them had obtained a work


**Table 5.1** Distribution of the reasons to immigrate according to the different characteristics (in %)

Note: Economically active population only

contract before migrating. Thirty percent of the women only mention family reasons to explain their arrival in Switzerland (11% for men). Professional migration is predominant among every group of migrants except for those from South America and West Africa, who are primarily concerned with family migration.

#### *5.4.2 Integration in the Labour Market: Descriptive Results*

Table 5.2 provides descriptive statistics according to the four selected indicators. For women migrants, we observe a lower level of improvement in integration into the labour market, a higher level of unemployment, a slightly higher level of shortterm contracts and a lower use of skills compared with men. In a country with low unemployment rates and high wages, the fact that almost 20% of migrants declare not having improved their situation in the labour market after the move most likely reflects the migrants' difficulty in integrating into the labour market, which is more an issue among women than among men.

Table 5.2 also shows that migrants who arrived for professional reasons show better labour market outcomes compared with those who arrived for either family or other reasons. In particular, they declare a better use of skills and an improvement of the situation in the labour market compared with their situation prior to migration. Concerning the improvement of their own professional position, professional


**Table 5.2** Description of the indicators of integration into the labour market according to gender, reason to immigrate, and the presence of a job or job offer before migration (in %)

Note: a Economically active population only, also excluding migrants seeking a job. b After the exclusion of independent persons and owners of their own enterprises. c Total numbers can differ according to the indicator due to the definition of the population (for example, economically active or occupied) and missing values

migrants who had arrived without a contract are slightly more satisfied than are those who arrived with a job or a job offer. Because of the relatively low level of unemployment in the country, the proportion of persons seeking a job is low for every group. However, this proportion is the highest among migrants who arrived for professional reasons without a work contract, for family reasons only or for other reasons.

Overall, approximately 85% of active migrants have a permanent contract. A higher proportion of those who arrived for professional reasons but without a job at the time of the move also have a job with a short-term contract/without a work contract compared with migrants who arrived for other reasons (including family reasons).

#### *5.4.3 Relationship Between Reason for Migration and Labour Market Performance. Results of Logistic Regressions*

After considering the confounding factors, the logistic regressions show that the reason to migrate is strongly associated with the level of improvement in the labour market following migration, the risk of unemployment, and the extent to which previously acquired skills are used but not with the type of contract.

In more detail, migrants who have arrived for family reasons or for other reasons have a smaller probability of improving their work conditions following migration than do migrants who arrived for professional reasons with a work contract (O.R. 0.46\*\*\* and 0.61\*\*, respectively, Table 5.3). Migrants who arrived in Switzerland for both professional and family reasons do not present any significant variation concerning work conditions compared with those whose migration is only motivated by professional reasons. As expected, the results confirm that migrants who arrived for family reasons can more frequently encounter difficulties finding work that matches their personal aspirations. Interestingly, among professional migrants, having a contract at the time of arrival does not ensure the highest level of satisfaction. The interpretation of this result must consider the fact that the labour market functions better in Switzerland than in most of the countries of origin, having a wide range of professional opportunities and high wages. However, despite the relatively favourable situation, acceptable employment in the labour market is not ensured for every group of migrants, in particular for family or migrants who arrived for nonprofessional reasons.

Compared with the reference group, economically active migrants who arrived for family reasons (O.R. 2.21\*) or for other reasons (O.R. 1.40\*\*) more frequently state that their knowledge and skills are poorly used in their current work. Such results are in line with the hypothesis that family migrants face difficulties with integrating into the labour market in a position that corresponds to their skills. Among migrants who arrived in Switzerland for professional reasons, the risk of not adequately using their skills is higher among those without a job or job offer at the


Results of a logistic regression on indicators of the labour market

114


Improvement of work

conditions Confidence

O.R

Intervals

P

O.R

Intervals

P

O.R

Intervals

P

O.R

Intervals

P

Currently unemployed

Confidence

contract Confidence

Poor use of skills

Confidence

Limited contract/absence of


Source: Migration-Mobility Survey 2016. Weighted results (normalized weights). \* p<0.05; \*\* p<0.01; \*\*\* p<0.001 time of migration compared with those with a job (O.R. 1.36\*\*), suggesting that if job opportunities exist for this category of migrants, there is a risk of skills mismatch.

The labour market in Switzerland privileges permanent contracts in many sectors. Those with short-term contracts are primarily the academic sector, some international organizations and sectors with primarily seasonal activities. These sectors hire a high proportion of migrants; hence, the reason why short-term contracts are more frequent among migrants than among natives.8 The reason for migration does not significantly influence the probability of having a short-term contract.

As expected, the risk of having to seek a job is higher for all groups (except for those who arrived for both professional and family reasons) compared with migrants arriving with a job contract. It is highest among those who arrived for professional reasons but without a job (O.R. 2.54\*\*) or for other reasons (O.R. 2.57\*\*).

Significant differences are then observed between migrants arriving for professional reasons without a job contract and those having secured a contract before the migration, the former being systematically in a less favourable position. This situation can be explained by the fact that migrants arriving with a contract had the prior possibility to choose whether to migrate based on knowing what type of job they would have in Switzerland. The decision to move is therefore made based on the position that is offered. Those arriving without a job have no prior information concerning a potential future job and therefore are not able to make such an informed choice.

In addition to the effect of the reason to migrate, the model also shows the effect of different sociodemographic characteristics on the level of integration. Compared with men, and after considering the other confounding factors, women have a lower probability of improving their position in the labour market through the migration and a slightly higher probability of not using their skills adequately and of having a limited contract. They are also significantly more concerned by being unemployed. These results confirm those based on the descriptive analysis (Table 5.2) and suggest that even after considering the role of the reason for migration, women are disadvantaged on the labour market compared with men. Compared with migrants aged less than 30, older migrants present a lower probability of declaring an improvement of their work conditions. However, they also present a lower risk of poorly using their skills and a lower risk of having a short-term contract. Younger migrants thus appear more frequently satisfied with their work conditions, although they have not yet been able to find a job that matches their skills. This situation can be partially explained by the expectations of those young migrants towards ascending professional mobility, which make them consider their current position a temporary state (cf. theory of career mobility (Sicherman and Galor 1990)). Migrants who arrived in 2008 or before also show an increased probability of declaring having

<sup>8</sup>According to the Swiss Federal Statistical Office, the proportion of workers with contracts of limited duration is 8% among Swiss citizens and 11.3% among foreigners (Swiss Federal Statistical Office, Swiss Labour Force Survey, https://www.bfs.admin.ch/bfs/en/home/statistics/cataloguesdatabases.assetdetail.je-f-03.02.01.23.html. Accessed 5 June 2018).

improved their working conditions compared with those who arrived in the last 2 years. Recent migrants also present a higher risk of having a short-term contract.

Concerning the country/region of origin, non-European migrants present the highest probability of having contracts of limited duration. Southern Europeans more frequently declare having improved their situation on the labour market. Those concerned foremost by underuse of skills are Portuguese and French. There is therefore a paradox concerning the Portuguese migrants who declare an improvement in their professional position although their skills are underused. This paradox is most likely explained by the fact that such migrants generally come from a country with low professional opportunities at the time of their migration and then find better working conditions in Switzerland although they do not use all of their professional skills.

Results concerning the permit of residence and civil status are rarely significant, meaning that these dimensions do not affect the integration on the labour market. The only significant results refer first to a small decrease in the probability of having improved their professional situation among holders of annual resident permits (B permit) and second to a significant increase in the risk of a short-term contract among the groups with a B permit (O.R. 1.74\*\*) or other permit (short-term permit (L permit) in general, O.R. 8.46\*\*\*).

As far as the level of education is concerned, one can observe a lower risk of not using the acquired skills (O.R. 0.67\*\*\*) and a higher improvement of the work conditions (O.R. 1.28\*) for tertiary-educated migrants compared with others. Those who studied in Switzerland present a slightly higher level of unemployment (O.R. 1.85\*) and a low level of underuse of skills (O.R. 0.49\*\*\*), confirming the results of Akresh (2006) for the United States. The presence of a relative in Switzerland prior to the migration and being a parent do not significantly affect the indicators. Finally, those having received support for the move declare a higher level of satisfaction (O.R. 1.47\*\*\*) compared with those not having had such support. In general, all of the results are in line with what would be expected.

#### **5.5 Conclusions**

This paper aims at providing a better understanding of the extent to which the reason to migrate affects current migrants' situation on the labour market (the so-called structural integration), as measured by different indicators. Based on the Migration-Mobility Survey, this study assesses the relationship between reason to migrate and integration. Among the 6000 respondents 3210 are active on the labour market. The question referring to the reason to migrate, which is included in the survey, also provides the self-declared rather than the administrative reason, the latter being the information that is generally available in national statistics.

Descriptive results show that gender and the existence of a work contract or job offer prior to migration significantly influence the reason to migrate. Compared with men, women arrive more frequently in Switzerland for family reasons, often after the arrival of their partner, and generally without a work contract. Conversely, most male migrants move to Switzerland for professional reasons, and almost half already have a work contract prior to arrival, meaning that for one of two male migrants, the move to Switzerland is prepared and anticipated. The existence of a contract is most frequent among highly educated migrants and among those coming from neighbouring countries and is least frequent among West African and South American migrants. One of the results of our analyses is the diversity of situation according to origin, particularly differences between migrants from Southern Europe and Western Europe and between migrants from the Southern Continents (West Africa and South America) and from industrialized countries.

Overall, migrants consider themselves relatively well integrated in the labour market: in particular, three-quarters of the men and almost two-thirds of the women declare an improvement in their work conditions after migration, their unemployment rate is low, fewer than 20% of migrants have short-term contract or work without a contract, and approximately 60% use their skills at work. Such results are not surprising given the favourable state of the Swiss labour market and the characteristics of newcomers. The results are in line with those presented by the OECD and European Union (2015) for comparable indicators and translate to a rather good working situation for migrants in Switzerland compared with other countries. However, after considering confounding factors using a logistic regression, one can observe differences in the level of integration according to the reason for migration. Not only family migrants but also migrants who arrived in Switzerland for other reasons present a lower probability of job satisfaction and a higher risk of underuse of skills compared with those arriving with a work contract. Significant differences are also observed between professional migrants with or without a work contract, the latter facing more difficulties, according to all of our indicators (in particular concerning the use of skills and the probability to be unemployed), with integrating into the labour market in a good position.

Results are in line with those of the rare studies mentioned before, particularly in North America (CIC 1998), which have explored the relationship between reasons for migrating and labour force outcomes. The results contradict the few studies that specifically use professional wages as an indicator of integration (Nekby 2002). However, due to the lack of data, we were not able to investigate income as an indicator of structural integration. Overall, the results confirm that tied or family migrants, often the woman in a partnership, face more difficulties than men do with entering the labour market and obtaining a job corresponding to their skills and that satisfies them.

The subjective evaluation of migrants' position in the labour market, particularly in terms of the level of satisfaction and improvement in their conditions following migration, depends upon both their current position and that preceding migration. Some groups of migrants, particularly those from Portugal, do not use all their skills at work, which can diminish their remuneration and lead to frustrations; however, they declare an improvement in their situation. This paradox can be explained by the poor working conditions before migration in the period following the 2008 financial crisis. The absence of professional prospects in the country of origin can lead them to positively evaluate their current situation, although this evaluation would not always objectively be positive. Reflecting more widely, one can suggest that the self-estimated situation on the labour market post-migration also depends upon previous experiences and not only upon the current situation.

Results obtained can contribute in at least two ways to the debate on migration and integration policies, at least in Switzerland if not also in European countries.

First, results highlight the importance of the motive for migration in influencing the position migrants have in the labour force. Those who have migrated for family reasons, particularly women, and those arriving from non-European countries are particularly concerned by lower work conditions. Such a result justifies specific integration policies for non-professional migrants, tied migrants9 in particular. It also raises concerns regarding how women can be better integrated into the labour market. The difficulty for migrants to integrate into the labour market in a position that corresponds to their skills not only leads to a waste of competences in the labour market (brain waste) but also can result in personal dissatisfaction with the move to the recipient country. Further analyses, which are not presented here, show that family migrants, particularly women, declare a lower level of satisfaction concerning the move compared with migrants who have moved for professional reasons.

Second, the analyses conducted among family migrants demonstrate the importance of gender in professional outcomes. The results clearly demonstrate that when the family migrant is a man, he is more frequently able to obtain a professional position leading to an improvement in his working conditions (compared with his situation before migration) compared with a woman. This result clearly raises the issue of gender equality in the labour market.

#### **References**


Akresh, I. R. (2008). Occupational trajectories of legal US immigrants: Downgrading and recovery. *Population and Development Review, 34*(3), 435–456. https://doi. org/10.1111/j.1728-4457.2008.00231.x.

<sup>9</sup> In the largest cities of Switzerland, initiatives already exist to try to integrate tied, family or female migrants into the labour market. Generally, those initiatives are private, occasionally with public subsidies and focussed on highly skilled tied migrants.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Chapter 6 Employment Trajectories of Recent Immigrants in Switzerland**

**Elena Vidal-Coso**

#### **6.1 Introduction**

The employment and socioeconomic integration of immigrants in destination countries has attracted considerable public and policy concern. Understanding how successfully immigrants from different origins integrate in host labour markets and the specificities of employment and occupational attainment of female and male migrants are relevant issues for further policy implementations. Indeed, academic interest has broadly focussed on immigrants' employment status and the types of job they hold in comparison with native workers. Likewise, scientific investigation has attempted to discern the effect of residence length in the convergence of immigrants' labour market performance compared with that of the native-born (Chiswick 1978; Chiswick et al. 2005; Akresh 2006, 2008). However, although a highly relevant issue, studies on labour mobility of immigrants between their home country and host country are scarce due to the longitudinal data requirements for such analysis, which is only available for some countries (some interesting studies are Akresh 2006, 2008 for the United States of America; Chiswick et al. 2005 for Australia; Bauer and Zimmerman 1999 for Germany; Rooth and Ekberg 2006 for Sweden; and Stanek and Veira 2009; Aysa-Lastra and Cachón 2013; Simón et al. 2014; Vidal-Coso and Miret-Gamundi 2014 for Spain.)

Employing the longitudinal information from the new data on migration in Switzerland, the Migration-Mobility Survey, this chapter focusses on immigrants' trajectories with respect to their employment status at the time of migration and during the process of settling in Switzerland. Moreover, the analysis discerns how

E. Vidal-Coso (\*)

Institute of Demography and Socioeconomics IDESO, University of Geneva, Geneva, Switzerland

nccr – on the move, Neuchâtel, Switzerland e-mail: Elena.Vidal@unige.ch

© The Author(s) 2019 123

I. Steiner, P. Wanner (eds.), *Migrants and Expats: The Swiss Migration and Mobility Nexus*, IMISCOE Research Series, https://doi.org/10.1007/978-3-030-05671-1\_6

the integration into the host labour market varies across origins and gender. This survey is the first in the country containing retrospective information on immigrants' employment status before migration together with their first and their current employment status in Switzerland. The richness of the information contained in the survey allows searching for the main explanatory factors with respect to immigrants' participation in the labour market at different points in their migratory experience.

The investigation is particularly interested in analysing whether the observed inequalities by origin in the labour market integration of immigrants after arrival and during settlement in Switzerland result from differences in terms of human capital across origins. However, if in contrast labour market disadvantages exist and persist over time for those migrants more distant linguistically and culturally, regardless of their human capital characteristics and previous experience in the labour market, this situation would confirm the segmentation and discrimination postulates. This chapter also aims to investigate whether there are differences between the genders in immigrants' labour market attainment. In particular, the availability of information in relation to the migration process allows us to estimate whether post-migration labour performance of women is more determined by family characteristics (partnership, couple's migratory trajectory and presence of children) and reasons for migration (i.e., professional or family-oriented migration) than that of men because female-tied migration is associated with lower labour attachment due to the persistence of the preference for men's careers (Boyle et al. 2001, 2009).

Switzerland constitutes the perfect case study for achieving our research interests because of its unique structural and institutional characteristics. On the one hand, Swiss migration policy has reoriented towards a selective model favoring the highly skilled citizens in an increasingly globalized scenario. Some authors (Afonso 2004; Favre 2011; Wanner 2004) identified, at the beginning of the 2000s, a shift in the occupational profile of the foreign labour force in Switzerland. They pointed out the overrepresentation at the top of the occupational structure (Favre 2011) of the recently arrived highly skilled immigrants, in speciall of those from the European Union (EU) and North America (Golder 2001; Laganà 2013; Liebig et al. 2012; Wanner et al. 2002; Widmer 2005). Nevertheless, they also maintain that the entrance of low-skilled immigrants has not concluded and that migration is dual in terms of human capital, with a concentration of new immigrants at the bottom of the occupational structure. Empirical evidence pointing to disadvantages concerning the labour performance of specific groups of immigrants in Switzerland is abundant, linked not only to their lower educational level but also to discriminatory practices and to the segmented nature of the Swiss labour market (Ebner and Helbling 2016; Fibbi et al. 2003; Widmer 2005; Vidal-Coso and Ortega-Rivera 2017). On the other hand, some of these studies also found gender-specific differences in labour and occupational outcomes among immigrants, with women presenting less favourable labour attainment than men did. Whereas the dynamism of the Swiss labour market would act to facilitate the labour insertion of men and women partners, public attitudes towards employment remain highly gendered in the country (Levy and Widmer 2013; Stadelmann-Steffen 2007). Among OECD

countries, Switzerland has one of the highest participation rates of its female population but at the same time one of the lowest female full-time employment rates (Dutu 2014). The scarce and expensive provision of external childcare encourages mothers either to withdraw from employment or to work part-time (Stadelmann-Steffen 2007). Deficiencies in conciliatory measures reinforce inequalities across socio-economic groups, which can be even more crucial for immigrant families in the absence of a family network. Moreover, previous analysis demonstrates the importance of different job opportunity structures in explaining the heterogeneity of how women in Switzerland adapt their labour supply to their family circumstances depending upon their national origin (Vidal-Coso 2018). To summarize, due to the country's economic prosperity, job opportunities and wage standards, immigrants are expected to successfully integrate into the Swiss labour market. Thus, this study provides the first opportunity to explain the degree of success of the integration of immigrants into the Swiss labour market considering their migration motivations, their family circumstances during their mobility into Switzerland, and their human capital and labour situation in origin and destination.

The chapter tries to enlarge the empirical evidence of the Migration-Mobility Nexus. Although the research is based in the paradigm of immigrants' integration into the host society through patterns of labour market participation, it aims to emphasize the notion of migration as a mobility process and of the individuals' trajectories within the context of an international division of labour and human capital in a contemporary globalized migratory scenario. Indeed, Switzerland constitutes a perfect case for analysing this Migration-Mobility Nexus because the country is characterized by a dual regime of migration. The chapter aims to understand to what extent this dual regime affects mechanisms of inclusion and exclusion of migrants in labour markets.

This chapter is structured as follows. Section 6.2 reviews the theoretical framework of the labour mobility of migrants. The data, the variables and the sample used are described in Sect. 6.3. Sections 6.4 and 6.5 present the research results. Finally, the chapter concludes with a discussion Sect. (6.6).

#### **6.2 Theoretical Background and Research Objectives**

The labour market mobility of immigrants has been widely analysed from different theoretical perspectives. First, the *functionalism* and *neoclassical* traditions focus on the importance of *human capital* and *time of residence* as key determinants of the differences of individuals in labour market attainment (Borjas 1994; Chiswick 1978). This perspective maintains that, upon their arrival in the host country, immigrants suffer a disadvantage relative to natives that can affect different aspects of their labour market incorporation, such as employment, wages and devaluation in the occupational category. This disadvantage has been attributed to the difficulties of immigrants in transferring formal education, employment experience, and training obtained abroad (Chiswick and Miller 2009; Clark and Drinkwater 2008).

Moreover, some authors have highlighted the lack of country-specific skills on arrival as a key factor explaining differences in economic success. Limited knowledge about the functioning of the labour market or a lack of fluency in the host country's language might represent an obstacle to finding better job opportunities for immigrants (Chiswick and Miller 2002; Dustman and Fabri 2003; Clark and Drinkwater 2008). A key factor in determining new immigrants' labour performance is the extent to which their education, pre-migration labour market experience, and training obtained abroad are valued in the destination country (Blackaby et al. 2002; Clark and Drinkwater 2008; Kanas and Van Tubergen 2009). Moreover, the level of transferability of human capital across countries depends upon the economic and cultural proximity between country of origin and country of destination (Akresh 2006). From this perspective, the assimilation process entails that the initial disadvantage should decline over time. As immigrants settle into the receiving country, they adapt their skills to the requirements of the destination labour market, improve their knowledge of the host country's language, and acquire local education and training. These adjustments can eventually improve their employment prospects. Consequently, a U-shaped pattern of employment and occupational mobility for immigrants is expected, with an occupational downgrading and higher risk of unemployment upon arrival and a recovery in employment probabilities and conditions during settlement in the host country (Chiswick et al. 2005).

The second theoretical framework considered in this investigation is the *structural or dual labour market* theory (Piore 1979; Thurow 1975), which offers an alternative explanation of immigrants' employment after migration. According to this view, labour markets are divided into primary and secondary sectors. The primary sector offers stable jobs, relatively high wages, and better working conditions. In contrast, the secondary sector is characterized by unstable jobs with low salaries and worse job conditions. This perspective predicts greater employment instability and precariousness and consequently higher risk of unemployment in the secondary segment and little intersegment mobility, particularly upward mobility, from secondary to primary segment occupations. Furthermore, Piore (1979) and Kalleberg and Sorensen (1979) identified the segmentation of the labour force by migrant status or national origin of workers. Various researchers, such as Heath and Cheung (2007), refer to an "ethnic penalty" or the process in which immigrants from a particular national or ethnic origin experience some type of disadvantage to their labour performance. This situation involves a greater risk of unemployment and of a limitation on their labour prospects to access the secondary segment of the labour market compared with natives with similar human capital and sociodemographic profiles (Rooth and Ekberg 2006 for the Swedish context). This approach maintains that employment disadvantage and occupational downgrading of some groups of migrants is expected to persist over time, regardless of the duration of the residence and the adaptation of skills to the host country's labour market requirements. In a very similar line of argumentation, such researchers state that cultural or social differences between individuals from different cultures can cause employers to prefer applicants from their own culture or with higher affinity (Ebner and Helbling 2016). To cope with this problem, employers rely on the observable characteristics of applicants, such as gender, age, national origin and ethnic group, to infer their expected productivity, which should correlate with the perceived average productivity of the group. A case study for the Swiss case can be found in Fibbi et al. (2006).

The third theoretical perspective considered is *social capital* theory (Palloni et al. 2001; Massey et al. 1993). The starting hypothesis is that once someone in a person's networks migrates, the ties of friendship and kinship are transformed into a resource for gaining access to employment at the destination, particularly in those considered "immigrant jobs" (Massey et al. 1994) because social and family networks make a migrant's economic requirements less urgent, facilitating an optimization in the job search. However, Portes and Sensenbrenner (1993) noted that social networks might have negative consequences for migrants and that the same mechanism that gives rise to labour insertion at a destination can also constrain occupational opportunities and labour mobility to certain types of jobs. In this sense, Mahuteau and Junankar (2008) confirm the negative effect of social networks leading to poorer occupational labour outcomes.

Finally, the labour performance of migrants has largely been assumed to vary with individuals' family circumstances and reasons to migrate. Accordingly, migrants who migrate to join their relatives at the destination are commonly expected to decline in their occupational attainment in comparison to employmentmotivated migrants. Mincer (1978) analysing "Family Migration Decisions", stated that the migration of "tied" partners was motivated by a desire to maximize family income rather than by individual professional interests. This motivation would place them in a less favourable position in the host labour market. The author determined that tied movers were primarily women and that those women were more likely to be unemployed or out of the labour force and have lower earnings than were otherwise comparable immigrant women who were primary movers. As a matter of fact, it was usually the husband who stimulated employment-motivated family migration and who had the most to gain from the family relocation (Boyle et al. 1999). Currently, the massive entrance of young women into the domains of higher education and professional careers, together with the reversal in the gender gap in education (Esteve et al. 2016), explains the emergence of dual-income couples, in particular among people with a university degree. Accordingly, the "trailing spouse" resulting from migration can no longer be assumed, particularly among highly skilled migrants. However, although differences between men and women in terms of labour market outcomes after migration have decreased over time, they have not completely disappeared. In contrast with the commonly held view that highly skilled couples are very egalitarian, arrangements consolidating gender inequality following family migration might be present among skilled in addition to among unskilled migrants. Indeed, the persistence of the preference for men's careers (Boyle et al. 2001) in family migration decision making makes it necessary to estimate whether the post-migration labour performance of women is more determined by the reasons for migration than is that of men.

This chapter analyses immigrants' labour market trajectories in Switzerland considering their employment situation in the country of origin and throughout their settlement in the country of destination. The research has three specific objectives. First, it provides new insight to understand the dynamics of immigrants' insertion in the host labour market, focussing on their employment status at the time of migration and during the process of settlement in Switzerland. In doing so, the analysis aims to search for differences by birthplace and gender in the individual's probability of being employed in relation to being unemployed or inactive. Furthermore, for those employed at the time of the survey, the analysis estimates their probability of working part-time.

Second, this research is particularly interested in discerning whether the observed differences in terms of employment status result from the differences among the considered groups of immigrants in terms of skill composition and of the degree of assimilation of their human capital to the requirements of the Swiss labour market. However, labour disadvantages existing and persisting over time for some immigrant groups, regardless of their human capital characteristics and level of assimilation to Switzerland, would confirm segmentation and discrimination postulates.

Third, professional and family reasons for migration are not necessarily incompatible. Thus, this research will assess to what extent the probability of unemployment, economic inactivity and part-time employment is higher for tied or family-motivated migrants. Or if on the contrary, family-motivated mobility following the primary migrant partner helps the migrant to successfully integrate into the host labour market, due to less urgent economic requirements and a greater ability to be selective about the types of jobs he/she takes. Moreover, the research aims to demonstrate whether gender dispositions are essential for explaining family migration labour market outcomes. If they are essential, then women's labour market participation would be more determined by their family circumstances and the reason for migration than would be that of men. This gender perspective is indispensable for analysing the Swiss case. In effect, whereas the dynamism of the Swiss labour market should facilitate the labour insertion of both partners, the gendered patterns of the female labour supply might generate or reinforce gendered employment arrangements among migrant couples.

#### **6.3 Data and Methods**

The data source used in the empirical analysis is the Swiss Migration-Mobility Survey. From the original sample, this research focusses on the following nine groups: Germany/Austria, France, Italy/Spain, Portugal, UK/North America, India, South America, Africa and Asia. This grouping yields a final sample selection of 5823 immigrants – 3118 immigrant men and 2705 immigrant women. The data is weighted for descriptive and multivariate analysis. This data source is the best option for the analytical purposes of this investigation because it includes retrospective information about employment trajectories from the period prior to migration. It includes information on the immigrant's employment status at three crucial points: before migration to Switzerland, immediately upon arrival, and at the time of data collection. This information allows analysis of the dynamics in terms of employment status at the time of migration and during the process of settling in Switzerland. Nonetheless, the Migration-Mobility Survey is composed of single cross-sectional data. As Chiswick et al. (2005) noted, by using cross-sectional data, we could incur a bias in the longitudinal effect that immigrants experience in their labour market achievements. This bias could be a consequence of changes over time in the quality of immigrant cohorts (Borjas 1985), selectivity in return migration, third-country migration or abandonment of the labour market (Constant and Massey 2003) or changing economic cycles (Aslund and Rooth 2007). However, in this research, migrants only arriving recently, between 2006 and 2016, are considered. More specifically, immigrants considered in this analysis have a median duration of residence of 5 years in Switzerland. Consequently, the recent nature of the immigration included in the analysis minimizes the effect of any possible bias.

Using a multinomial probit model, we analyse differentials by birthplace and gender in terms of labour market inclusion, both immediately after their arrival in Switzerland and at the time of data collection. Multinomial probit regression is an extension of probit regression that is applied to categorical variables with more than two categories. We estimate the likelihood of immigrants to be unemployed and to be inactive relative to being employed (which is the baseline category). Thus, the multinomial model is appropriate here, given the unranked nature of the outcome. Moreover, binary probit regression analysis is applied to investigate the immigrant's probability of working part-time for those individuals employed at the time of the interview. Beta coefficients, the standard errors and the level of signification are specified in the models. Predictive margins from the binary and multinomial models are displayed for interactions between gender and nationality and between gender and reasons for migration.

With respect to explicative human capital variables, we include the level of education (primary or lower, secondary and tertiary) and previous occupation status in the origin country. To control the potential effects of language skills on labour market performance, we include a variable that identifies whether the main language of the individual corresponds to any of the Swiss languages, English or to another foreign language. When models focus on the immigrants' employment status at the time of the survey, whether the educational level was validated in Switzerland and whether the individual is proficient in the local language are also considered.

Reason for migration with its interaction with gender is included in the models. This variable is coded in four categories: professionally motivated migration, family-motivated migration, professionally and family-motivated migration (including immigrants who declared both reasons of migration, professional and family) and migration motivated by other reasons (e.g., lifestyle). Likewise, age (continuous and squared) and retrospective information concerning family characteristics (children and partnership) are also considered in all models. Social network at arrival is included as a covariate in the analysis of employment status at arrival. When the analyses relate to the moment of data collection, i.e., current employment status, controls for years residing in the country (continuous) are included.

#### **6.4 Descriptive Findings**

Table 6.1 displays the employment status of immigrant men and women by country of birth in three different moments of their migratory experience: before migration, immediately after their arrival in Switzerland, and at the time of the survey. According to the observed trajectories in terms of labour market situation, recent male migration successfully assimilates into the Swiss labour market after settlement in the host country. Indeed, although men present, after arrival in Switzerland, lower levels of full-time employment and higher percentages of unemployment in comparison to their labour situation prior to migration, this less favourable labour performance is a temporary process of adjustment to the host country. Effectively, in comparison with their situation in their countries of origin, the majority of male groups present, at the time of the survey and after having resided in the host country, higher employment percentages, particularly of full-time employment, and lower unemployment percentages. Consequently, accordingly to the assimilation postulates, settlement in Switzerland results in a general gain in the employment prospects for immigrant men.

However, heterogeneity across origins in post-migration labour market integration could also be observed among immigrant men. On the one hand, men from Germany, Austria, France, Italy, Spain, Portugal, and Asia are the most successful in their employment incorporation in Switzerland. The initial devaluation in employment status is minimal for them, and the gain in terms of employment is very steady for these groups of migrants after residing in Switzerland. On the other hand, lower percentages of full-time employment and higher unemployment among immigrant men from South America and Africa indicated more-disadvantaged labour insertion for them. Finally, other reasons for labour market inactivity are residual among male immigrants. To summarize, although integration into the Swiss labour market has been very positive in general for recent male migration, this statement is not true for those born in Africa and South America. Indeed, their migration to Switzerland does not improve the disadvantaged employment status they presented before migration.

The employment trajectories of immigrant women during their migratory process to Switzerland follow a different pattern; mobility into Switzerland has caused a reduction of their labour market attachment. Indeed, with the exception of women from Germany/Austria and Asia, percentages of full-time employment among immigrant women are markedly reduced after migration in comparison to the




132

**Table 6.1**


Germany/

Austria

France

Spain

Portugal

America

India

America

Africa Asia

Italy/

UK/North

South

Source: Migration-Mobility Survey 2016. Weighted results (normalized weights) situation in origin. Moreover, a notorious rise in the share of part-time employment during their migratory process is observed, particularly among women from Germany/Austria, France, Italy/Spain, Portugal and South America. Unemployment percentages have also increased among female immigrants from UK/North America, India, South America and Africa. Finally, the increase in the post-migration percentages of women who care for the home/family, particularly of non-European women, is also notorious. Furthermore, the migration to Switzerland has increased the existing differences before the mobility process in terms of participation in the labour market between the genders. Effectively, the labour participation of women in their countries of origin was already lower than was that of their male counterparts because women presented lower percentages of full-time employment, higher percentages of part-time employment and a higher share of inactivity for family reasons. These existing differences by gender were accentuated after migration to Switzerland. Whereas employment prospects after settling in Switzerland improved in general for immigrant men and even exceed the employment percentages in the country of origin, the recovery in labour market participation after residing in the country is not so evident for immigrant women, at least in terms of full-time employment.

Table 6.2 displays the employment and occupational characteristics of immigrant men and women at the time of data collection. From this table, it is interesting to highlight the highly skilled occupational profile of male immigrants from the German-speaking countries, France, UK/North America and India. Table 6.2 also shows that all employed immigrants, men or women, and independently of their origin declare that their professional situation has improved in relation to that in the country of origin at the time prior to migration. However, although the feeling of professional progression after moving to Switzerland is systematic among all immigrants, some heterogeneity could be observed across origins. Male migrants from Germany and Austria, France and UK/North America express lower levels of improvement and a higher level of status quo. In fact, these groups are highly educated, many of them with director and management positions. Thus, one possible interpretation of the results is that these migrants were characterized by relatively good conditions prior to the migration and, consequently, the differences between the evaluation of the situation in the home country and in Switzerland are smaller. In contrast, despite their relatively low insertion in terms of employment, occupational category and higher overqualification, immigrants from India, South America and Africa, but also from Italy, Spain, Portugal, and Asia express very high levels of improvement in their professional situation. Gender-specific patterns among employed immigrants also appear in Table 6.2. For example, there are higher proportions of directors and managers among employed men than among employed women, and there are major levels of overqualification among the latter.


Occupational characteristics of immigrants at the time of interview by origin and gender (in %)

6 Employment Trajectories of Recent Immigrants in Switzerland


136


Germany/

Italy/

UK/North

South

Source: Migration-Mobility Survey 2016. Weighted results (normalized weights)

#### 6 Employment Trajectories of Recent Immigrants in Switzerland

#### **6.5 Multivariable Results**

#### *6.5.1 Employment Status Immediately After Migration to Switzerland*

The multinomial probit models in Table 6.3 analyse the main determinants of the employment status of male and female immigrants immediately after their arrival in Switzerland. The models provide the likelihood of being unemployed or being inactive, rather than being employed, the reference category. The predicted probabilities for men and women by their birthplace and reason for migration are displayed in Figs. 6.1 and 6.2.

The primary conclusion from Table 6.3 is that, compared with German and Austrian immigrants, immigrants from South America and Africa had the lowest probability of being employed after arrival to Switzerland because coefficients for these groups indicated their higher probabilities of unemployment (coefficients β 0.33 and 0.86, respectively) and inactivity (coefficients β 0.52 and 0.62). In the same vein, men from Italy/Spain, Portugal and Asia present a relatively higher likelihood of unemployment, and those born in the UK and North America higher probabilities of inactivity, although differences between them and the reference group of immigrants from Germany and Austria are not large. In contrast, immigrants arrived from France and India present relatively lower risks than German/Austrians do of being unemployed or inactive. Does this disadvantage in employment probabilities for some groups disappear when controlling educational level, primary language, occupation status in origin and the remaining socio-demographic characteristics? The results in successive models indicate that the differences across groups remain almost unaltered, although the highest unemployment probability for the Portuguese and the inactivity of UK/North American immigrants became nonsignificant. Effectively, although Table 6.3 predicts that the differences across groups decrease after having introduced the control variables, the mentioned differences among immigrants in terms of employment status remain visible (Fig. 6.1). Moreover, the employment probabilities increase for all immigrant men, many of them presenting high percentages of employment (approximately 80% and reaching 90% for French male immigrants). In contrast, men from Africa and South America continue to present the lowest employment levels. Indeed, African male unemployment rises to 20%, and the unemployment rate is 15% for men from South America.

Women are more likely to be unemployed and inactive than men are, and results for the interaction term show that the unemployment is particularly high for women from the UK and North America (coefficient β 0.71) and from India (1.67). In relation to female immigrants from Germany and Austria, the reference group, the remaining immigrant women present a higher probability of inactivity, whereas differences for French and Asian women are no longer significant. For women, Fig. 6.1 clearly indicates that inequalities in employment status across groups are more pronounced than in the case of men, with very low expected probabilities of






Employment status at arrival

Unemployment

Model 1

B

SE

Sig. B

SE

Sig. B

SE

Sig. B

SE

Sig. B

SE

Sig. B

SE

Sig.

Model 2

Model 3

Inactive

Model 1

Model 2

Model 3

Source: Migration-Mobility Survey 2016. Weighted results (normalized weights) Statistical significance: ns = non-significant; \* *p* < 0.10; \*\* *p* < 0.05; \*\*\* *p* < 0.01

**Fig. 6.1** Predictive margins of employment status immediately after migration to Switzerland by origin and gender

Note: *Obtained from multinomial models in* Table 6.3

Source: Migration-Mobility Survey 2016. Weighted results (normalized weights)

employment for women from India or South America, who present higher levels of unemployment and inactivity. Women from Spain/Italy, UK/North America and Africa also present low levels of employment (approximately 50%) in comparison with the highest employment levels of women from Germany/Austria, France,

**Fig. 6.2** Predictive margins of employment status immediately after migration to Switzerland by reason for migration and gender

Note: *Obtained from multinomial models in* Table 6.3

Source: Migration-Mobility Survey 2016. Weighted results (normalized weights)

Portugal and Asia. Once controls are introduced, the differences by origin are only slightly diminished, and the female employment levels remain strongly heterogeneous across groups, with French women presenting a twice as high percentage of employment as do Indian women.

With respect to the effect of the human capital variables, a higher probability of unemployment and inactivity is observed among secondary educated immigrants, whereas differences between tertiary and primary or lower educated are not significant. Moreover, the results predict better employment probabilities for those who speak English or one of the Swiss languages in comparison to those immigrants who declare another foreign language as their primary language. With respect to the occupation status before migrating to Switzerland, individuals who in their country of origin were directors or managers are more likely to be employed after arrival. Furthermore, as expected, individuals who were not employed before migration present a lower probability of being employed once in Switzerland. This result corroborates that employment status in the country of origin is very likely to continue immediately after arrival (the coefficient β for unemployment is 2.03 for previously unemployed, and the coefficient for inactivity is 1.88 for previous students or in training and 2.29 for inactive from the country of origin). Age at arrival is not significant, whereas having children before migration is associated with higher risk of inactivity. This result is consistent with González-Ferrer (2011), who argues that the temporal sequence of migration and key family lifecycle events might help us in explaining the post-migration employment patterns of migrants.

Moreover, immigrants already having a social network in Switzerland on arrival are more likely to be unemployed or inactive. This point is also true for those immigrants arriving together or after their partners, who are more likely to be seeking a job or inactive. These results are in line with the social networks postulates that having social or family network at the destination is related to having lessurgent economic requirements to find a job after migration and eventually optimize their occupational outcomes. These results could also explain the higher risk of unemployment and inactivity for family-motivated immigrants. However, the fact that inactivity levels are higher when the tied migrant is a woman might also be indicative that gender dispositions remain to be determinant for explaining labour market outcomes after migration. In effect, as seen in Fig. 6.2, post-migration labour insertion for women is more linked with reasons for migration than it is for men.

#### *6.5.2 Employment Status After Settling in Switzerland*

Table 6.4 displays the models analysing the employment status of immigrant men and women at the time of the survey. This analysis allows investigating differences by origin in the labour integration considering settling in the Swiss labour market and society, even when the focus is on immigrants relatively recently arrived. Model 1 again indicates heterogeneity in the immigrants' employment status based on their origin. Effectively, β coefficients for Africans (1.04), South Americans (0.58), Portuguese (0.40), UK/North Americans (0.33) and Asian (0.21) indicated relative higher probabilities of unemployment for them, compared to the reference group. Apart from the Portuguese, these groups also have a higher likelihood of inactivity. In contrast, Italians and Spaniards have improved their relative position in terms of employment status in comparison to their status immediately after arrival in Switzerland (Fig. 6.3). Once we annulled the differences across groups in terms of human capital and occupational status at origin, differences in the probabilities of unemployment in Model 2 only remain significant for Africans. Moreover, this immigrant group presents higher risk of unemployment, as coefficients indicate in Model 3. Likelihood of inactivity remains significantly higher for immigrants from UK/North America (0.29) and from Africa (0.11), although for this last group, differences turn not significant in Model 3. Moreover, the assimilation assumption of a more successful labour insertion for all groups after settlement in the host country appears to be corroborated by higher employment percentages compared with those observed for them after arrival in the host country. This point is consistent with the significant effect of years residing in Switzerland in reducing the individual's probability of unemployment and inactivity. To summarize, the percentages of employed immigrants displayed in Fig. 6.3 are higher in comparison to those observed after the arrival in Switzerland. Likewise, the differences across origins in terms of employment status are less significant at the time of the interview than are those observed at the beginning of the migratory experience in Switzerland (Fig. 6.1).

Women are more likely to be economically inactive, and although gender does not affect unemployment likelihood, significantly greater risks are observed in the interaction term for some groups. Expected probabilities displayed in Fig. 6.3 cor-




**Table 6.4** (continued)


Current employment status

Unemployment

Model 1

B

SE

Sig. B

SE

Sig. B

SE

Sig. B

SE

Sig. B

SE

Sig. B

SE

Sig.

Model 2

Model 3

Inactive

Model 1

Model 2

Model 3

Source: Migration-Mobility Survey 2016. Weighted results (normalized weights) Statistical Significance: ns = non-significant; \* *p* < 0.10; \*\* *p* < 0.05; \*\*\* *p* < 0.01

**Fig. 6.3** Predictive margins of employment status of immigrants at the time of the interview by origin and gender

Note: Obtained from multinomial models in Table 6.4

Source: Migration-Mobility Survey 2016. Weighted results (normalized weights)

roborate these findings. Effectively, although immigrant women generally improved their chances of being employed when their employment levels are compared with those after arrival, some of them continue to present high percentages of unemployment and inactivity at the time of data collection. This point is particularly true for

Indian women but is also true for immigrant women from the UK/North America, South America, Africa and Asia. In contrast, employment margins for female immigrants from German-speaking countries, France, Italy, Spain and Portugal indicate a high level of employment insertion in the Swiss labour market.

With respect to the effect of the human capital variables, not-significant differences are found by individual educational level. Moreover, the main language or proficiency in the local language is not associated with unemployment, but Englishspeaking migrants and those not proficient in the host local language are more likely to be inactive than employed. However, a clear effect of educational validation on successful assimilation into the host labour market is found. Finally, unemployment and inactivity risk remain higher after settling in the country for those who were not-employed before migration.

With respect to the other demographic and migratory characteristics, we can mention that age is not significant for unemployment, although it reduces the probability of being inactive. Partnership is not significant for the probability of unemployment but migrating together with the partner is associated with a higher likelihood of inactivity, as is having children. Finally, as was observed in the labour market participation immediately after arrival, family-motivated immigrants, particularly women, present a higher risk of unemployment and inactivity. Figure 6.4 shows that gender differences in employment status related to the reason for migration increased in comparison with those observed in Fig. 6.2 for their status after arrival. This result might indicate that motivation for entrance for men has a temporary effect on labour participation in the host society, whereas for women, it has deeper consequences for employment prospects that remain after settlement in the country.

**Fig. 6.4** Predictive margins of employment status of immigrants at the time of the interview by reason for migration and gender

Note: Obtained from multinomial models in Table 6.4

Source: Migration-Mobility Survey 2016. Weighted results (normalized weights)

Table 6.5 and Fig. 6.5 shows that, independently of their birthplace, immigrant women are more likely than immigrant men are to work part-time at the time of data collection. Moreover, although differences by gender have decreased after covariates were introduced in successive models, the female likelihood of part-time employment remains significantly higher than that of men. Indeed, Fig. 6.5 shows that whereas the expected percentage of part-time employment among immigrant men is situated at approximately 10%, female part-time rises to 30% for many of the origins considered. Furthermore, heterogeneity could be observed across origins even when the differences among groups decrease after considering their human capital, employment characteristics in the country of origin and at the time of the survey, and their family and migratory characteristics. Effectively, a higher likelihood of part-time employment is observed for men and women from South America and Africa and for women from Portugal. In contrast, immigrants from India and France are less likely to work part-time.

Likelihood of part-time employment is negatively correlated with educational level. In the same vein, immigrants who did not validate their education obtained abroad, overqualified employees and those without a job offer before migration present a higher probability of working part-time. Swiss language speakers are the most likely to be working part-time, although proficiency in a local language is not associated with the outcome. As expected, immigrants in part-time employment or those looking after their families before migration present higher likelihood of parttime employment. In relation to those holding director or manager roles at the time of the interview, self-employed and unskilled employees are the most likely to work part-time. Furthermore, age and years of residence in Switzerland are not significant in predicting part-time employment.

Finally, whereas partnership and the couple's migratory process are not associated with our outcome, having children is a strong determinant of part-time employment for women, particularly for mothers who had their children after migration (Fig. 6.6). In contrast, differences are not observed between fathers and childless men in their probability in terms of working part-time. Finally, likelihood of parttime employment is higher among family-motivated migrants than among profession-related migrants. Moreover, gendered patterns also appear in relation to the effect of the motivation for migration on part-time employment in Switzerland. Indeed, Fig. 6.7 clearly shows that the likelihood of working part-time is more linked with reason for migration for women than for men.

#### **6.6 Conclusions and Discussion**

Despite the short time span involved in the length of residence of immigrants considered in this research, a median of 5 years since arrival, results are consistent with the assimilation hypothesis because they point to an improvement in terms of labour market insertion during the process of settlement in Switzerland. Nevertheless, significant differences have emerged by gender that broadly justify the decision to



E. Vidal-Coso


Model 1

Model 2

Model 3

Model 4

Statistical Significance: ns = non-significant; \* *p* < 0.10; \*\* *p* < 0.05; \*\*\* *p* < 0.01

**Fig. 6.5** Predictive margins of part-time employment of immigrants at the time of the interview by origin and gender

Note: Obtained from probit models in Table 6.5

Source: Migration-Mobility Survey 2016. Weighted results (normalized weights)

**Fig. 6.6** Predictive margins of part-time employment of immigrants at the time of the interview by children and gender

Note: Obtained from multinomial models in Table 6.5

Source: Migration-Mobility Survey 2016. Weighted results (normalized weights)

**Fig. 6.7** Predictive margins of part-time employment of immigrants at the time of the interview by reason for migration and gender, obtained from multinomial models Note: Obtained from multinomial models in Table 6.5

Source: Migration-Mobility Survey 2016. Weighted results (normalized weights)

include the gender perspective in the analysis and in the interpretation of the results. The first aim was to analyse dynamics in terms of employment status at the time of migration and during the process of settling in Switzerland by origin and gender. In the male case, results corroborate the influence of the length of residence in the country and the adaptation to the host labour market in reducing the individual's risk of unemployment or inactivity. Consequently, employment probabilities at the time of the survey are clearly higher for all groups of immigrants than they are immediately after migration. Moreover, the investigation was particularly interested in analysing whether the differences by origin in the immigrants' employment prospects in Switzerland resulted from the differences in terms of human capital composition among them. Or if, on the contrary, disadvantages in the labour market integration of some groups persist after the human capital differences are controlled for. In this respect, segmentation hypothesis could explain the heterogeneity in the level of employment of immigrants recently arrived, and the fact that differences by origin persist even after controlling for their human capital, their occupational status in the country of origin, and their socio-demographic and migratory covariates. Therefore, results indicate some degree of segmented labour insertion of recent migrants upon arrival in Switzerland. In contrast, no differences across groups were obtained in terms of employment status after settlement in the country, in line with human capital and assimilation postulates. Only African immigrants continue to present a significantly higher risk of unemployment, regardless of their education and their level of assimilation into the host labour market.

The evolution of the employment status of immigrant women also points to some degree of progression in their employment probabilities after settlement in the country; therefore, assimilation postulates are also corroborated for them. However, in comparison with men, they are, in general, more likely to be unemployed or inactive. Only women from Germany/Austria, France and Portugal present levels of employment similar to their male counterparts. Moreover, as results show, a meaningful proportion of these women are working part-time. In contrast, significant percentages of inactivity among some groups of female immigrants, such as among those from the UK/North America, India, South America, and Africa, clearly indicated lower post-migration participation in the labour force for these women. Furthermore, although differences in terms of employment levels among immigrant women decrease after controlling for their human capital characteristics, employment status in their countries of origin, family circumstances or motives for migration, heterogeneity across groups in terms of labour force participation remain in the female case.

Finally, the richness of the information in the data source in terms of the migration process allowed us to corroborate that post-migration employment is lower for tied migrants and family-motivated migrants. Even when the reason for migration has the same effect for men and women, margins obtained indicated that the employment prospects of women are more determined by the reason for migration than are those of men. Furthermore, based on the results obtained, family-motivated migration has only temporary effects on labour market integration of male migrants, whereas it harms employment prospects for women more permanently. In fact, women's inactivity levels and part-time employment remain very high after settlement in Switzerland.

The chapter yielded evidence of the Migration-Mobility Nexus because it considers migration and migrant labour market integration mobility processes that start in the society of origin and evolve during settlement in the host society. Although the study contributes to enhancing the understanding of the decisions immigrants make to attain a successful incorporation in the host country labour market, it is not without limitations. First, a limiting feature of this study is that the immigrants have resided for such a short period in Switzerland, a median of 5 years. Although biases due to emigration, death or abandonment of the labour market increase with the length of time considered, availability of longitudinal panel data following individuals through a longer period would provide a more accurate understanding of the immigrants' adjustment process. Moreover, although analysis corroborated that immigration to Switzerland is, in general, an advantageous experience in terms of employment prospects, further research is needed to better understand other aspects of immigrants' labour market integration. For example, descriptive analysis showed that less-advantaged immigrant groups in terms of human capital and employment status at origin are precisely those expressing higher levels of progression in their professional situation in Switzerland in comparison to that in the country of origin, despite their less favourable position in the Swiss occupational scale. Therefore, more-accurate analysis is necessary to corroborate the neoclassical argument that the subjective measure of professional success after migration is caused by differences in the rate of return to human capital between home and destination countries. Given the economic prosperity, job opportunities and high wage standards of Switzerland, the country would be a very special case to study. However, regardless of the richness of the information included in the Migration-Mobility Survey, one of its limitations is the absence of retrospective information in relation to immigrants' wages and positions within an international index of occupational scale. The availability of this information would facilitate a cross-country comparison and a quantification of immigrants' occupational and earnings trajectories. Finally, the prospect of immigrants returning to their source country or of permanent settlement in Switzerland is, in the author's opinion, determined not only by the success of their insertion in the host labour market but also by individuals' ability to be truly integrated into the host society. As stated in Vidal-Coso and Ortega-Rivera (2017), this point is particularly relevant for Switzerland, in which labour integration is facilitated by immigration policies, whereas societal and political integration remain restricted. As in many other countries, economic ground in Switzerland is strongly dependent upon not only attracting but also retaining skilled workers. In the same vein, challenges faced by women to combine family life and paid work should be carefully addressed.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Chapter 7 Does the Recognition of Foreign Credentials Decrease the Risk for Immigrants of Being Mismatched in Education or Skills?**

**Marco Pecoraro and Philippe Wanner**

#### **7.1 Introduction**

Migrants are generally found to earn less than natives do upon arrival in the host labour market (e.g., Chiswick 1978; Borjas 1994). Various explanations have been proposed, such as the fact that the quality and compatibility of foreign human capital might differ from those of local human capital (e.g., Friedberg 2000) or simply because of labour market discrimination (e.g., Lang and Lehmann 2012). Alternatively, because they have information on the labour markets in both the country of origin and the host country, they should be able to compare their expected labour market outcome in both countries (Borjas and Bratsberg 1994). Thus, they might be ready to accept a poorer situation than would their native counterparts in the host country but improve their labour market outcome when they expect a worst position in the country of origin. Migrants are then expected to work in a lower position than the position of natives with the same observed education and experience levels. Such a *downgrading* appears to be a widespread phenomenon in developed economies and means that immigrants with foreign-acquired skills are paid less than are natives for a comparable set of measured skills (Dustmann et al. 2016).

M. Pecoraro (\*)

nccr – on the move, Neuchâtel, Switzerland e-mail: Marco.Pecoraro@unine.ch

P. Wanner Institute of Demography and Socioeconomics IDESO, University of Geneva, Geneva, Switzerland

nccr – on the move, Neuchâtel, Switzerland e-mail: Philippe.Wanner@unige.ch

© The Author(s) 2019 161 I. Steiner, P. Wanner (eds.), *Migrants and Expats: The Swiss Migration and Mobility Nexus*, IMISCOE Research Series, https://doi.org/10.1007/978-3-030-05671-1\_7

Swiss Forum for Migration and Population Studies, University of Neuchâtel, Neuchâtel, Switzerland

However, skills mismatches among migrants are an issue in how they represent a "brain waste" for the migrant and for the host economy.

Among the factors explaining immigrant downgrading in the forms of overeducation (when workers have more education than is required for their jobs) or skills mismatch more generally are the imperfect transferability of foreign credentials and skills acquired abroad (Chiswick and Miller 2008, 2009). This problem often results from poor knowledge of the foreign educational system held by the employers in the host country (e.g., Buzdugan and Halli 2009). Consequently, employers minimize uncertainty by prioritizing workers educated in the home country or by hiring overeducated migrants. This handicap for migrants can be partially reverted by means of foreign credential assessment. In other words, the recognition of foreign credentials might be a means to avoid or overcome situations of overeducation (see for instance Nielsen 2011; Pecoraro 2011).

The literature on the possible effect of recognizing skills acquired abroad is rather scarce. Brücker et al. (2016) studied the employment and wage effects of occupational recognition. Based on a linked dataset from the IAB-SOEP Migration survey and the German social security data, their empirical analysis shows gains from occupational recognition in the German labour market.1 Accounting for the dynamic structure of their data, they also indicate a faster assimilation of immigrant earnings relative to native earnings resulting from occupation recognition. A more recent study by Tani (2017) also investigated whether assessing foreign education increases returns to schooling of migrants in Australia. Using data from the Longitudinal Survey of Immigrants to Australia (LSIA), he indicates substantial wage improvements when foreign qualifications are officially assessed. Assessing foreign education provides, then, a clearer signal of productivity, improving the transferability of human capital and migrants' economic assimilation.

In parallel, a significant number of papers have examined the determinants and consequences of migrants' overeducation, with a particular focus on the wage effects (e.g., Chiswick and Miller 2008, 2009; Green and McIntosh 2007; Lindley 2009; Nielsen 2011). Few focussed on the migrants' strategies to avoid overeducation or skills mismatch. However, the effect of the recognition of the foreign qualification on the mismatch is not documented in the literature. This chapter aims at filling this gap with original data from an immigration country such as Switzerland.

The aim of this chapter is to investigate the relationship between foreign credential recognition and both educational and skills mismatches. To our knowledge, such a relationship has never been investigated. We focus on the migrant population living in Switzerland, which represents one-third of the country's workforce, according to the Swiss Federal Statistical Office.2 The availability of new data from the Migration-Mobility Survey allows us to measure a mismatch based on the

<sup>1</sup>The IAB-SOEP Migration Sample is a household survey conducted jointly by the Institute for Employment Research (IAB) in Nuremberg and the German Socio-Economic Panel (SOEP) at DIW Berlin.

<sup>2</sup> 31% in 2017, cf. https://www.bfs.admin.ch/bfs/fr/home/statistiques/catalogues-banques-donnees/communiques-presse.assetdetail.2633986.html. Accessed 25 May 2018.

migrant's self-assessment of job matching in terms of education and actual skills. For this purpose, after a review of the literature (Sect. 7.2) and a description of data and methods (Sect. 7.3), we initially measure the incidence of both educational and skills mismatches among groups of (recent) migrants defined according to the place of origin and the status of recognition of the foreign diploma (Sect. 7.4.1). In a second step, we aim at verifying to what extent there is a significantly positive relationship between the foreign-acquired education (whether or not officially recognized) and the probability of being overeducated or in a situation of skills mismatch. Finally, the reasons explaining educational mismatch and skills mismatch are investigated (Sects. 7.4.2 and 7.4.3). The chapter closes with a conclusion (Sect. 7.5).

#### **7.2 Review of the Literature**

Various theoretical models have been advanced to explain the presence of overeducation in the labour market (for overviews, see Hartog 2000; Leuven and Oosterbeek 2011). Among them is the human capital model (Becker 1964). According to it, skills learnt in formal education constitute an incomplete measure of the human capital endowment, which includes other important elements such as early ability (either acquired or innate, before entering the schooling system), skills acquired through training on the job, and labour market experience. Workers with different levels of education can provide the labour market with the same amount of human capital. In line with Becker's model, overeducation might then substitute for a lack of other elements of the human capital endowment (Sloane 2003). In other words, although the human capital endowment is vaster than the *educational knowledge*, the existence of overeducation does not imply excess human capital per se and could be associated with a lack of *non-educational knowledge*. This explanation is particularly prevalent for migrants who, to some extent, might lack knowledge of the local language, lack knowledge concerning local legislation, or face a lack of networks within the local market.

Sicherman and Galor (1990) proposed an adaptation of the human capital model in which overeducation is considered an opportunity for occupational upgrading and is thus a temporary state at the early stage of a career. According to this model of career mobility, the wage penalty for being overeducated is compensated for by future wage growth through promotion or on-the-job training. Based on US data, the authors found support for the career mobility model, which was also demonstrated by subsequent studies (e.g., Robst 1995; Rubb 2006; Sicherman 1991). Although the empirical evidence on the link between overeducation and upward mobility is not clear-cut in the context of European labour markets (e.g., Baert et al. 2013; Büchel and Mertens 2004; Korpi and Tåhlin 2009), a recent study by Grunau and Pecoraro (2017) confirms Sicherman and Galor's prediction for Germany. By analogy, according to Sicherman and Galor (1990), migrants can be assigned to jobs for which they are overeducated and can accept low wages during the first years in

the host country, with the hope of improving their position and earnings in the following years.

It is not surprising to observe an increased level of educational mismatch among migrants compared with natives (Aleksynska and Tritah 2013; Leuven and Oosterbeek 2011). This point is particularly true for cohorts of immigrants who arrived in the first decade of the twenty-first Century in the United Kingdom (Lindley 2009), a result that is confirmed based on 13 European countries by Prokic-Breuer and McManus (2016) except among workers with the highest cognitive proficiency. Piracha and Vadean (2013) and Visintin et al. (2015), among others, propose reviews of the literature on migrant educational mismatch. Visintin et al. (2015) used data on almost 700,000 workers in 86 countries to better describe the migrant groups experiencing overeducation in Europe. According to the authors, this situation is generally more common among migrants from the EU15 and Asia but not among those from Africa or South America.

Among the reasons suggested for the higher likelihood of migrant overeducation, the imperfect international transferability of human capital is mentioned by Chiswick and Miller (2009) in the case of foreign-born men in the United States and by Nieto et al. (2015) for immigrants from non-EU countries. Tani et al. (2013), Sanromá et al. (2015) and Basilio et al. (2017) confirm the importance of this interpretation for immigrants to Australia, Spain and Germany, respectively; they show evidence of significant variation in the wage returns from nominally equivalent qualifications obtained in different countries.

Other alternative explanations have been suggested. One frequently mentioned is that overeducation might be due to discrimination in the destination labour market (Battu and Sloane 2004; Nielsen 2011). Inadequate language skills of migrants is another factor highlighted by Green and McIntosh (2007) and Prokic-Breuer and McManus (2016). Piracha et al. (2012) also argued that migrants' overeducation in the destination country is to a large extent explained by the matching status prior to migration, a worker being overeducated in the country of origin having a higher probability to be in the same situation after migration. This finding is in line with a human capital model of migration in which skills are not always fully utilized in the home country (Nowotny 2016). Thus, overeducated migrants are negatively selfselected, that is, have a lower skill level than non-migrants.

In Switzerland, regulated professions such as medicine and law require a formal recognition of foreign qualifications. According to the State Secretariat for Education and Innovation (SERI), which oversees the recognition procedures, "A profession is deemed to be regulated when legal or government provisions stipulate that a specific qualification is required to work in the given profession". This requirement not only affects occupations in the health sector (e.g., physician, psychologist, and assistant in community health) and in the law sector (e.g., barristers and solicitors) but also concerns various other occupations (such as taxi driver, teacher, and civil engineer). In the case of differences between the migrant's foreign education and the corresponding diploma acquired in Switzerland, the SERI can require an aptitude test to verify the skills content. Occasionally, additional years of education are required.3 This requirement can discourage people from initiating the procedure when the upgrading process is too complicated.

For the non-regulated occupations, recognition is not necessary. In this case, the employer decides whether to hire the migrant. The employer also might ask the migrant to attend additional education or on-the-job training. To some extent, the migrant can request an official assessment of his or her credentials to meet Swiss standards, which can help to find a job corresponding to his or her actual skills. Such a process is generally proposed by unemployment offices or organizations that facilitate migrants' integration in the labour market.

Switzerland presents rather low rates of educational mismatch in international comparisons (Quintini 2011). Consequently, the high proportion of migrants in its population tends to increase the average extent of overeducation because migrants are more affected than natives are (Pecoraro 2011). However, Switzerland is an interesting case to study in this context, in particular because of the availability of survey data that provide information on educational and skills mismatch for recently arrived migrants.

#### **7.3 Data and Methods**

The Migration-Mobility Survey 2016, described in Chap. 2 of this book, was used to estimate the link between job mismatch and the place of education on one hand, and the recognition of foreign qualifications on the other hand. As a reminder, the sample includes 11 groups of migrants from different countries or regions of birth. Among countries that were excluded were the Balkan countries and the Asian countries, India excepted. Therefore, the survey is not representative of the entire working population of foreign nationality, although the largest groups of migrants living in Switzerland, representing together 68% of the active migrant population, are part of the survey. Moreover, the survey oversampled migrants who arrived in the last 2 years. Weights are computed to make the data representative of the migrant population living in Switzerland (see Chap. 2). When interpreting the results, we must remember that the survey covers the migrants who arrived in the last 10 years in Switzerland.

For this chapter, migrants without any formal education are excluded because we are primarily interested in the transfer of foreign-acquired education into the Swiss labour market. Due to a technical problem, the survey did not collect any information on the recognition of foreign qualifications among PhD holders; therefore, this group is also excluded. Only migrants with a job are included in the analysis (4,190 persons).

<sup>3</sup>SBFI, recognition of foreign qualifications, https://www.sbfi.admin.ch/sbfi/en/home/bildung/ recognition-of-foreign-qualifications/recognition-procedure-on-establishment.html. Accessed 14 June 2018.

The respondents were invited to indicate the country in which the highest educational qualification was obtained (question D2). Countries are grouped in four broad regions: education acquired in Switzerland (reference), in a EU28/EFTA country, in another OECD country, or in another country in the rest of the World.

Concerning the recognition of foreign qualification, one question is about the official requests made by the respondents to obtain a certificate of equivalence when their highest educational qualification was acquired abroad. The following answers are included: (1) Yes, the certificate was obtained; (2) Yes, but the certificate was not obtained; (3) Yes, but the procedure is not yet complete; (4) No, it was not necessary; and (5) No, for other reasons.

Educational and skills mismatches are both derived from the worker's selfassessment method. Although all measurement methods tend to induce measurement error (see, e.g., Hartog 2000; Leuven and Oosterbeek 2011, for the pros and cons of each existing measurement method), the subjective method presents the advantage of relaxing the assumption that all jobs within a given occupation have the same requirements.

The measure of educational mismatch is based on the comparison between the highest level of education obtained by the respondent and the answer to the following question: "What type of education do you feel is most appropriate for your current job?" The respondent could answer as follows: (1) No formal educational qualification; (2) Compulsory education; (3) Higher secondary education not giving access to universities (or similar); (4) Vocational education and/or training; (5) High school-leaving certificate giving access to universities (or similar); (6) Advanced technical and professional training; (7) Bachelor or equivalent; (8) Master or equivalent; or (9) PhD or equivalent.

Overeducation (resp. undereducation) is defined as situations in which the highest level of education achieved is higher (resp. lower) than the one that is considered the most appropriate for the current job. This measurement approach is standard in the overeducation literature (see, e.g., Duncan and Hoffman 1981; Hartog 2000; Leuven and Oosterbeek 2011).

Skills mismatch is computed based on the following question: "On a scale from 0 'not at all' to 7 'to a very high extent', to what extent are your knowledge and overall skills utilized in your current work? By knowledge and overall skills, we mean your formal education and the skills you obtained while working (on-the-job


**Table 7.1** Coherence between educational and skills mismatch (in %)

Sample size: 4,171

training)." This question is similar to those generally used in the overskilling literature (see, e.g., Allen and van der Velden 2001; Green and McIntosh 2007; Mavromaras et al. 2012). Answers ranging from 0 to 3 are viewed as reflecting skills mismatch because answers ranging from 4 to 7 are viewed as reflecting a good or perfect use of skills.

In line with the literature, the consistency between both indicators is relatively low (Table 7.1); only 54% of the respondents report an adequate education and adequate skills, and 33% of the respondents report adequate skills but are either undereducated (13%) or overeducated (20%). Moreover, 10% report both skills and educational mismatches (9% overeducation, 1% undereducation), and 3% report a situation of skills mismatch while being adequately educated (Table 7.1). This low degree of consistency is explained by the fact that both dimensions reflect different, but also complementary, notions of human capital utilization. Educational mismatch is a rather objective measure but solely based on the worker's educational credential, whereas skills mismatch translates largely into satisfaction in terms of skills utilization on the job. The latter indicator can also reflect dissatisfaction concerning the tasks that are given to the migrant worker.

Different regression models were applied to the data to determine the statistical associations between the key variables under study (country of education and request for recognition of diploma) and the risk of either educational mismatch or skills mismatch. Our expectation concerning these associations is formulated in accordance with Chiswick and Miller's (2009) evidence of imperfect international transferability of human capital and with Brücker et al. (2016) results showing improved labour market adjustment due to occupational recognition; the recognition of foreign credentials is expected to be negatively related to overeducation and skills mismatch more generally. Standard control variables were integrated into the models. Those confounding factors are the following:


Because the dependent variables present different forms, two different models are tested. We use a multinomial logit regression model to identify the determinants of educational mismatch. This model is appropriate to express the probability of both overeducation and undereducation, the alternative being an absence of mismatch. We assume that educational mismatch for individual *i* is determined by the following model: *Mi*\* = *Oi* α + *Xi* β + *ei*, with *i* = 1, 2, ..., *N*. *O* is a vector including the key variable under study (either country of education or recognition of diploma), *X* is a vector including control variables (including a constant), and *e* is a term reflecting the unobservable component of educational mismatch. Given that the latent model is not observed, we define the polytomous variable *M* as the realization of three possible states:

$$\mathcal{M}\_{i} = j = \begin{cases} 1 & \text{if } \mathcal{M}\_{i} < \mu\_{1} \qquad \text{(adequate education)} \\ 2 & \text{if } \mu\_{1} \le \mathcal{M}\_{i} < \mu\_{2} \text{ (underedLocation)} \\ 3 & \text{otherwise.} \qquad \text{(overreduction)} \end{cases}$$

The probabilities for individual *i* to be in situation *j* are

$$\mathbb{P}\left(M\_i = j \mid O\_i, X\_i\right) = \frac{\exp\left(O\_i \alpha\_j + X\_i \beta\_j\right)}{\sum\_{h=1,2,3} \exp\left(O\_i \alpha\_h + X\_i \beta\_h\right)}.$$

We rely on three specifications to estimate the determinants of educational mismatch and of whether foreign credentials are transferable in the host labour market. Therefore, in addition to the confounding variables, each specification includes additional explanatory variables that measure education acquired abroad and the

<sup>4</sup>SFSO, regions of analyses, https://www.bfs.admin.ch/bfs/fr/home/statistiques/themes-transversaux/analyses-spatiales/niveaux-geographiques/regions-analyse.html. Accessed 14 June 2018.

status of foreign credential recognition: the country of education (specification 1), the status concerning the recognition of a foreign diploma (specification 2), and an interaction between the recognition of a foreign diploma and the country of education (specification 3). The relative risk ratios are presented along with the level of significance (\* < 0.10; \*\* < 0.05).

We use a logistic regression model to explain the probability (*p*) of skills mismatch according to different factors, including whether education is acquired abroad (specification 1′), whether foreign education has been recognized (specification 2′) or both (specification 3′). The formula is the following:

$$\operatorname{logit}\left(p\right) = \ln\left(\frac{p}{1-p}\right) = X\_i\beta,$$

where *X* is a vector of explanatory variables (including a constant), and the exponential value of *β* provides the odds ratios associated with each of the explanatory variables in the three specifications.

Given the cross-sectional nature of the data and the empirical methods outlined above, we cannot address adequately the issue of non-random selection into the application process for recognition as stressed by Brücker et al. (2016) and Tani (2017). Accordingly, the expected results should best be interpreted conservatively as descriptive. Finally, all descriptive and regression analyses incorporate individual weights to consider the sampling design of the Migration-Mobility Survey and thus obtain reliable estimates concerning the population of interest. Accordingly, we calculate robust standard errors using the "linearization" variance estimator based on a first-order Taylor series linear approximation (Eltinge and Sribney 1997).


**Table 7.2** Educational mismatch and skills mismatch according to the status of recognition of foreign qualification (in %)

Sample size: 4,184

#### **7.4 Results**

#### *7.4.1 Incidence of Educational and Skills Mismatches*

Table 7.2 shows the proportion of respondents who are adequately educated, undereducated or overeducated according to the status of recognition of foreign qualification. This table also displays the incidence of skills mismatch. The proportion of migrants with adequate qualifications is the highest among those educated in Switzerland (75%) or among those who obtained recognition (69%). It is the lowest among those who requested recognition but did not obtain it (36%) and among those who did not ask for recognition for other reasons (28%).

The group of those who did not ask for recognition for other reasons is also characterized with a high incidence of overeducation (56%), which also affects approximately 45% of the persons who asked for recognition but did not obtain it. By contrast, overeducation is rather low (less than 25%) among migrants who did not need recognition or obtained it. The incidence of overeducation is the lowest among persons who obtained their diploma in Switzerland.

Skills mismatch, which arises in 13% of the sample, is less frequent among migrants educated in Switzerland (3%) or having obtained recognition (7%) than among other groups, in particular those not having obtained recognition (33%) or not having requested it (28%).

Table 7.3 provides the distribution of educational and skills mismatch according to the region of origin, which was self-reported by the respondents. The share of persons with adequate education is higher among those from North America (69%)


**Table 7.3** Education mismatch and skill mismatch according to the region or origin (in %)

Sample size: 4,171


**Table 7.4** Reasons for educational mismatch and skills mismatch, among those concerned (in %)

171

Note: Multiple answers possible

Source: Migration-Mobility Survey 2016. Weighted results

and Asia (most of them being Indian migrants, 78%), who generally arrive in Switzerland with a job contract from multinational companies or universities. This share is also high (64%) among migrants from countries bordering Switzerland (Austria, France, Germany, and Italy). It is lower among migrants from another European country (primarily Portugal and Spain, 43%) and from South America (38%). The latter group has the highest incidence of overeducation (53%), whereas the incidence of undereducation is the highest among migrants from other EU28/ EFTA countries (19%).

The highest rate of skills mismatch is observed among South Americans (26%) and migrants from other EU28/EFTA countries (20%). This rate is less than 10% among migrants from countries bordering Switzerland, Asia, other Europe and North America.

Migrants who were either overeducated or mismatched in skills were asked to provide the reasons for their current situation (Table 7.4). For both indicators of mismatch, the main reason is the wish to avoid unemployment, i.e., the wish to work even when its cost is to be overeducated or mismatched in skills. Approximately 30% of the respondents reported this reason. The second reason, which constitutes approximately 20% of the migrants, refers to the lack of local language skills. Lack of jobs with corresponding qualifications, family obligations and a change of career are other reasons that are mentioned.


172


(1) UE

OE

UE

OE

UE

OE

(2)

(3)


**Table 7.5** (continued)


(1) UE

OE

UE

OE

UE

OE

(2)

(3)


groups are men, vocational education and/or training (highest level of education), on-the-job training: no, no job before immigration, does not understand the local language at all, does not speak the local language at all, C permit, birth countries bordering Switzerland (Italy, Germany, France, Austria), Single, Lake Geneva (Vaud, Valais, Geneva), highest level of education acquired in Switzerland. Individuals without any formal education or with a PhD are excluded Source: Migration-Mobility Survey 2016. Weighted results (normalized weights)

#### *7.4.2 Factors Explaining Educational Mismatch*

Estimation results on the determinants of educational mismatch are presented in Table 7.5. We first discuss the estimates of relative-risk ratios for the control variables before commenting on the estimates associated with the country of education and the recognition of a foreign diploma.

Gender is significantly associated with the risk of both overeducation and undereducation; women are less likely to be undereducated but are more likely to be overeducated compared with men. A low level of education increases both risks, whereas a high level of education increases the risk of being overeducated but diminishes the risk of undereducation.

On-the-job training is statistically significant, and can be viewed as an investment that improves human capital accumulation. That is, it is positively associated with the risk of undereducation but is negatively associated with the risk of overeducation.

Compared with a situation in which migrants are unemployed before migration, those migrants working for either the same or another company are significantly less likely to be overeducated. By contrast, having a job before migration is not statistically associated with the risk of undereducation. This result can be explained by the fact that migrants arriving in Switzerland after an episode of unemployment or immediately after obtaining their diploma are more inclined to work even when their educational qualifications do not correspond to the job requirements. Most likely, those migrants consider the current job in Switzerland an opportunity for occupational upgrading.

The results also show that a low knowledge of the local language is not related to overeducation but is negatively associated with the risk of undereducation. Moreover, holding an annual permit B, compared with a long-term permit C, is negatively associated with the probability of overeducation.5 Although surprising, this result can be explained by the fact that holders of a resident permit B generally fill jobs for which there exist labour shortages (Pecoraro and Ruedin 2017). A shortterm permit L, in contrast, increases the probability of undereducation.

Compared with migrants originating from one of the neighbouring countries, the probability of undereducation is higher among migrants from South America (except in specification 2) and migrants belonging to an OECD country outside Europe (except in specification 1). The risk of overeducation is higher among other EU28/EFTA migrants, migrants from an OECD country outside Europe (specification 2), and from South American countries.

When those confounding factors are controlled for, we find that the place of education plays a significant role in the case of both undereducation and overeducation (see specification 1). For a migrant, having obtained his diploma in a country from the rest of the World significantly decreases the probability of being undereducated,

<sup>5</sup>Given the close correlation between permits and the duration of residence in Switzerland, the association between years since migration and the probability of educational mismatch is generally not significant. The same applies when examining the determinants of skills mismatch.

compared with a situation in which the highest diploma was obtained in Switzerland. In contrast, having obtained the diploma not in Switzerland increases the risk of being overeducated. Such an increase is observed among migrants educated in a EU28/EFTA country. By contrast, the risk is reduced for those having obtained their diploma in another OECD country. The latter result can most likely be explained by the specificities of North Americans, who generally arrive in Switzerland for work purposes with high-quality educational credentials. For the migrants belonging to the group of "other countries", the situation is different because this group is more heterogeneous, with an overrepresentation of migrants arriving in Switzerland with reasons other than professional (see Chap. 5 of this book).

In specification 2, we observe that the status of non-recognition of the foreign diploma is significantly related to the risk of overeducation. First, compared with the migrants holding a Swiss diploma, those who obtained recognition present a similar risk of overeducation. This risk is however significantly higher for those who considered recognition unnecessary (the relative risk of being overeducated is expected to increase by a factor of 1.67 given the other variables in the model are held constant) and for those who did not complete recognition (the relative risk is expected to increase by a factor of 2.82 *ceteris paribus*) or did not obtain it (the relative risk is expected to increase by a factor of 3.15 *ceteris paribus*). It is the highest among migrants who did not ask for recognition (the relative risk is expected to increase by a factor of 5.65 *ceteris paribus*), confirming the importance of the recognition of a foreign diploma in the host labour market. When focussing on the interactions between both variables (specification 3), we observe that those educated in EU28/EFTA countries face a higher risk of overeducation when recognition was not obtained. The same result is observed for migrants educated in other countries, in particular when recognition is not obtained or not requested for other reasons. For this group, the recognition of the diploma also leads to a situation in which the likelihood of overeducation does not significantly differ from that of migrants who have obtained a Swiss diploma. Results concerning undereducation are less clear and more difficult to address.

#### *7.4.3 Factors Explaining Skills Mismatch*

The determinants of skills mismatch were also examined using three specifications (Table 7.6), with the same variables on the right-hand side as the ones used for educational mismatch. Again, we describe the results related to the control variables before focussing on the two dimensions of interest, i.e., the country of education and the recognition of a foreign diploma.

The likelihood of skills mismatch is higher among women compared with men, among migrants from other EU28/EFTA countries, from other OECD countries (in specification 1′ only) and from South America compared with migrants from countries bordering Switzerland. Compared with migrants who graduated from a vocational track, the odds ratio is lower among those with a tertiary-level education


**Table 7.6** Results of logit models. Determinants of skills mismatch (odds ratios)


#### **Table 7.6** (continued)


#### **Table 7.6** (continued)

#### **Table 7.6** (continued)


Note: The outcome variable is a binary variable for the risk of skills mismatch. Linearized SE in parentheses, \*\* p < 0.05, \* p < 0.10. Reference groups are men, vocational education and/or training (highest level of education), on-the-job training: no, no job before immigration, does not understand the local language at all, does not speak the local language at all, C permit, birth countries bordering Switzerland (Italy, Germany, France, Austria), single, Lake Geneva (Vaud, Valais, Geneva), highest level of education acquired in Switzerland. Individuals without any formal education or with a PhD are excluded

Source: Migration-Mobility Survey 2016. Weighted results (normalized weights)

(bachelor or master). Having a job before migration decreases the probability of skills mismatch, which confirms the results observed for the determinants of educational mismatch.

Having controlled for those control variables, the country in which the diploma was acquired appears to be significantly associated with the risk of skills mismatch (specification 1′). The estimated odds ratio is greater than five for those having completed their education in a EU28/EFTA country or in the rest of the World compared with the case in which the diploma is acquired in Switzerland. The risk of skills mismatch is also higher for those with a diploma from another OECD country, but the related estimate is not significant. Once again, the relatively good situation of this group on the labour market can be explained by the characteristics of the migrants from North America. Given the job opportunities and the wages in their country of origin, those migrants generally arrive in Switzerland when they have an assurance of obtaining a job that provides them better working conditions than they can expect in the country of origin. Such conditions are generally granted in the absence of skills mismatch. Moreover, among North American migrants, the circulatory migrants who are employed by their (multinational) companies are overrepresented; according to the survey, among North American migrants who had a job at the time of migration, 43% worked for the same company as before migration (job transfer) and then present a low risk of skills mismatch when reallocated in a new country.

Compared with a holder of a Swiss diploma, a migrant having obtained the recognition of a foreign diploma has a higher risk of skills mismatch (specification 2′), but the related estimate is not significant. Other groups have a higher risk of skills mismatch, significantly different from that of the reference population. This point is particularly true for people who have not completed recognition or who did not obtain it (compared with the migrants holding a Swiss diploma, the odds of being mismatched in skills is greater than ten in both cases). The interaction between both variables (specification 3′) confirms those results, with the highest estimate of the odds ratio among people from the rest of the World not having obtained recognition.

#### **7.5 Conclusion**

Our analyses aimed at examining the relationship between the recognition of a diploma and the risk of skill mismatch. Before further discussion of the overall results, we must mention some limitations. First, analyses based on survey data are prone to different forms of social desirability bias, in particular for questions related to the level of education. Second, surveys on migrants are generally difficult to perform because this population is very mobile, which is particularly true for recently arrived migrants. Another issue when surveying recently arrived migrants is related to their capacity to understand the local language and to fill in survey questionnaires. However, as mentioned in Chap. 2, the survey was specifically designed for migrant populations, with a questionnaire translated into the migrants' mother tongue.

The main results are the following. First, concerning the incidence of overeducation, our results partially contradict those of Visintin et al. (2015), who observe that overeducation is more common among migrants originating from the EU15 and Asia but not among those from Africa and South America. Although the migrants from EU28/EFTA without common boundaries with Switzerland are shown to exhibit a high rate of overeducation (almost 40%), South American workers have the highest rate at more than 50%. The latter, who represent a small group of migrants in Switzerland, most likely suffer from a lack of support from their migrants' community in terms of professional integration. Moreover, migrants from Asia face a low incidence of overeducation compared with the reference group (i.e., migrants from EU28/EFTA countries bordering Switzerland), a result that can be explained by the highly educated profile of this group, largely composed of migrants from India.

Second, the significant association between the place of education and the risk of both skills and educational mismatches confirms the hypothesis of an imperfect international transferability of human capital mentioned by Chiswick and Miller (2009). Although educational systems abroad are most likely as good as the one in Switzerland in a large range of countries, foreign diplomas do not secure access to related jobs in Switzerland. Therefore, the recognition of foreign credentials can be viewed as a strategy to decrease the risks of overeducation and skills mismatch, in particular for the regulated professions.

Finally, by focussing on the link between job mismatch and foreign-acquired education, our analysis provides results that can have direct implications for the administrations in charge of the policies on the recognition of foreign diplomas. We show that the recognition of foreign credentials is expected to reduce the risk of skills or educational mismatch. Although this relationship are unlikely to be causal, specific policies that aim at encouraging requests for recognition of non-regulated professions and strategies to facilitate the process (for instance by promoting access to additional courses) might improve wages and, by extension, job satisfaction for migrants. Further research confirming that the causal effect of credential recognition is indeed beneficial to immigrants' assimilation in the host labour market is needed more than ever.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Part IV Social Life and Political Participation**

## **Chapter 8 Immigrants' Feelings of Attachment to Switzerland: Does the Cantonal Context Matter?**

**Salomon Bennour and Anita Manatschal**

#### **8.1 Introduction**

Immigration, increasing ethnic diversity, and managing the integration of foreigners into host societies are among the most pressing issues of contemporary societal and political debates. Consequently, academic interest in these topics is not only increasing but also spreading across social science disciplines, which hitherto rather neglected these subjects. Numerous studies focus on how receiving countries address and react to immigration, for instance, by studying migration policies such as immigration, integration policies, or natives' attitudes towards immigrants. Much less is known about immigrants'1 attitudes towards the host country or society and how these attitudes are affected by the political and societal receiving context to which immigrants are exposed. Existing research suggests that national identity, in the sense of a feeling of attachment to a country, can assume the role of an overarching identity that reduces social distances between distinct groups in a society, diminishing social conflicts and contributing to social cohesion (Transue 2007; Verkuyten and Martinovic 2012).

Given the potential role of national identity as a "social lubricant" (Miller 1995; Reeskens and Wright 2014), the purpose of this study is thus to examine the factors that enhance immigrants' feelings towards their new country of residence in terms of national identification. More specifically, we are interested in how cantonal norms of inclusion or exclusion affect immigrants' national identity in terms of their

<sup>1</sup>Throughout this chapter, we use the terms immigrants and non-citizens interchangeably. Note that only one of the 5855 respondents included in our analysis is not a non-citizen (meaning the respondent indicated Swiss nationality), implying that de facto our results apply to non-citizen immigrants living in Switzerland.

S. Bennour (\*) · A. Manatschal

University of Neuchâtel, Neuchâtel, Switzerland e-mail: salomon.bennour@unine.ch; anita.manatschal@unine.ch

<sup>©</sup> The Author(s) 2019 189

I. Steiner, P. Wanner (eds.), *Migrants and Expats: The Swiss Migration and Mobility Nexus*, IMISCOE Research Series, https://doi.org/10.1007/978-3-030-05671-1\_8

feelings of attachment to Switzerland. Our main focus will be on the crystallization of these norms in cantonal integration policies. Alternative measures to capture the degree of cantonal exclusivity or inclusivity towards immigrants include, first, popular attitudes towards immigrants aggregated at the cantonal level and, second, cantonal vote shares of conservative right parties.

Research on integration, immigration and citizenship policy is abundant. Whether qualitative or quantitative in orientation, the predominant approach of these studies is based on cross-country comparisons of national policies (e.g., Brubaker 1992; Bjerre et al. 2015; Favell 2001; Goodman 2010; Koopmans et al. 2012). This narrow, nation-state focus is increasingly criticized (Castles 2010) by scholars emphasizing striking policy differences (e.g., between Italian regions or Swiss cantons) and by bustling integration policymaking activities (e.g., in the US states) at the subnational level (Condon et al. 2016; Manatschal 2012; Wisthaler 2016). Subnational policy heterogeneity creates different living conditions for immigrants and, therefore, a non-negligible heterogeneity in integration possibilities (Cattacin and Kaya 2001). A large number of studies focus further on determinants of integration policies such as immigration history, meaning path-dependency (Favell 2001), or the influence of far-right parties (Koopmans et al. 2012). However, few studies analyse integration policy outcomes. For instance, Cebolla-Boado and Finotelli (2011) evaluate the effect of integration policies on labour market integration, and Manatschal and Stadelmann-Steffen (2013) scrutinize policy outcomes on immigrant educational inequality across cantons. Our study addresses the two gaps identified above: First, we try to understand the effect of integration policies on immigrants from a thus far neglected perspective, their feelings of national attachment. Second, by focussing on cantonal integration policy, the study complements the emerging body of research emphasizing the relevance of subnational integration policy heterogeneity.

In addition to cantonal integration policy, we also account for the potential effects of societal norms on immigrants' national identification in terms of majority attitudes towards immigrants. Several scholars scrutinize the different determinants of public attitudes towards immigrants (e.g., xenophobia), investigating, for instance, the effect of the type of the welfare state (Koopmans 2010) or the effect of news media (Boomgaarden and Vliegenthart 2009). Similarly to the research on integration policy, outcome studies focussing on how inclusionary or exclusionary popular attitudes affect immigrants' attitudes or behaviour in turn remain scarce. Only a handful of studies aim at analysing the outcomes of natives' attitudes on noncitizens' location choices (Slotwinski and Stutzer 2015), return migration (de Coulon et al. 2016) or labour market discrimination (Waisman and Larsen 2007). Our study adds to this research because we are interested in the more immediate effect of welcoming or hostile attitudes towards immigrants on non-citizens' sense of national identity.

Beyond these specific contributions to the literature, this chapter also speaks directly to the Migration-Mobility Nexus standing at the core of this edited volume (see Chap. 1). First, the study follows the sedentary premises of classical migration research when focussing on how societal norms affect immigrants' integration

outcomes (Castles 2010; Dahinden 2016). Second, mobility scholars argue that governance of migration and mobility occurs at subnational, national or supranational levels. By focussing on the subnational policy level, this study circumvents the narrow focus on the nation-state, which is predominant in classical migration studies (Wimmer and Glick-Schiller 2006). Third, this chapter also contributes to this nexus by questioning what makes immigrants feel attached to their host country. This emotional connection is likely to be related to higher intentions to remain in Switzerland rather than to leave again. In this sense, the findings of this study can also help to better understand potential drivers of individual (im-)mobility trajectories. This study builds on different data sources. The main data source is the Migration-Mobility Survey conducted in 2016, from which we draw our dependent variable, immigrants' feelings of attachment to Switzerland. This survey clearly surpasses other population surveys in Switzerland because it is the only comprehensive and nationally representative survey focussing explicitly on immigrants. The survey is conducted in six different languages, which also guarantees the inclusion of non-language-assimilated immigrants. Even more importantly for the present study, the large sample of 5973 immigrant respondents facilitates individual analyses at the level of the Swiss cantons, which is hardly feasible with general population surveys. To measure cantonal integration policies, we draw on data collected by Manatschal (2011). Public attitudes towards immigrants at the cantonal level, in turn, will be captured by direct democratic vote results (e.g., the national vote on "mass immigration" from 2014), which are less distorted by social desirability than are survey data on xenophobia, and cantonal statistics on conservative-right voting rates. To empirically test the relationship between societal norms at the cantonal level and immigrants' individual national identities, we apply multilevel regressions suitable for cross-sectional analyses of this type, accounting for the pertinent individual and cantonal control variables discussed in the literature.

This chapter is organized as follows. Section 8.2 discusses the current state of the research and develops theoretical hypotheses concerning the assumed relationship between inclusionary/exclusionary societal norms and the national identification of immigrants in Switzerland. Section 8.3 presents the data sources and explains the research design and methods used in this study. The results of our multilevel regression analyses are presented in Sect. 8.4. Section 8.5 concludes.

#### **8.2 Theoretical Background**

National identity can be understood as an "awareness of affiliation with the nation" (Hjerm 1998, p. 337) that increases the possibility to "feel at home" (Keane 1994). "National identity" relates to a broader lexical field in the literature, with scholars using different terms such as "host-country identity" or "majority identity" to refer to the same concept. Although we use the term "national identity" in this chapter, it can be understood as a synonym for the abovementioned terms. In surveys, national identification is typically measured with questions such as "how close do you feel

to country XY?" This understanding of national identification as closeness or affiliation with a country can be demarcated from two other forms of positive perceptions of the nation. First, patriotism implies pride in the nation, e.g., in national democratic institutions. Second, nationalism reflects an uncritical and blind attachment to the nation, which is often combined with a sense of national superiority with respect to other countries (Coenders et al. 2004; Green et al. 2011).

National identity, in turn, can decrease social distance between distinct subgroups sharing the same overarching identity and, thus, diminish intergroup conflicts because individuals with distinct religious, ethnic, or home country-related identities can continue to reunite under the same overarching identity (Transue 2007). Identification with the country one lives in, meaning national identification, can take over the role of such an overarching identity. Given its role as a social lubricant (Miller 1995; Reeskens and Wright 2014), national identification by immigrants can be a remedy for social problems resulting from ethnic diversity. Indeed, immigrants' attachment to the host-country identity helps to reduce intergroup conflict via two main mechanisms. First, national identification among immigrants decreases social distance from the majority group by embracing, partly or entirely, the overarching identity (Gaertner et al. 1989). Second, negative feelings across minority groups can also shrink due to host-country identification because that identification unites a plethora of subgroup identities under an encompassing identity (Transue 2007; Verkuyten and Martinovic 2012). More generally, trust and solidarity develop more easily between individuals sharing common values related to the same identity (Doosje et al. 1999). In summary, this type of a shared overarching identity, which can be approximated by national identification, can be a solution to tensions arising from ethnically diverse societies because national identification creates a common sense of belonging (Moran 2011; Transue 2007), reduces social distance between states' residents and helps to develop national solidarity (Kymlicka 2015). Because of its beneficial effects on social cohesion, scrutinizing more closely the feeling of attachment to the host country appears highly relevant for contemporary states facing growing ethnic diversity. The first contribution of our chapter to the existing research is thus that we try to understand the effect of welcoming or hostile reception contexts on immigrants from a thus far neglected perspective, their feelings of national attachment.

A broad range of research examines the interplay between host and origin country identities. Traditionally, and in contrast to the idea of the overarching identity delineated above, scholars have depicted dual identity as impossible, meaning that identification with a specific ethnic identity is inversely related to identification with other subgroups (Verkuyten and Martinovic 2012; Gecas and Burke 1995; Hall and du Gay 1996; Simon 1997). More-recent research contradicts these earlier findings, claiming that, in reality, identities are not exclusive but rather overlapping and overarching (Bauböck 2002). For instance, support for this argument is provided by the increasing acceptance of dual citizenship attribution in many European countries (Blatter 2011; Vink and Bauböck 2013). Phalet and Swyngedouw (2002) examine the long-distance affiliation sustained by Turk and Moroccan communities in Belgium with their homelands through transnational identities. Based on this empirical evidence, the authors develop a composite understanding of identities in the context of a contemporary European state. Nandi and Platt (2015) study minorities in the UK, showing how three different identity levels (ethnic, regional and national) interact with one another. They emphasize that groups are not homogeneous in terms of identities, which can be composed of different affiliations. Based on the interaction of different identities, Verkuyten and Martinovic (2012) delineate a four-field typology of possible identity configurations: First, *assimilation* occurs when immigrants favour the host-country identity over the ethnic identity of the country of origin. Immigrants maintaining both identities as salient represent cases of *integration* or *dual-identity*, whereas a major support for the ethnic identity implies *separation*. *Marginalization* describes individuals without any specific attachment to either of the countries. These studies show that identities are not fixed and unique but rather are fluid and hyphenated.

A variety of individual factors determines the national identification process. Being able to speak the local language and being citizen of a country (Maxwell 2010) are all factors that are positively associated with the national identification of immigrant minorities with the host land. Concerning duration of stay in the host country, assimilation theory (Alba and Nee 2009) stipulates that, with time, immigrants' identities converge towards national identification with the destination country. Some studies confirm this theory, finding that the length of time since immigrants' arrival is positively correlated with national identity (Manning and Roy 2010; Nandi and Platt 2015; Platt 2014). However, other scholars find that the correlation between length of stay and host-country identity is U-shaped (Reeskens and Wright 2014), implying that recently arrived and long-time resident immigrants are the most attached to their host country. In addition, Reeskens and Wright (2014) find that a higher level of education coincides with less national group identification because highly educated individuals tend to have more cosmopolitan and transnational identities. However, this finding is partly weakened by studies considering the interplay between education and national identification (e.g., Verkuyten and Martinovic 2012). On the one hand, highly educated individuals tend to have more contacts with nationals, which increases overarching group identification; on the other hand, highly educated individuals are also more likely to notice discrimination towards themselves or, more generally, their subgroup.

According to the literature, the receiving context matters for immigrants' national identification in three respects: immigrants' perceived discrimination, natives' antiimmigrant attitudes and an ethnic conception of the nation. There is a negative relationship between perceived discrimination and minorities' national identity (Jasinskaja-Lahti et al. 2009; Maxwell 2010; Verkuyten and Martinovic 2012). Indeed, "perceived discrimination prevents minority members from developing a sense of belonging to an overarching national in-group" (Jasinskaja-Lahti et al. 2009, p. 109). As Maxwell (2010) summarizes, discrimination appears to be more justified in a context in which anti-immigrant feelings are widespread (Fetzer 2000; Joppke 2005; Paskeviciute, A., & Anderson. C. (2008). Friendly Territory: Opinion Climate, Discontent, and Immigrant Political Action in Europe. *Paper presented at the American Political Science Association conference, Boston MA, August 28–31*). In addition, natives' xenophobic attitudes undermine the inclusion of immigrants as a legitimate part of the society (Koopmans et al. 2005; Banting and Kymlicka 2006; Maxwell 2010), which can directly hamper immigrants' attachment to the host country. Moreover, Hjerm (1998) suggests that there is a positive association between xenophobic attitudes and an ethnic conception of national identity among the majority group. Not surprisingly, such a national identity based on ethnic principles leads to a greater gap in national group identification between majorities and minorities, with natives identifying more strongly with the nation than, for instance, immigrant minorities (Staerklé et al. 2010). Ultimately, natives' anti-immigrant feelings are more likely to develop when an ethnic conception of national identity prevails, which in turn paves the way for justifying discrimination. This interplay of factors suggests that the reception environment of immigrants in the host country in terms of hostile/ welcoming native attitudes is relevant to immigrants' feelings of national identity.

Few studies scrutinize the role of institutional settings, such as multiculturalist, integration and citizenship policies, with respect to the host-country identification of immigrants. Outcomes of these different manifestations of migration policy can be studied through two theoretical lenses. First, an instrumental-legalistic perspective, based on the economic rational-choice model, stipulates that individuals react to institutional rules defined by migration policies based on a purely rational costbenefit calculation (Van Hook et al. 2006). For instance, according to this logic, a non-citizen would move from the canton of Nidwald to Zurich because access to naturalization is comparatively easier in Zurich. Because our study does not relate to material benefits but rather to the receiving political context in terms of atmosphere, we follow a second theoretical approach instead, which highlights policies' "symbolic boundaries" (Lamont and Molnár 2002). The national self-understanding of a state defines who belongs to a community, who does not (meaning the boundaries of nation states), and consequently the reception environment of a given territory (Koopmans et al. 2005). Van Hook et al. (2006) describe the importance of feeling welcomed at arrival to understand, for instance, naturalization intentions. Therefore, by giving a "warm handshake" or a "cold shoulder" (Reeskens and Wright 2014) to immigrants, the country's discursive opportunity structure (Koopmans et al. 2005) leads immigrants to construct an "emotive understanding of membership" (Bloemraad 2013, p. 201) generated by governments. The literature discusses different political institutions as important drivers of national identification, such as multiculturalist policies (Wright and Bloemraad 2012), access to citizenship (Ersanilli and Koopmans 2010) and integration policies (Dinesen and Hooghe 2010), without however reaching unanimous conclusions, as we further elaborate upon below.

Drawing on the notion that national self-understanding and reception environments crystallize in territories' integration policies (Koopmans et al. 2005, p. 6), the chapter transfers this approach to the subnational level. This chapter addresses the puzzle of how the subnational reception environment, measured amongst others through cantonal integration policies, affects the feeling of attachment to the host country. As mentioned above, the literature finds consensus on the negative correlation between hostile natives' attitudes towards immigration and immigrants' identification with an overarching identity. However, mixed findings emerge when scholars address the effects of inclusive or exclusive integration policies. Some studies find no effect of inclusionary policies on national identification (Hjerm 1998; Dinesen and Hooghe 2010; Wright and Bloemraad 2012). Contradicting these results, more-accessible citizenship policies yield positive effects on minorities' identification with the host country according to Ersanilli and Koopmans (2010). Multiculturalism and citizenship policies have attracted great attention from scholars. At the same time, little is known about the precise role of more-comprehensive integration policies. Only two papers address this specific type of policy (Dinesen and Hooghe 2010; Reeskens and Wright 2014), and these studies do not find any effect of integration policies on the feeling of attachment to the host country.

One major limitation of these studies is that they focus unanimously on national policies, which might not be the most adequate analytical level when investigating interpersonal relationships or personal attitudes (Gundelach and Manatschal 2017; Kesler and Bloemraad 2010). National identification is an individual attitude that develops when exposed to local, regional, and national discourses and policies. Given that individuals' everyday lives are more or less closely tied to the specific locality and canton in which they live, we can assume that the regional reception and integration policy environment is decisive in shaping an individual's national identification. The second contribution of this chapter, therefore, concerns the political unit under study. Complementing the abovementioned studies addressing receiving contexts and national identification exclusively at the national level, this paper scrutinizes the pronounced but often neglected subnational integration policy heterogeneity and how that heterogeneity affects the national identification of noncitizens in Switzerland. Thus, this paper adds to the recently emerging literature on subnational integration policy outcomes (e.g., Condon et al. 2016; Gundelach and Manatschal 2017; Manatschal and Stadelmann-Steffen 2013).

Based on the theoretical reflections delineated above, our first hypothesis stipulates that *national attachment of immigrants increases in regions with more-inclusive cantonal reception contexts in terms of integration policy and popular attitudes*. With the exception of integration policy, this hypothesis is in line with the results reported in other policy fields (e.g., naturalization and multiculturalism) on how such reception contexts affect national identification. Drawing on the "assimilation theory" argument, according to which identification with an overarching identity emerges only after a certain period of time (Alba and Nee 2009; Manning and Roy 2010; Nandi and Platt 2015; Platt 2014), we deem it necessary to further refine the positive relationship stipulated in hypothesis one. Reception contexts might not have an immediate effect on national identification but rather act as a catalyst. Our second hypothesis, therefore, postulates that *inclusive cantonal reception contexts in terms of integration policy and popular attitudes amplify the positive effect of duration of stay on immigrants' national identification*.

We refrain from formulating explicit hypotheses concerning the homeland identification of immigrants. As the literature suggests, all possible combinations of home- and host-land identification occur in reality (Verkuyten and Martinovic 2012). In other words, an increase in host-land identification does not necessarily imply a decrease in homeland identification. We will nevertheless complement our main analyses on national identification and test, in an exploratory manner, how cantonal reception contexts in terms of cantonal integration policies or popular attitudes affect immigrants' homeland identification.

#### **8.3 Data and Research Design**

To empirically test our theoretical expectations, our research builds on three data sources. The dependent variable, immigrants' feelings of attachment to Switzerland, is drawn from the Migration-Mobility Survey conducted in 2016. This survey clearly surpasses other population surveys in Switzerland because it is the only comprehensive and nationally representative survey focussing explicitly on immigrants that is conducted in six different languages, which also guarantees the inclusion of non-language-assimilated immigrants. Even more importantly for the present study, the large sample of 5855 non-citizen respondents available for our particular analyses guarantees sufficient respondents at the level of the individual cantons. Consequently, the survey allows for individual analyses at the cantonal level, which is hardly feasible with general population surveys. To measure cantonal integration policies, we draw on data collected by Manatschal (2011). Public attitudes towards immigrants at the cantonal level, in turn, are preferably captured by cantonal statistics on right-wing voting rates or direct democratic vote results (more specifically, the vote against "mass immigration"2 from 2014), which are less distorted by social desirability than are survey data on xenophobia.

To measure our main dependent variable, immigrants' feelings of attachment to Switzerland, we use an item from the Migration-Mobility Survey in which respondents had to answer the following statement: "On a scale from 0 (no feeling of attachment) to 7 (strong feeling of attachment), to what extent do you have a feeling of attachment to Switzerland?" This eight-point scale item can be used as a finegrained continuous factor. Promisingly, only a very low number of missing answers (23) is listed, leaving us with a sample of 5855 respondents for our analyses. As the left-skewed distribution in Fig. 8.1 shows, the sample under scrutiny tends to feel rather attached to Switzerland. Fewer than 3% of all respondents indicated that they do not feel attached to Switzerland at all (0), whereas approximately 20% expressed a strong feeling of attachment (7). More than 60% of the questioned individuals ranged their attachment between 4 and 6, whereas only 17.4% indicated a relatively low attachment to Switzerland, ranging from 1 to 3.

Cantonal integration policies serve as the main independent variable in this study. We measure cantonal integration policy using an index built by Manatschal (2011). Conceptually, this index captures subnational integration policies along different policy dimensions that measure the ease or difficulty of accessing civic, political, socio-structural, as well as cultural and religious rights and obligations

<sup>2</sup>This popular initiative that aimed at limiting immigration by means of quotas was accepted by 50.33% of Swiss people and 17 cantons out of 26 in a national vote held on February 9, 2014.

Note: Immigrants' feelings of attachment on an 8-point scale (0–7), mean = 5.83, standard deviation = 1.77

Source: Migration-Mobility Survey 2016 (weighted results)

**Fig. 8.2** Restrictiveness/inclusiveness of cantonal integration policy, Switzerland 2011 Source: Manatschal (2011), own illustration

based on 24 indicators overall. The overall additive Integration Policy Index (IPI) is a continuous variable ranking from −0.54 (most restrictive policy pole) to 0.76 (most inclusive policy pole; for a detailed discussion see Manatschal 2011; Manatschal and Stadelmann-Steffen 2013). The index reveals a clear divide along Switzerland's language regions, as shown in Fig. 8.2.

French-speaking cantons have in general more-inclusive integration policies than do German-speaking cantons, with the canton of Ticino lying in between. This linguistic policy pattern reflects a cultural-historical coinage, with French-speaking cantons being influenced by France's more inclusive *jus soli* citizenship conception and German-speaking cantons (and Ticino) reflecting Germany's (and Italy's) traditionally more-restrictive *jus sanguinis* citizenship tradition (Manatschal 2012). In line with the literature documenting the path dependency of national integration regimes (Brubaker 1992; Favell 2001), cantonal integration policies reflect path dependent, meaning fairly stable, policy frameworks over time (Manatschal 2012; similar findings are reported in Sager and Thomann 2017 for the related policy field of cantonal asylum policy). We also, however, consider testing more recent cantonal policy data. Currently, there is no more-recent comprehensive cantonal integration policy measure available, but there is an index for the single policy field of naturalization policy (Arrighi and Piccoli 2018), which we will use as an alternative cantonal policy measure in this study.

In addition to cantonal integration policy, we use alternative measures that capture the degree of cantonal exclusivity or inclusivity towards immigrants. First, popular attitudes towards immigrants aggregated at the cantonal level (measured in terms of cantonal yes-shares to the national vote against "mass immigration" from 2014) capture the cantonal reception environment. Second, the most recent federal elections in 2015 are used to extract cantonal vote shares of conservative right parties based on the following parties: Swiss People's Party (SVP), the League of Ticino, the Citizens Movement of Geneva, the Swiss Democrats and the Federal Democratic Union (Strijbis 2014). We gathered additional contextual variables at the cantonal level to control for other influencing factors. The unemployment rate for foreigners and the GDP per capita grasp the economic realities of the cantons. The geographic context is approximated by the cantonal degree of urbanity. The foreign-born share reflects the ethnic diversity in each canton. To account for the cultural-linguistic heterogeneity in Switzerland, a dummy variable is included to control for the majority language in a canton (1 = German-speaking, 0 = otherwise).

In addition to these contextual factors, we control for the following variables at the individual level. We gather socio-demographic factors such as age, gender and duration of stay in Switzerland (in years). The question of education has been recoded to obtain a grading from the primary to the tertiary level. Another aspect that is relevant to the feeling of attachment is the ability to speak the local language. Therefore, a six-point scale question assessing individual language skills is included in the analysis. The type of residence permit also matters for this study because it creates different living conditions for non-citizens. The remaining variables address daily life experiences in Switzerland, which likely affect respondents' feelings of attachment to the country. An eleven-point scale question describes respondents' general satisfaction with their decision to move to Switzerland. In addition, a dummy variable is used to measure whether the respondent felt discriminated against in the last 2 years due to his/her origin. Having a social network primarily in Switzerland or in the home country can have divergent effects on the feeling of attachment. Consequently, a five-point scale question is included in our analysis identifying the geographical location of a respondent's social network from "all good friends live in Switzerland" (1) to "all good friends live abroad" (5). More details on all of the variables, data sources, operationalization of the single variables and descriptive statistics can be found in Table 8.2 in the Appendix.

Given the hierarchical structure of the data, we apply a multilevel analysis3 (random intercept models for the baseline model and random slopes and intercepts for the interaction model), implying that individual behaviour can vary between cantons. Cantons represent level-2 units, and individual observations nested within the 26 cantons are the level-1 units. Such multilevel models are moreover very useful if the number of respondents per canton is low because the level-1 estimators are not only rooted in the observations of a particular contextual unit but also 'borrow strength' from the other level-2 units (Steenbergen and Jones 2002). Additionally, these models allow for the modelling of macro-level characteristics (in the present case, the cantonal integration policy and the additional cantonal control variables).

#### **8.4 Empirical Results**

Table 8.1 shows the results of the baseline model. Model 1 confirms that the correlations at the individual level are in line with some of our expectations. Age is positively correlated with the dependent variable. Surprisingly, duration of stay is negatively correlated with the feelings of attachment to Switzerland without being statistically significant. Educational attainment is also not significantly correlated with the outcome. Concerning residence permit, compared with the reference category of Permit C (also called "settlement permit"), Permit B (a residence permit valid for one [non-EU/EFTA] or 5 years [EU/EFTA]) appears to increase feelings of attachment. Interestingly, even non-citizens with the L Permit, a shorter residence permit allowing a year of stay maximum, feel more strongly attached to Switzerland than do members of the reference category. Diplomats, civil servants and their family members (the latter holding a Ci permit) express no greater feelings of attachment to Switzerland than do those with the regular C permit. However, the residence permit estimates are not statistically significant. As expected, individuals who are satisfied with their life in Switzerland and who speak the language fluently tend to feel more strongly attached to the host country. Similarly, if a person has experienced discrimination in the last 2 years, her/his feeling of attachment is significantly reduced. Social networks also matter because immigrants who have all their good friends abroad exhibit significantly lower levels of national identification compared with those with important social ties in Switzerland.

Concerning the cantonal variables, interestingly, none of them significantly affects non-citizens' attachment to Switzerland. With respect to the cantonal reception environment, the direct democratic vote variable (i.e., the yes-share to the "mass immigration initiative", with higher shares representing more-immigranthostile attitudes) and the conservative right share do not appear to have any direct

<sup>3</sup>We consider the dependent variable – feelings of attachment to Switzerland – an integer variable because it is scaled on eight points, which we evaluate as sufficiently fine-grained to conduct such an analysis. We run alternative multilevel models using an ordinal response, which did not alter our results (models not reported but displayed in Appendix).


**Table 8.1** Multilevel regression analyses on feelings of attachment to Switzerland


#### **Table 8.1** (continued)

Note: Unstandardized beta coefficients (standard error in parentheses); \*p < 0.1; \*\*p < 0.05; \*\*\*p < 0.01. Glarus and Appenzell Innerrhoden do not appear in the models because neither had any office runners from a conservative right party. Model 1 uses only random intercepts, whereas Models 2, 3 and 4 use random intercepts and slopes. Variables "GDP p.c.", "Foreigners' unemployment rate" and "Urbanity" were z-transformed prior to modelling. IPI = cantonal Integration Policy Index

Source: Migration-Mobility Survey 2016. Weighted results. The contextual variables "Urbanity", "GDP p.c.", "German-speaking canton", "Yes share of "mass immigration" vote", "Conservative right" have been made available by the Swiss Federal Statistical Office (SFSO) and "Foreigners' unemployment rate" by the SECO.

effect on immigrants' feelings of attachment. Cantonal integration policy points in the expected positive direction; however, this result is not statistically significant. Thus, we find no support for our first hypothesis.

Given that there is no direct effect of cantonal reception contexts on immigrants' national identity, we turn to our second hypothesis and test whether cantonal reception contexts instead have a catalytic function. For instance, Model 1 in Table 8.1 does not account for effective reception context exposure. Thus, this model might not adequately capture the potential policy or public attitude effects. The Migration-Mobility Survey does not include any direct item capturing individuals' perceptions of the cantonal reception context as inclusive or exclusive. Instead, Model 2 in Table 8.1 does indirectly model context exposure for the cantonal integration policy factor by introducing a cross-level interaction term between immigrants' duration of stay and the cantonal Integration Policy Index. As expected, the interaction term is significantly positive. In other words, the positive effect of length of stay on immigrants' feelings of attachment is significantly stronger in cantons with more-inclusive integration policies, which corroborates our more nuanced second hypothesis. This finding suggests that effective exposure to cantonal policies is required for the expected positive effect of inclusive cantonal integration policy on attachment to unfold. Models 3 and 4 in Table 8.1 display the interaction terms between duration of stay and our two additional measures of cantonal reception contexts capturing natives' attitudes towards immigrants. The results are generally consistent with those shown in Models 1 and 2. Concerning the interaction terms with natives' attitudes, only Model 4 displays a slightly negative and statistically significant result. In line with our second hypothesis, the positive effect of duration of stay on attachment to Switzerland is amplified in inclusive cantons, meaning those with lower yes shares of the "mass immigration" vote.

Figure 8.3 illustrates the policy interaction effect with a marginal effect plot that helps to better determine how the two independent variables jointly alter non-citizens' feelings of attachment to Switzerland (Berry et al. 2012; Brambor et al. 2006). The marginal effect of time spent in Switzerland on immigrants' feelings of attachment is insignificant when the IPI is at its lowest value. The increasing slope illustrates how the marginal effect on attachment to Switzerland turns significantly positive in moreinclusive cantonal integration policy contexts. In other words, Fig. 8.3 shows that the positive effect of length of stay on attachment to Switzerland observed in Table 8.1 increases even further in cantons with increasingly liberal integration policies. The same pattern appears in Figs. 8.4 and 8.5, which are based on Models 3 and 4 in Table 8.1. Despite the insignificance of the interaction term per se, Fig. 8.4 shows that up to a conservative right share of 37.1%, which represents a substantial share of 75% of all of the cantons, the interaction term is statistically significant and slightly negative. Concerning integration policies, more-welcoming native attitudes towards immigration (in terms of a lower vote share for the conservative right) appear to amplify the positive effect of length of stay on attachment to Switzerland. The catalyst role of the cantonal reception context is also observable in Fig. 8.5, which follows the same trend as Fig. 8.4. Overall, these results confirm our second hypothesis that more-inclusive/welcoming political contexts act as catalysts by amplifying the positive effect of time spent in Switzerland on immigrants' national attachment.

We conducted several robustness checks that support the main results of our analyses. Because the strongest results emerge from the interaction between period

**Fig. 8.3** Marginal effect of duration of stay on feeling of attachment by IPI Notes: Marginal effect plot based on Model 2 in Table 8.1 Source: Migration-Mobility Survey 2016. Weighted results

**Fig. 8.4** Marginal effect of duration of stay on feeling of attachment by conservative right share Notes: Marginal effect plot based on Model 3 in Table 8.1 Source: Migration-Mobility Survey 2016. Weighted results

**Fig. 8.5** Marginal effect of duration of stay on feeling of attachment by yes share to "mass immigration vote"

Notes: Marginal effect plot based on Model 4 in Table 8.1 Source: Migration-Mobility Survey 2016. Weighted results

of residence and IPI, these checks will focus on the reception context in terms of integration policy.

First, we tested the models with another dependent variable, feelings of attachment to the country of origin (Table 8.3 in the Appendix). In line with the theoretical expectations, respondents with all of their good friends abroad accordingly feel closer to their country of origin than do immigrants who have strong social ties in Switzerland. Non-citizens with no local language knowledge feel more strongly bound to their homeland than do individuals who have better language abilities. Lower satisfaction with the decision to move to Switzerland is further correlated with a stronger feeling of attachment to the country of origin. These last three variables follow a trend opposite to that observed in our main models, indicating clear individual differences between host- and homeland attachments among non-citizens in Switzerland. In contrast to the dependent variable, feelings of attachment to Switzerland, one could expect that inclusive cantonal integration policies coincide with a lower feeling of attachment to the homeland. However, cantonal integration policy is never statistically significant even when interacted with duration of stay in Switzerland. Thus, more-inclusive integration policies do not appear to sway immigrants' feelings away from their country of origin.

The second robustness check used the Citizenship Index (CI) (Arrighi and Piccoli 2018) instead of the Integration Policy Index to measure cantonal policy contexts (Table 8.4 in the Appendix). Before its inclusion in the models, the CI was z-transformed to make it as comparable as possible to the IPI. The two indices are weakly but significantly correlated at 0.40 (Pearson's correlation, p < .05). The CI ranks the Swiss cantons according to the ease or difficulty with which non-citizens can access ordinary naturalization. Model 1 in Table 8.4 shows a statistically significant negative correlation between CI and feelings of attachment. The interaction term in Model 2 in Table 8.4 does not reveal the same positive trend observed in Model 2 in Table 8.1. Indeed, the CI does not act as a positive catalyst by amplifying the positive effect of duration of stay on immigrants' feelings of attachment to Switzerland. Third, to test the catalyst-function of inclusive integration policy with other determinants of national identification, we checked additional interaction terms between the IPI and both educational attainment and satisfaction with the decision to move to Switzerland (Table 8.5 in the Appendix). Only individuals with a secondary education display a statistically significant and positive estimate when interacted with integration policies. Fourth, to account for individual integration into the labour market, a dummy variable (1 = employed, 0 = unemployed) was created (Table 8.6 in the Appendix). Model 1 in Table 8.5 contrasts with our main results because of the negative effect of integration policies on feelings of attachment to Switzerland. However and in line with our main results, the IPI is positively correlated with the outcome variable when interacted with duration of stay in Model 2 in Table 8.6. Interestingly, the R2 of this model is greater than the one drawn from our main analyses, although the inclusion of this new variable leads to approximately 2000 missing observations.

#### **8.5 Conclusion**

This chapter set out to answer the question of how the reception context in Swiss cantons affects immigrants' national identity in terms of feelings of attachment to their destination country. To understand the reception context in Swiss cantons, we considered different manifestations of societal norms of inclusion or exclusion in the Swiss cantons as reflected in cantonal integration policies, public attitudes towards immigrants, and conservative-right vote shares. As the research discussed in this chapter clearly shows, the identity question is complex. Identities are not fixed and unique but fluid and overlapping. This complexity is most likely best captured by the notion that identity equals the sum of all the different attributes and affiliations that make a person unique and that help *identify* this particular person and no one else out of all possible people (Maalouf 1998, p. 16).

Despite this complexity with respect to the identity question, the chapter shows that it can be very revealing to isolate and scrutinize a single aspect of identity among specific individuals, such as immigrants' national identification with their country of residence. This aspect can play an important role as an overarching identity by uniting groups that can have very different ethnic, religious, or cultural identities, as is often true in immigration societies. Despite its integrative force as a "social lubricant" (Miller 1995; Reeskens and Wright 2014) that can bring different people together, diminish social distances, and contribute to social cohesion and mutual trust, we know surprisingly little about whether the reception context can nurture national identification among immigrants. This chapter offers important contributions to the existing research by addressing the neglected question of how reception contexts at the level of Swiss cantons, which represent the immediate everyday life context individuals are exposed to and within which feelings of attachment to the country of residence evolve, affect immigrants' national identity.

From a merely descriptive perspective, non-citizens indicate an overall high attachment to Switzerland. Corroborating assimilation theory (Alba and Nee 2009), the results of our multilevel analyses show that with time in terms of older age immigrants' national identification with Switzerland increases. Further in line with the theoretical expectations, immigrants' national identification increases with higher levels of satisfaction with their decision to move to Switzerland, language ability, and their close social networks being primarily located in Switzerland. Finally, immigrants who have experienced discrimination based on their country of origin over the past 2 years expressed significantly lower levels of national identification.

This last individual finding is particularly relevant for our study; it relates individual experiences to the larger reception context, which is the focus of our chapter, showing that an exclusive social reception context does indeed impair individual attitudes towards the destination country. However, at the contextual level of Swiss cantons, we observe no direct effect of the reception context on immigrants' national identity. Neither cantonal integration policy nor immigrant-sceptic attitudes as expressed in vote results or conservative-right vote shares are per se significantly related to non-citizens' attitudes towards Switzerland. As we argue in this chapter, the reception context might not yield a direct effect on individuals but rather might affect immigrants' national identification in an indirect (i.e., catalytic) way. As the results of our cross-level interaction models suggest, an inclusive reception context in terms of liberal cantonal integration policies and, although to a weaker extent, less xenophobic attitudes and a low share of conservative-right parties amplify the positive effect of time spent in a canton on national identification. In other words, non-citizens' national identification flourishes best over time in those cantons with a receptive integration policy context. Conversely, this process of becoming closer over time to the nation that the immigrants live in appears to completely halt in those cantons with the most exclusive reception contexts.

Relating our findings to the literature, we can say that "feeling welcome at arrival" (Van Hook et al. 2006) is certainly important, but it appears insufficient for immigrants to develop a feeling of attachment to Switzerland. Our findings appear to imply that the "warm handshake" and "cold shoulder" (Reeskens and Wright 2014) need time to be perceived by non-citizens. The same holds for integration policies, which matter only when they are actually perceived, which requires effective exposure to these policies over time. Future studies using more-direct measures for policy exposure and based on more-subtle research designs involving quasiexperiments or panel data are needed to substantiate or refute the causal claims developed in this chapter. Nevertheless, a recent study supports the argument that time and effective exposure to societal norms is needed for exclusive or inclusive reception contexts to be effective. This research shows that inclusive cantonal integration policies can prevent immigrants from emigrating again once they have been exposed to these policies, but they cannot attract them because immigrants are not more likely to migrate to inclusive cantons (Bennour, S. (2017). Do integration policies matter? An exploratory study on foreigners' mobility and cantonal attractiveness in Switzerland. *Presented at the IMISCOE Conference, Rotterdam, 27 June*). In other words, integration policies in particular, and reception contexts more broadly, appear to be neither very effective prior to nor on arrival, but they affect immigrants' attitudes after a certain period of exposure.

Apart from future studies based on more-subtle research designs to prove or refute our preliminary findings, it would also be important to validate these findings using, for instance, more-recent data on integration policy and considering evenmore-nuanced subnational units, such as municipalities. As we argued in this chapter, and in line with the literature documenting the path dependency of national integration regimes (Brubaker 1992; Favell 2001), cantonal integration and migration policies reflect path dependency, meaning fairly stable policy frameworks over time (Manatschal 2012; Sager and Thomann 2017). Consequently, we do not expect these policies to be completely different in 2014 when compared with 2008, which was the reference year of the index used in this chapter. Our robustness check using the narrower but more recent cantonal citizenship law index by Arrighi and Piccoli (2018) appears to confirm this expectation. Nevertheless, it would be interesting to replicate the analysis with more-recent cantonal integration policy data as soon as they are available. The second concern about the level of analysis appears more pertinent. As shown in this contribution, reception contexts vary strongly across Swiss cantons. However, the immediate surroundings people are exposed to on a daily basis are ultimately the local contexts. Given that local reception contexts can vary considerably within the same canton (e.g., rural versus urban municipalities), future research should scrutinize questions about immigrants' national identification from a local perspective, considering the reception context of municipalities.

Concerning the identity question, it would be interesting to further disentangle the general trend emerging from this study. Does the reception context affect the identity of female and male, refugees and highly skilled immigrants, children and adult noncitizens similarly, or are certain groups, for instance, particularly vulnerable groups, more "receptive" to the regional reception context? In a future study, it would also be revealing to connect this research more strongly to the complexity of the identity question. Although the focus of this study was on national identification with the country of destination, the additional robustness checks concerning homeland orientation revealed an interesting pattern. We observed contrasting individual profiles of people who identify with Switzerland (with typically close networks in Switzerland and high levels of satisfaction with the decision to move to Switzerland) and with their country of origin (with rather close social networks abroad and low levels of satisfaction with the decision to move to Switzerland). The cantonal reception context, in turn, did not yield a contrasting effect on homeland identification because it did not significantly alter the homeland identification either directly or over time. Further research is needed to understand whether inclusive cantonal reception contexts and integration policies are conducive to the development of dual-identities ("integration" according to Verkuyten and Martinovic 2012). Although the findings of our study are only a starting point, they clearly highlight the symbolic and catalytic nature and potential for inclusive reception contexts to foster national attachment and eventually social cohesion in contemporary immigration societies.

#### **Appendix**


**Table 8.2** Codebook (variables, descriptive statistics, operationalization and sources)


**Table 8.2** (continued)


#### **Table 8.2** (continued)

**Table 8.3** Multilevel regression analyses, alternative outcome: feeling of attachment to country of origin



#### **Table 8.3** (continued)

Note: Unstandardized beta coefficients (standard error in parentheses); \*p < 0.1; \*\*p < 0.05; \*\*\*p < 0.01. Glarus and Appenzell Innerrhoden do not appear in the models because neither had any office runners from a conservative right party. Model 1 uses only random intercepts, whereas Model 2 uses random intercepts and slopes. Variables "GDP p.c.", "Foreigners' unemployment rate" and "Urbanity" were z-transformed prior to modelling. IPI = cantonal Integration Policy Index

Source: Migration-Mobility Survey 2016. Weighted results. The contextual variables "Urbanity", "GDP p.c.", "German-speaking canton", "Yes share of "mass immigration" vote", "Conservative right" have been made available by the Swiss Federal Statistical Office (SFSO) and "Foreigners' unemployment rate" by the SECO


**Table 8.4** Multilevel regression analyses, alternative index: citizenship index


#### **Table 8.4** (continued)

Note: Unstandardized beta coefficients (standard error in parentheses); \*p < 0.1; \*\*p < 0.05; \*\*\*p < 0.01. Glarus and Appenzell Innerrhoden do not appear in the models because neither had any office runners from a conservative right party. Model 1 uses only random intercepts, whereas Model 2 uses random intercepts and slopes. Variables "GDP p.c.", "Foreigners' unemployment rate", "Urbanity" and "Citizenship Index" were z-transformed prior to modelling

Source: Migration-Mobility Survey 2016 (weighted results). The contextual variables "Urbanity", "GDP p.c.", "German-speaking canton", "Yes share of "mass immigration" vote", "Conservative right" have been made available by the Swiss Federal Statistical Office (SFSO) and "Foreigners' unemployment rate" by the SECO


**Table 8.5** Multilevel regression analyses, alternative interaction terms


#### **Table 8.5** (continued)

214


#### **Table 8.5** (continued)

Note: Unstandardized beta coefficients (standard error in parentheses); \*p < 0.1; \*\*p < 0.05; \*\*\*p < 0.01. Glarus and Appenzell Innerrhoden do not appear in the models because neither had any office runners from a conservative right party. Model 1 uses only random intercepts, whereas Model 2 uses random intercepts and slopes. Variables "GDP p.c.", "Foreigners' unemployment rate" and "Urbanity" were z-transformed prior to modelling. IPI = cantonal Integration Policy Index.

Source: Migration-Mobility Survey 2016 (weighted results). The contextual variables "Urbanity", "GDP p.c.", "German-speaking canton", "Yes share of "mass immigration" vote", "Conservative right" have been made available by the Swiss Federal Statistical Office (SFSO) and "Foreigners' unemployment rate" by the SECO.


**Table 8.6** Multilevel regression analyses, control for individual employment


#### **Table 8.6** (continued)



Note: Unstandardized beta coefficients (standard error in parentheses); \*p < 0.1; \*\*p < 0.05; \*\*\*p < 0.01. Glarus and Appenzell Innerrhoden do not appear in the models because neither had any office runners from a conservative right party. Model 1 uses only random intercepts, whereas Model 2 uses random intercepts and slopes. Variables "GDP p.c.", "Foreigners' unemployment rate" and "Urbanity" were z-transformed prior to modelling. IPI = cantonal Integration Policy Index

Source: Migration-Mobility Survey 2016 (weighted results). The contextual variables "Urbanity", "GDP p.c.", "German-speaking canton", "Yes share of "mass immigration" vote", "Conservative right" have been made available by the Swiss Federal Statistical Office (SFSO) and "Foreigners' unemployment rate" by the SECO

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Chapter 9 Who Feels Disadvantaged? Reporting Discrimination in Surveys**

**Daniel Auer and Didier Ruedin**

#### **9.1 Introduction**

In a world increasingly characterized by growing ethnic diversity, questions of inter-group relationships and social cohesion find their way into the political debate (Green-Pedersen and Otjes 2017; Van der Brug et al. 2015). When individuals are treated differently along ethnic, cultural or legal lines depending upon their group membership, *discrimination* is a frequently used and arguably often correct label. However, between and even within disciplines of social science, there is little agreement on what exactly discrimination is or, moreover, on how it can be measured. Cleavages in the understanding of discrimination become particularly visible when the discriminating person can provide a plausible justification for their action (e.g., blame shifting; see Campbell et al. (2012)) or when victims of discrimination are either unaware of their disadvantage or do not feel disadvantaged at all. In this chapter, we focus on the *perception* of discrimination because if policymakers try to maximize individual well-being, they should find it valuable to know who feels discriminated against and under what circumstances. Moreover, knowledge about the drivers of perceived discrimination and the reporting of discrimination when

D. Auer (\*) WZB Berlin Social Science Center, Berlin, Germany

nccr – on the move, Neuchâtel, Switzerland e-mail: Daniel.Auer@wzb.eu

D. Ruedin University of Neuchâtel, Neuchâtel, Switzerland

University of the Witwatersrand, Johannesburg, South Africa e-mail: Didier.Ruedin@unine.ch

© The Author(s) 2019 221 I. Steiner, P. Wanner (eds.), *Migrants and Expats: The Swiss Migration and Mobility Nexus*, IMISCOE Research Series, https://doi.org/10.1007/978-3-030-05671-1\_9

Authors are listed in alphabetical order and contributed equally to the design, analysis, and writing of this chapter. Replication material is available at <osf.io/b2zf6>.

asked in a survey enables policymakers to implement tailored means to diminish perceptions of unfair treatment and discrimination. Indeed, we argue that perceptions of discrimination are often more important for the well-being of people than 'objective' discrimination, although the two will be related to one another.

For discrimination to occur, we need at least two actors. One of these actors unfairly treats the other based on an irrelevant criterion such as ethnicity, country of origin, or gender. Theories of justice provide answers concerning what constitutes unfair treatment and what criteria can be considered irrelevant (e.g., Rawls 1999; Sen 2009). For instance, Rawls introduces the concept of the 'veil of ignorance' to highlight that criteria that have no bearing on people properly performing their job cannot be considered just – assuming a hiring situation in this case. For instance, the skin colour of a worker has no bearing on the productivity of the worker, so selecting workers based on skin colour is unjust. In contrast, selecting manual workers on their dexterity is just because it affects the productivity of the worker. In other words, if a criterion for selecting a worker is irrelevant for that person's productivity, we can say that the differential treatment based on that criterion is unfair and discrimination occurs. Objectively speaking, discrimination does not require witting action; nor does the person affected have to perceive it as unfair. In this chapter, we present a non-exhaustive list of (unintentional) factors that influence individual awareness of discrimination or the perception of unequal treatment as unjust. This ostensible unawareness of discrimination has important consequences for discrimination research and possible policy responses to discriminatory behaviour because nescience about being the victim of discrimination blurs our understanding about the extent of unfair treatment.

At the same time, we argue that social scientists and policymakers should pay attention to *perceptions* of discrimination because these are linked to well-being, poor health, and ultimately social cohesion (Versey and Curtin 2016; Simona et al. 2015; Hanefeld et al. 2017). We therefore create a model to predict the individual propensity to report discrimination when asked in a survey. Conceptually, people can feel discriminated against in situations in which there is no objective discrimination – consider a situation in which the treatment of different groups is objectively the same – or they might not feel discriminated against in situations in which there is objective discrimination. The latter can occur when individuals do not perceive differences in treatment or do not consider these differences unjust, for instance, because they have internalized social roles that naturalize such differences.

We argue and demonstrate that, among recent immigrants in Switzerland, the perception and reporting of discrimination relates to reported difficulties during immigration, how immigrants succeed economically, how ethnic minority groups are included in politics in the country of origin, and other individual characteristics. We show that immigrants differentiate between discrimination at work and in public, indicating that immigrants perceive greater discrimination when it comes unexpectedly and that reporting discrimination in a survey is the result of various individual circumstances rather than the haphazard response to a yes-or-no question. We thereby empirically picture aspects of the Migration-Mobility Nexus (compare Chap. 1 in this volume) by highlighting that the exclusionary logic of the societal sphere is substantially influenced by the inclusionary logic of the economic sphere. In other words, whether discriminatory treatment of immigrants becomes visible depends in part upon an implicit trade-off with economic gains. Moreover, we stress that societal exclusion along the lines of the Swiss dual regime of migration and mobility (see Sect. 9.2) is decisively blurred by individual contexts. Although studies have shown that the dual immigration regime distinguishing between EU/EFTA and non-EU/EFTA nationals provides a powerful proxy for societal exclusion (e.g., Auer and Fossati 2018), we stress that individual perception of attachment and welcoming can mitigate or even counter statements about discrimination that appear to be valid on the aggregate level.

#### **9.2 Context: Immigration to Switzerland**

Despite a trend towards more-expansive policies (Ruedin et al. 2015), Switzerland has relatively restrictive naturalization policies that contribute to a high share of foreign citizens. Traditionally, immigration was regulated restrictively as a function of the needs of the economy (Piguet 2004). After World War II, Switzerland experienced a substantive growth in the immigrant population, helped by a booming economy and guest worker recruitment in Italy, Spain, Portugal, Turkey, and Yugoslavia. To prevent workers from settling, the guest-worker programmes were designed to rotate workers. Increasing competition for these workers with other Western European countries, combined with growing pressure to respect worker rights by civil society and international organizations, however, led to family reunification and settlement of a growing number of immigrants (e.g., Ruedin 2015). After a gradual loosening of its immigration policies, Switzerland introduced a tiered labour market model in the early 1990s (Becker et al. 2008). Swiss immigration policy sought to balance the diversification of migration due to emerging trends of globalization with the highly emotional public debate around *Überfremdung* – a "fear of domination by foreign influences", implying too many foreigners and foreigners who are "too foreign" in the sense of cultural distance. Immigrants from countries with perceived cultural closeness (Western Europe, USA, Canada, Australia and New Zealand) were granted preferential access. In the late 1990s, the wars in (former) Yugoslavia brought an unprecedented number of refugees to Switzerland, partly an indirect consequence of existing ties through guest-worker programmes. The "three-circles" system was reduced to two tiers (European Union vs. rest of the world) in 2002, which led to high inflows from European countries, particularly from Portugal and Germany (Becker et al. 2008).1 The presence of

<sup>1</sup>When nationality is taken as the criterion – the data that are readily available over time – the growth in the immigrant population post-1970s correlates strongly with its diversity. Using the Herfindahl index to express the diversity of the population, we observe an increase from 0.07 in the mid-1970s to 0.25 in the early 2010s. After the early 2010s, the Herfindahl index stopped increasing as much as it did previously.

immigrants – foreign citizens and their children – has long been politicized (Ruedin and D'Amato 2015). As in other countries, a significant part of the population has negative attitudes towards foreigners (Pecoraro and Ruedin 2016). However, Switzerland also has a long tradition of human rights and justice as exemplified by the International Red Cross (Ruedin and D'Amato 2015).

This contradictory understanding as a country inherently shaped by and being open towards immigration on the one hand and a subordination of restrictive immigration policies to an economic rationale on the other hand make Switzerland an interesting place to conduct research on discrimination. In the public sphere, racism is commonplace (Ruedin 2015; Efionayi-Mäder and Ruedin 2017). Exploiting a natural-experimental setting in which municipalities voted over naturalization applications, Hainmueller and Hangartner (2013) show that immigrants from former Yugoslavia and Turkey had a substantially lower likelihood of being granted Swiss citizenship than equivalent immigrants who originated from northern or western European countries. Moreover, Switzerland is no exception concerning discrimination against immigrants in the labour market, as has been shown in a recent meta-analysis of experimental evidence (Zschirnt and Ruedin 2016). The systematic disadvantage of immigrants in the hiring process (Fibbi et al. 2006 and Zschirnt, E. (2018, March 23). *Ethnic discrimination in the Swiss (German) labour market – first results from a correspondence test*. Presented at the NCCR Research Day #5, Neuchâtel) is also reflected in sharp native-immigrant wage differentials (Henneberger and Sousa-Poza 1998; Iseni et al. 2014) and longer unemployment durations of foreigners (Auer et al. 2018). Recently, Auer and Liechti argued in their presentation *What does effect-heterogeneity of active labour market programmes reveal about discrimination?* at the annual IMISCOE Conference 2018 in Barcelona that statistical discrimination by employers dominates over taste-based discrimination by showing that immigrants' potential benefits from participation in active labour market programmes exceeds those of natives. According to the authors, such a finding can only materialize if employers allow their stereotypes to be influenced. Such a "pragmatic approach" towards discrimination is in line with Switzerland's largely economy-driven immigration policies and reflects the evidence of surprisingly deliberate statistical discrimination reported in the social domain in Switzerland (Fibbi et al. 2018).

#### **9.3 Theoretical Considerations**

There is extensive literature on ethnic discrimination and a clear recognition that perceptions of *discrimination* do not necessarily match "*objective*" measures of discrimination (Blank et al. 2004). However, no agreed-upon understanding exists of why some individuals feel more discriminated against than others do or are more likely to report discrimination in a survey when asked. Existing research on perceptions of discrimination has highlighted variances within ethnic groups but has provided mixed evidence with respect to individual socio-demographic variables (Zainiddinov 2016; Alanya et al. 2017; Flores 2015; McGinnity and Gijsberts 2016; Hopkins et al. 2016). Zainiddinov (2016) highlights that most existing research on perceived discrimination focusses on Black Americans, although more-recent research also includes other minority groups, notably Hispanics and Asians. People with lower socio-economic status tend to report more discrimination, as do older respondents. The evidence for gender differences is mixed, with most studies finding no substantive differences. The association between levels of education or income and perceived discrimination has been reported to be positive, negative, or indeed curvilinear – which we take as inconclusive evidence, as does Zainiddinov (2016). Alanya et al. (2017) examine perceived discrimination in several Western European cities, highlighting that the experience of immigrant integration might also matter, although they find substantive differences between the cities examined. In this chapter, we expand on this literature by highlighting plausible drivers of such *perceptions*, although we leave the development of a full theory for future research (compare Swedberg 2014). In our view, such a theory should start with the threat framework generally used as an explanation for intercultural relationships and attitudes towards foreigners (Blumer 1958; Bobo and Hutchings 1996). In the threat framework, immigrants are perceived as unwanted competitors for scarce resources such as jobs or social benefits. In a zero-sum understanding of these resources, the presence and size of different groups in society is perceived as a threat and can lead to discrimination. Although notions of threat are usually applied to the majority population, we focus here on immigrants at the receiving end.

In virtually all societies, there are accounts of "objective" discrimination being overseen, ignored, or "justified" by individuals. That is, people do not identify inequality in the first place or do not perceive unequal treatment as discriminatory ranging from gender discrimination to taxation justice, access to goods and services, hiring, or less materialistic aspects such as active participation of persons with special needs in sports clubs. We argue that the discrimination against immigrants follows the same mechanism. Although immigrants and their descendants are often treated to their disadvantage in labour markets or elsewhere (Heath and Cheung 2007; Neumark 2016; Zschirnt and Ruedin 2016; Oswald et al. 2013), their individual *perception* of discrimination can deviate substantially from objective measures (Hopkins et al. 2016). We argue that discrimination can be analysed as both an *objective*, systematic unfair differential treatment of two or more groups and as a subjective *perception* of such a treatment, in which case the latter is influenced by individual characteristics within three key areas among others. We explicitly leave the influence of context and the interaction between context and individual characteristics to future research. Hopkins et al. (2016) demonstrate that immigrants' perceptions of discrimination in the USA vary very little across localities, suggesting that the influence of context alone can be more intricate than a direct relationship – or indeed that individual-level factors such as we examine in this chapter are more important for perceptions of discrimination and for reporting such perceptions in a survey.

*First*, we expect that people who are more likely to be discriminated against (objectively) are more likely to report the experience of discrimination in a survey when asked. This area is close to conventional residual studies that highlight the unexplained gap, for instance in labour market outcomes, after controlling for a number of socio-economic characteristics (e.g., Auer et al. 2017; Ballarino and Panichella 2015). Although individual traits such as country of origin affect the propensity to be exposed to discriminatory behaviour, these *objective* (read: general) drivers of discrimination are external – not within the scope of individuals to change. This point is likely most apparent in the case of racial and ethnic differences such as skin colour (Zainiddinov 2016).

*Second*, because of material gains and improved labour market outcomes, people might *accept* a certain degree of discrimination. Here we assume that some immigrants might come prepared to face unjust treatment and are willing to tolerate the unpleasant experience as part of the "costs" of immigration that are compensated for by the (financial) benefits of immigration. Similarly, immigrants might accept discrimination as a cost they pay so that their children can reap the benefits of immigration. It is also possible that immigrants internalize roles in which differential treatment appears "natural" and therefore accept that treatment. To an extent, this internalization of social roles can be considered a cost to immigrants. Because longterm strategies and hopes of a better future are difficult to capture, we focus here on more-immediate gains. It follows that immigrants whose labour market situation has improved after migration can be expected to be more willing to *accept* differential treatment and are hence less likely to report discrimination in a survey. This argument is in line with findings by Zainiddinov (2016) that older Muslims in the United States are less likely to report discrimination; they might be "tolerating" unfair treatment and accept whatever is necessary to the extent that their children born in the country are treated equally and benefit from migration to the United States. Importantly, *acceptance* implies some form of trade-off or bargain between current or future economic (or social) benefits at the expense of potentially unfair treatment. Thus, not perceiving or reporting discrimination because a person accepts or tolerates it stands in opposition to the subsequent expectation according to which individuals can develop a genuinely positive feeling towards the host country without any trade-off attached.

*Third*, we expect *attachment* to the country of destination in general to be associated with the likelihood of reporting discrimination when asked in a survey. People who are positive about the country of destination arguably possess a higher "tolerance" concerning critical aspects about the society of the host country and, hence, are less likely to perceive treatment as unjust and discriminatory. In contrast to the influence of acceptance as presented above, attachment to the country of destination does not necessarily entail economic gains. Rather, people can feel attached to or identify with a country of destination if their habitus and worldview are compatible with the dominant cultural aspects of the destination country. This point suggests that socialization in the country of origin shapes how the situation in the country of destination is perceived and hence how discrimination is perceived and reported when investigated in a survey. However, attachment and discrimination are likely to be interdependent; people who are discriminated against are less likely to feel attached to the country of destination, but those who feel

attached to the country of destination are less likely to perceive acts as discriminatory and report discrimination in a survey. It is inherently difficult to assess which of the two concepts drives the other, particularly with cross-sectional data. To (partly) overcome this issue, as we will elaborate below, we differentiate between two types of attachment proxies; on the one hand, there is straightforward reporting of feelings of attachment. On the other hand, we introduce variables that are associated with feelings of attachment but that capture situations prior to or during the migration process, including socialization in the country of origin. By definition, these elements are not or are less likely to be confounded with possible subsequent incidents of discrimination in the host country and, hence, are less prone to bias the relationship between attachment and reporting of discrimination. In other words, positive or negative experiences during the migration process should be related to the reporting of unfair treatment in the host country without being biased by an already developed feeling of attachment that could blur the perception of discrimination or by a form of alienation from the host country due to incidents of discrimination.

#### **9.4 Data and Method**

To examine who is more likely to feel discriminated against and how potential drivers affect perceptions of discrimination, we use new data from the Migration-Mobility Survey that cover detailed information on socio-economic traits, subjective well-being, and the migration experience of approximately 6000 recent immigrants in Switzerland (Migration-Mobility Survey 2016; see Chap. 2). We use three direct questions about the experience of discrimination in different situations. A general measure of perceived discrimination asks, "Have you experienced situations of prejudice or discrimination in Switzerland in the last 24 months?" Two further questions ask specifically about experienced discrimination in the workplace (including education facilities) or during leisure time and in the more general public sphere.2

We use a range of predictor variables to capture the three stated mechanisms for differences in perceptions or reporting discrimination: *drivers*, *acceptance*, and *attachment*. These variables are presented in Table 9.1; the question wording can be found in Table 9.4 in the Appendix.

Initially, as shown in Table 9.1, individual propensity to being discriminated against varies with group membership (e.g., Auer et al. 2018; Ebner and Helbling 2016). We expect that individuals who belong to groups objectively more often discriminated against report discrimination more often. Such *drivers* typically

<sup>2</sup>These questions can entail individually experienced discrimination and situations in which someone in the vicinity of the respondent was discriminated against. Although such experiences could bias the effect of individual-level drivers of discrimination such as education, they should not affect individual perception of discrimination per se.


**Table 9.1** Predictor variables and expectations

include local language proficiency (depending upon the Swiss language region), which provides a key driver of integration success and is expected to reduce exposure to discrimination (Föbker and Imani 2017; for the labour market, see Auer 2018). Being born within the European Union has been shown to be a powerful approximation for "cultural proximity" (Ruedin 2018b). At the same time, existing networks among family and friends or among work colleagues have been shown to be beneficial in various ways, ranging from increased well-being (Portela et al. 2013) to higher labour-market access probability (Bonoli and Turtschi 2015) and less exposure to discrimination (Jasinskaja-Lahti et al. 2006).3 Moreover, we consider geographical origin by including region-fixed effects in all models. We thereby consider the literature on ethnic hierarchies or ethnic rankings based on socio-economic differences and ethnic traits among immigrants (Hagendoorn 1995).

*Drivers* of discrimination are usually observable by other people, for instance in job applications, during administrative dealings but also in private everyday interactions. Hence, they have the potential to generally increase the propensity to be discriminated against. If we assume that people who are unfairly treated on average also have a higher propensity to report such treatment, we must control for these individual characteristics in our statistical models to render the remaining – unobservable – traits meaningful determinants of perceived discrimination. One issue with testing what drives individual perception of discrimination is the direction of the relationship. Does a happy immigrant experience less discrimination or is a lessdiscriminated-against immigrant happier? We try to minimize this bias by focussing on aspects that occurred prior to migrating or during migration to Switzerland in our main analysis. Bearing in mind the possibility of a biased reflection of past events, which is inherent to all survey information, reported discrimination can only be the result of these aspects and not vice versa.

<sup>3</sup>We refrain from defining a specific pathway because both networks as a decreasing factor for discrimination and discrimination as a decreasing factor for networks are possible.

Hence, we capture *acceptance* of discrimination with changes in individual economic opportunities compared with the situation in the country of origin. We argue that if the job situation of a person has improved since migrating to Switzerland, incidents of discrimination are not perceived as such as often (or perceived as less severe) because of the compensating ability of economic success. The same mechanism applies for feeling right for a job in terms of qualification.

Although accepting differential treatment is the result of a biased perception due to (material) gains, elements of *attachment* can reduce individual propensity to perceive and report discrimination. We argue that a lack of past socialization in the country of origin in terms of ethno-cultural diversity (approximated by low representation of ethnic minority groups in the national legislature; Ruedin 2009), decreases discrimination perceptions. The intuition is that if individuals have been confronted with diversity, the promotion of equal opportunities or antidiscrimination efforts in their country of origin, they are likely to have established some form of awareness of these issues (Ziller 2014). Subsequently, they should be more sensitive in terms of noticing discrimination in the country of destination. Conversely, the absence of adequate minority representation in the country of origin increases the propensity to be unaware or ignorant of issues of inequality and discrimination. Furthermore, a "smooth" migration process, that is, the reported absence of issues when entering Switzerland and when registering at various immigration offices, can shape the overall notion about feeling "welcome" in the host country. If this statement is true, individuals should identify more with Swiss society and perceive discriminatory patterns less often or as less "severe". We measure this aspect with a variable on satisfaction with the decision to migrate to Switzerland. Finally, we consider three aspects of potentially very strong confounders of discrimination perception. As we will briefly elaborate below, however, these aspects should be interpreted with special caution due to a direct reciprocal relationship with experienced discrimination. We capture whether the person feels attached to Switzerland. Intuitively, someone who lacks a certain subjective feeling of attachment is more likely to perceive situations as particularly inconvenient and, hence, reports discrimination more often. Conversely, attachment to Switzerland creates a 'blind spot' with respect to discrimination.

Table 9.2 presents descriptive statistics of the outcome and predictor variables. Overall, 35% of the recently arrived immigrants in Switzerland report having experienced situations of prejudice or discrimination in Switzerland in the preceding 2 years. The reported discrimination at work and in public situations is similar. With respect to the predictor variables, more than one-half of the respondents do not speak the local language. Approximately one-half of the immigrants come from outside the EU, and they often report neither personal (73%) nor professional (60%) network links to Switzerland prior to arriving (see Table 9.4 in the Appendix for question wordings and operationalization). On average, respondents are very satisfied with their decision to migrate (8.3 of 10). At the same time, feelings of attachment to Switzerland are largely positive but less pronounced (4.8 of 7). These


**Table 9.2** Descriptive Statistics

Note: The weighted mean, minimum, maximum, and median for each variable are given. For binary variables such as 'Female', the mean corresponds to the percentage

a 'worsened substantially' (=1) to 'improved substantially' (=5)

b 'Assembly representation' is 1 if the share of ethnic groups in the national legislature is perfectly proportional to the share of ethnic groups in the population, and 0 if it is perfectly disproportional c 'very problematic' (=0) to 'not problematic at all' (=7)

d 'not satisfied' (=0) to 'completely satisfied' (=10)

e 'no feeling' (=0) to 'strong feeling' (=7)

Source: Migration-Mobility Survey 2016. Weighted results

figures correspond roughly with the findings by Geurts and Lubbers (2017) that approximately 60% of immigrants in the Netherlands are positive about their migration decision and intend to stay permanently.

In terms of analytical strategy, we predict the likelihood of reporting discrimination in *general*, at *work*, and during *public* activities using three outcome variables in separate models. We depart from a baseline logistic regression model that captures fundamental individual *drivers* and includes fixed effects for the geographical origin. Additionally, we account for certain individual traits that might affect both the perception of discrimination and actually being discriminated against. These baseline characteristics include gender, age at arrival, the duration of stay in the host

country, and employment status (see, for instance, Zainiddinov 2016 and Alanya et al. 2017). Subsequently, we add variables to capture the *acceptance* mechanisms in the model. We add a battery of *attachment* variables that influence perception but are more sensitive with respect to the potential issue of reverse causality. Ex-post empirical analyses of survey responses are always biased by subjective interpretations of events, which is particularly problematic for establishing a (causal) mechanism. Does an immigrant who feels attached to Switzerland report less discrimination, or does an immigrant who is less discriminated against feel more attached to the country? We try to circumvent this reverse pathway issue by focussing on evaluations of past events. For instance, at the time when someone decided to migrate to Switzerland for employment, he or she was not exposed to discrimination by Swiss society. Hence, it is safe to assume that a certain response might be biased by subjective interpretation but not by our factor of interest, that is, discrimination. This assumption should hold for aspects with less clear-cut temporal distinction, such as reported issues during the migration process. It is possible that the newly arrived person has previously been exposed to discrimination; however, we consider this probability small and, not least, its effect limited given that respondents on average have remained in Switzerland more than 5 years. For the parts of the variables capturing *attachment*, however, the direction of the relationship is less clear. Therefore, we interpret these factors with special caution.

#### **9.5 Findings**

In this section, we highlight the most important findings based on the logistic regression models presented in Table 9.3. If not stated otherwise, we refer to the full regression models on discrimination in general, at work, and in public situations presented in Table 9.3. 4 Although most variables thought to be *drivers* were found to be weak predictors for reporting discrimination, the geographical origin indicates a high relevance of ethnic traits, with different statistical effects between discrimination at work or in public situations. Elements of both *acceptance* and *attachment* play a substantial role with respect to perceiving differential treatment as unfair.

With the exception of origin, the drivers of experiencing discrimination are only weak predictors. Variables such as gender, less education, or a lack of language proficiency are typical drivers of 'observable' discrimination; that is, they explain a substantial part of differences in labour market outcomes in terms of wages or unemployment duration (e.g., Auer et al. 2017; Koopmans 2016) and in other markets such as housing (e.g., Carlsson and Eriksson 2014). The results show that most of these variables have limited influence on individual perception of discrimination. For instance, immigrants who have been in Switzerland for a longer period are more likely to report discrimination. The predicted probability of experiencing discrimination of a person who stayed for 2 years is 49%, compared with 59% for an other-

<sup>4</sup>Table 9.5 in the Appendix presents partial models.


**Table 9.3** Regression results


**Table 9.3** (continued)

Notes: \*p < 0.1; \*\*p < 0.05; \*\*\*p < 0.01; logit coefficients, standard errors in parentheses. Outcome variables: experience of discrimination, experience of discrimination at work, experience of discrimination in public; shown are the log odds with predicted probabilities provided in the text. See Table 9.5 in the Appendix for additional models

Source: Migration-Mobility Survey 2016. Weighted results

wise equivalent person with a stay of 10 years.5 These results could be due to an increasing likelihood of experiencing an incident of discrimination with duration of stay. At the same time, it is possible that the effect of acceptance (that is, an improvement in a person's economic situation as discussed below) diminishes over time and less recent immigrants start to notice differential treatment and perceive it as unfair. In contrast, whereas a lack of language proficiency is associated with a higher propensity to report discrimination, the substantive difference is limited (predicted probabilities of 53 and 45%). It is possible that this result is biased by highly skilled immigrants working in international businesses and organizations, in which it is often more common to communicate in English rather than in one of the Swiss languages. Moreover, it is possible that a complete lack of language skills constrains the capability to notice discrimination, particularly in the spheres of social mistreatment. If we interact language ability with level of education, the sign of the coefficient suggests that less-educated immigrants without local language skills might indeed be more likely to report discrimination, although the standard errors around the estimate are rather large, particularly for discrimination in public. Although less-educated immigrants might be concentrated in economic sectors with many immigrants – perhaps objectively preferred in hiring (Auer et al. 2018) – these immigrants can however experience discrimination during their workday or outside work. Moreover, immigrants who report lacking a supportive professional network are substantially more likely to report discrimination in general (predicted probabilities of 53 and 59%). Intuitively, being unemployed only affects discrimination in general and in public situations, but not at work. The negative coefficient could be explained by a lower general activity of the unemployed (first highlighted in the well-known study by Jahoda and Zeisel 1933), that is, the potential number of situ-

<sup>5</sup>To calculate predicted probabilities, we set all binary variables to 0, all continuous variables to their mean and the region to neighbouring countries. We use discrimination in general as the outcome variable.

ations in which discrimination can occur diminishes when people are unemployed (predicted probabilities of 48% for unemployed people compared with 53% otherwise).

The apparent key determinant among the drivers is the immigrant's origin. We observe a constant increase in the propensity to report general discrimination when moving from immigrants originating from European countries (predicted probability of 53%) to North-American (57%), South-American (58%), and West-African countries (71%). This finding is in line with a large corpus of literature on the graded effects due to so-called ethnic rankings (Hagendoorn 1995; Zschirnt and Ruedin 2016). In contrast, the predicted probability of reporting discrimination for immigrants from Asian countries is less than what can be expected from ethnic hierarchies (46%), and we expect this probability to reflect different response behaviour with respect to the outcome variable.

Interestingly, the statistical effect of origin differs between reported discrimination in the workplace and in private situations. Although compared with European immigrants, immigrants from the Americas are significantly less likely to report discrimination at work, they are more likely to report discrimination in public situations; in general, the results are the opposite for immigrants from 'more distant' African countries. This result might capture effects of direct competition in the labour market versus working in occupations "reserved" for immigrants from culturally more distant countries.

In general, we find that individuals distinguish, to a surprising degree, between discrimination within and outside the work environment. For instance, respondents often report discrimination in the workplace but not in a public environment when they are less educated and lack local language skills. In contrast, age and duration of stay affect discrimination reporting in the public rather than the work sphere. Improvements in employment relative to the pre-migration situation reduce perceived discrimination in the workplace, whereas minority representation in the country of origin affects discrimination in the public context but not at work.

The reported experience during migration and attachment appear decisive with respect to reporting discrimination. The items of the *attachment* category are strong predictors of perceived discrimination. Both a currently higher level of attachment to the Swiss society (that is, "a certain feeling of attachment to Switzerland" – predicted probabilities are 69 and 35% for the minimum and maximum level of attachment observed, respectively) and positive past experiences during the migration process (that is, "no difficulties during the migration process" – predicted probabilities are 39 and 73% for the minimum and maximum level of attachment observed, respectively; for "overall satisfaction with the decision to migrate to Switzerland", predicted probabilities are 48 and 76% for the minimum and maximum level of attachment observed, respectively) significantly reduce the propensity to report incidents of unfair treatment. In Table 9.5 we include additional models to ascertain whether the reported statistical effects are robust against model specifications.

#### **9.6 Discussion and Conclusion**

In this chapter, we have examined who among recent immigrants to Switzerland is more likely to report discrimination in a survey when asked. After controlling for "observable" drivers, discrimination should occur, on average, with the same probability across individuals. However, immigrants have had different "unobservable" experiences and come with different socialization and notions related to their migration that are likely to affect their perception of discrimination – presumably beyond its actual occurrence. The findings in this chapter indicate that such elements of acceptance and attachment influence the perception of differential treatment as discriminatory in an essential way. We thereby also highlight the interdependency between different aspects of the Migration-Mobility Nexus, particularly that the exclusionary logic of the societal sphere is substantially influenced by the inclusionary logic of the economic sphere and that robust findings of discrimination along the Swiss dual migration and mobility regime at the aggregate level (i.e., EU/EFTA vs. non-EU/EFTA nationals) can be blurred by variation in individual perceptions and feelings of attachment to the host society. Although improvements in a person's economic situation – we refer to a higher acceptance – mitigate perceived discrimination, issues during the migration process or a lack of attachment – that is, by immigrants we assume to "anticipate" further problems ahead – are associated with substantially higher levels of perceived discrimination. This result indicates that studies capturing objective levels of discrimination, such as field experiments on hiring (Zschirnt and Ruedin 2016), can be inadequate to capture perceptions of discrimination relevant to well-being.

Despite the usual limitations and calls for cautious interpretation that we elaborate below, the findings are highly plausible and entail at least two main implications for research and for policymaking. First, policies focussing on "objective" discrimination might fail to address social cohesion – and, indirectly, the politicization of immigrants (Van der Brug et al. 2015). If discrimination is a blind spot in the perception of both policymakers and victims (and not least offenders), not only legal frameworks to mitigate unfair treatment but also other assessments such as surveys fail to capture the true extent of discriminatory behaviour and might ultimately be ineffective.

Second, research should generally focus more on perceptions of discrimination. From a purely empirical perspective, we can conclude that contemporary individuallevel measures will most likely fail to capture discrimination that goes beyond differences in wages for individuals with equal skills, for instance. In other words, analyses such as residual studies might provide us with valuable insights about observable disadvantages for observable individual traits, but we do not learn much about whether disadvantaged individuals *perceive* their lower wages as due to discrimination in society, to stay with the example. If they do, we must follow up and investigate why they fail to change their situation and what could be done from a policy perspective. If they do not, we are left wondering what led to a situation in which inequality that is not grounded on objective criteria is not perceived as unfair. Both aspects must be properly investigated and understood.

In this respect, one limitation of this study is also one of its key findings. It is safe to assume that reporting of discrimination is biased; who reports being discriminated against in a survey is a function of not only actual discrimination but also a complex array of individual characteristics. In this analysis, we tried to enumerate this selection bias. If we consider policy implications, establishing "objective" discrimination is not sufficient because from the perspective of the victims, what matters is the perception. In other words, we can imagine a world in which many are discriminated against according to our "objective" criteria but who do not feel that way (and the opposite – no objective discrimination, but feelings of discrimination). If individual well-being and social cohesion are the criteria of interest, we must take perceptions seriously.

Considering selection bias, we must assume that perception of discrimination correlates with reporting of discrimination. Although the assumption is a common one (because it is not verifiable) in all surveys – not only those on discrimination – the correlation might be blurred by increasing sensitivity to the question asked (Rosenfeld et al. 2015). This point is particularly true in situations in which there might be consequences for the respondent, such as when admitting to an illegal activity (which the respondent can indeed perceive as illegal and unjust), or when reporting minority status in a context in which violence and discrimination can be expected (Ruedin 2018a). Reporting discrimination in a Western country such as Switzerland most likely constitutes a somewhat sensitive topic but cannot be compared with admitting illegal activities. We argue that by considering key elements that drive individual perception, we also capture reporting behaviour to the same extent. In particular, we interpret the plausible differences between determinants of discrimination at work and in public situations as an indication that this assumption has merit.

In summary, we demonstrate how individual characteristics and experiences influence the perception of discrimination. Although a broad corpus of literature has shown that discrimination against immigrants is a widespread phenomenon (Zschirnt and Ruedin 2016 and Zschirnt, E. (2018, March 23). *Ethnic discrimination in the Swiss (German) labour market – first results from a correspondence test*. Presented at the NCCR Research Day #5, Neuchâtel), individual perceptions might deviate from this observation. It is therefore important to consider this additional dimension for both policymakers who seek to raise awareness of this matter and researchers who infer the occurrence of discrimination from observational data such as surveys or interviews. We conclude that contemporary individual-level measures and policy recommendations merely approximate discriminatory patterns and urge future research to consider factors that affect individual *perception* of discrimination.

#### **Appendix**


**Table 9.4** Question wording

Source: Migration-Mobility Survey 2016. Questionnaire


**Table 9.5** Additional models



Notes: \*p < 0.1; \*\*p < 0.05; \*\*\*p < 0.01; logit coefficients, standard errors in parentheses. Outcome variables: experience of discrimination, experience of discrimination at work, experience of discrimination in public; shown are the log odds

Source: Migration-Mobility Survey 2016. Weighted results

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Chapter 10 Skill Levels as a Political Resource: Political Practices of Recent Migrants in Switzerland**

**Metka Hercog**

#### **10.1 Introduction**

New, increasingly selective and temporary patterns of migration pose a challenge to social inclusion and migrant engagement. On the one hand, the temporary character of residing brings about different expectations from migrants and affects migrants' relationship with the environment they live in. It occasions questions about whether people who are often on the move develop different means of engaging and making sense of their belonging.

On the other hand, selective migration regimes in many developed countries lead to new residents being better educated than ever before (OECD/UNDESA 2013), which again has implications for migrants' ability to act. This chapter focusses on migrants' political engagement and links it with migrants' resources. After Van Hear (2014) proposed renewing attention on the part class plays in migration, we argue that resources and various sources of capital play an important role in political engagement of migrants. Because different groups control different volumes and compositions of capital and different forms of engagement require different forms of capital, we hypothesize that migrants' skill levels influence forms and aims of migrant engagement. Access to the political sphere can be gained by using skills to engage with the local environment and by using resources such as networks of organizations in which people are embedded. Although other forms of social difference, such as ethnicity, gender, generation or religion (de Rooij 2012; Giugni et al. 2014; Van Tubergen and Kalmijn 2005), have been emphasized as important factors in explaining migrants' mobilization, this chapter also calls for considering migrants'

M. Hercog (\*)

University of Basel, Basel, Switzerland

nccr – on the move, Neuchâtel, Switzerland e-mail: Metka.Hercog@unibas.ch

© The Author(s) 2019 243

I. Steiner, P. Wanner (eds.), *Migrants and Expats: The Swiss Migration and Mobility Nexus*, IMISCOE Research Series, https://doi.org/10.1007/978-3-030-05671-1\_10

resource levels in the context of migration experience another key factor. The main research question is how does migrant-specific socio-economic differentiation influence *forms of migrant engagement*?

Research on migrants' incorporation into the host society and transnational activities has mostly focussed on political integration predominantly related to electoral practices (for instance, Morawska 2009; Eggert and Giugni 2010). More-fluid patterns of mobility call attention to the need to observe residents' different approaches to engagement with the host country, which are not restricted to the practices of full citizens of the state. In addition to "conventional" political activities, such as voting in elections or involvement in political parties, Barnes and Kaase (1979) highlight the importance of more "unconventional" political activities, such as participating in petitions, demonstrations and consumer boycotts. Although these activities are largely no longer considered unconventional, they are not usually considered in research on migrants' political involvement (Però and Solomon 2010, p. 8).

This research relies on the exploration of political practices of recently arrived migrants to Switzerland. The Migration-Mobility Survey includes a module that focusses on the civic engagement and political and social participation of migrants in Switzerland and their countries of origin. Questions are specifically designed to include forms of engagement that are open to any resident of the state. We devote Sect. 10.2 of the chapter to emphasizing the need for a research approach that employs a broader understanding of engagement. This approach is specifically relevant for the study of migrant political engagement because targets and agencies of engagement relate to their political socialization and experiences of moving internationally. In addition, although migrants are not necessarily eager to gain access to the representative political sphere, they might participate in other areas of engagement (Sect. 10.3). Section 10.4 introduces the concept of "resource environment" in the context of political participation. We combine migrants' levels of individual resources with migration background and experiences of migration to explain political mobilization. We assess the effect of the various elements of the resource environment on forms of political participation by means of logistic regression (Sect. 10.6). Finally, in Sect. 10.7, we discuss our findings and identify the limitations of our approach.

#### **10.2 Understanding Immigrants and Political Participation**

Political participation is a principle of democracy. Such participation has been defined as "voluntary activities by ordinary people directed towards influencing directly or indirectly political outcomes at various levels of the political system" (Verba et al. 1995, pp. 38–39). The debates about political participation usually revolve around questions concerning who can legitimately make claims on the system. The growing presence of foreign residents makes the issue of political participation particularly complex. With political rights attached to citizenship, foreigners are largely excluded from the official participatory process, although they remain affected by the outcomes of political decisions in all areas of their daily lives. Switzerland has one of the highest proportions of non-citizen residents among nations counting more than five million inhabitants; in 2016, 24.6% of permanent residents in Switzerland did not have Swiss citizenship (Swiss Federal Statistical Office 2017). In a situation in which one-quarter of the residents are not allowed to participate in federal elections and have few other opportunities to participate politically, several questions pertaining to the quality of democratic representation come to mind: Not being able to vote, are foreign residents making claims towards the political system in other forms? Are they interested in the Swiss political system at all? These questions are relevant because society should have an interest in promoting mutual understanding and striving towards social cohesion.

In parallel with having a growing presence of residents with limited opportunities to influence the political system through conventional channels, Switzerland has, in the last decade, experienced increased anti-immigration rhetoric. In addition, although political rights of non-citizens are restricted by their official political membership status, this restriction does not prevent accusations being made against migrants that they are refusing to become involved in local societies. When civic engagement of recent migrants features in public debates, the topic is usually the alleged lack of interest from the migrants in contributing to civil society. Headlines in some popular German-speaking newspapers indicate the controversy: "How many Germans can Switzerland tolerate?"1 (Rüttimann 2007 in the *Blick*), "The elite is not integrated"2 (Schaffner 2012 in the *Tagesanzeiger*), and "A foreigner problem of a different kind"3 (Gerny 2012 in the *NZZ*). This backlash against immigrants is not reserved for low-skilled migrants who are often – also in other European countries and the United States – perceived by natives as less willing to work hard and as dependent upon scarce national resources (Larsen 2011; Senik et al. 2009). Instead of placing a burden on local resources, media and politicians alike are calling on migrants to contribute to civil society (Phillimore et al. 2018). In Switzerland, it is not uncommon to hear complaints about high-skilled migrants living in "parallel societies" or "expat bubbles" (Fournier 2012; Schneider-Sliwa 2013). In an interview from 2012 for the magazine Die Zeit, the former justice minister Simonetta Sommaruga spoke specifically of "too many high-skilled migrants who hardly speak any German. They are not integrated, and they cultivate this" (Daum and Teuwsen 2012). Responses from the migrants to persistent demands for integration vary from justifying their position with arguments of time constraints (Dacey 2012) of the short durations of their stays to noting the fallacy of these accusations (Oppliger 2013). The expansion of self-help and community groups and an increase in demand for volunteering opportunities in urban areas speak against the image of non-interested foreigners. Is it possible that the perception of migrants' political apathy is due to not focussing attention on the forms of participation in which migrants do engage?

<sup>1</sup>Translated by the author, "Wie viele Deutsche verträgt die Schweiz?" (*Blick*, 15 February 2007).

<sup>2</sup>Translated by the author, "Die Elite ist nicht integriert" (*Tagesanzeiger*, 5 May 2012).

<sup>3</sup>Translated by the author, "Ein Ausländerproblem der etwas anderen Art" (*NZZ*, 22 August 2012).

Because migrants in the prevailing discourse are considered from the perspective of nation states, either of home countries or of receiving countries, most studies observe migrants as objects of political interventions rather than consider them political agents. In addition, when migrants' activities are the subject of inquiry, the usual understanding of political engagement as being limited to electoral practices (Eggert and Giugni 2010; Morawska 2009) obscures political initiatives that occur in other public spheres. The general literature on citizens' political engagement suggests that the nature of political participation is changing in terms of channels, topics and targets (Giddens 2002; Norris 2002; O'Toole and Gale 2013). Social transformations at the global level have led to "new grammars of actions" (McDonald 2006) that are closer to everyday and lifestyle forms of activism. Some of the major driving forces behind "detraditionalization" (Giddens 2002) and the general citizen disengagement from conventional politics surrounding political parties have been located in social events such as the end of the Cold War and the development of new technologies. The Cold War had a destructive effect on the image of political authorities and on the mobilizing potential of political parties (Beck 1997). Contemporary forms of action that are more individualized and self-reflexive draw upon new technologies that allow for a global outlook and for a different perception of what matters in the world and at the same time enable a more creative repertoire of action (O'Toole and Gale 2013). Forms of action that are based on the "life politics" (Giddens 2002) of personal and self-reflexive engagement with the world are manifested in actions of political shopping, consumer boycotts, e-activism and veganism, to name a few. In addition to the conventional and unconventional forms of political participation, researchers characterize such forms as "post-conventional" (Stolle and Hooghe 2005). Bang (2004) sees participants engaging in such forms of participation as preferring to engage in concrete, local, self-actualizing political projects rather than participating in formally organized programmes with political ideologies. Political consumerism, for example, is supposed to appeal to people who view themselves as alienated from formal political settings (Micheletti 2004). Considering that the literature on citizens' political participation acknowledges a turn and diversification of political action, it is surprising that "unconventional" or "post-conventional" forms of participation are rarely discussed in the research on migrants' involvement. With an exploration of "truly" transnational political practices of migrants in the Netherlands, van Bochove (2012) provides a rare example of recognizing migrants as political agents beyond the conventional forms of engagement and beyond protests or illegal acts of resistance.

The dominant literature on migrant political engagement uses the approach of political opportunity structure (POS), which stays close to its original objective of explaining migrants' collective actions in either ethnic or cross-ethnic organizations (Koopmans et al. 2005; Koopmans and Statham 1999). This approach hardly considers the above-mentioned new patterns of political action, which are found to be more individualized, direct and expressive (O'Toole and Gale 2013). In Sect. 10.3, we explain how our empirical investigation tries to address the mentioned limitations.

#### **10.3 Research Approach for Broader Understanding of Engagement**

The data used for this study derive from the newly conducted Migration-Mobility Survey, which focusses on several aspects of migrants' living conditions in Switzerland (see Chap. 2). The survey includes a module on civic engagement and political and social participation of migrants in Switzerland and in their countries of origin. Questions were specifically designed to include forms of engagement that are open to any resident of the state. To be included in the survey, one had to have been born outside of Switzerland, to not hold a Swiss nationality and to have immigrated to Switzerland within the past 10 years. Based on these requirements, the respondents of the survey stay in Switzerland based on holding one of the following permits: short-term (up to 1 year), resident (1 or 5 years), settlement or diplomat status. Thus, our target population is not entitled to vote in general elections in Switzerland. Except for voting at the cantonal level in two cantons (Jura and Neuchatel) and at the communal level in communes in five cantons (Jura, Neuchatel, Vaud, Freiburg and Geneva) (EKM 2018), non-citizens are largely also excluded from voting at the local level. This situation makes it particularly compelling to accept a broad definition of political participation in an attempt to cover the previously mentioned description of voluntary activities directed towards influencing political outcomes by Verba et al. (1995).

Survey respondents were asked to identify from among the mentioned activities (see Table 10.1) those they did in the last 12 months with the objective of improving things or of helping to prevent things from going wrong.


There are different approaches to trying to improve things or help prevent things from going wrong. During the last 12 months, have you done any of the following…


Groups of activities based on the above response categories considered in this chapter:


Source: Migration-Mobility Survey 2016. Questionnaire

The responses include activities surrounding the established political entities that are open to non-citizens. They can make claims towards formal representative politics, try to influence political terrain by supporting certain political ideologies or even by joining a political party. Responding to accounts of general disengagement from party politics since the 1990s and given that party politics is likely to be even more inaccessible to migrants, the list includes other activities that might be targeting a wider range of institutions beyond the government and that also do not require delegates and representatives to act in the name of others. McDonald (2006) writes about global social movements in which activists are reluctant to submerge their identity into any organization or movement and rather engage in actions that are more expressive, hands-on and aim for direct effects. People can choose to express their alignments by joining a single cause, with ad hoc involvement relating to a specific issue or goal. They might join a demonstration, show alliance with a specific cause by displaying visual statements on their clothing or only add a campaign badge to a social media picture. These forms show political actions as closer to everyday and lifestyle decisions. Our approach, however, does not allow us to observe the motivations behind actions and stays within the investigation of forms of engagement.

In addition to exploring whether migrants do engage in these new patterns of political action, we are further interested in the attributes that distinguish the politically active in different forms of political engagement. For that purpose, we have created three major groups of activities based on the list of nine possible answers to the abovementioned survey question (see Table 10.1), namely (a) contacting, (b) expressive and (c) quiet forms of activity. The rationale for dividing activities in these three groups is based on the typology developed by Schulz and Bailer (2012). Contacting activities include contacting a politician, government or local government official; making a donation to a political campaign or to a political party; and working in a political party or in another organization or association. We consider the following activities expressive forms: wearing a campaign badge or sticker, participating in a lawful public demonstration or using the internet to express opinions. Signing a petition or boycotting certain products is grouped together under quiet forms of activities. We acknowledge that most of these activities can be performed with various intensities and time effort. However, some forms of activities are found to be less demanding that the others. De Rooij (2012) considers signing a petition, wearing a political badge or boycotting products low-cost acts, whereas contacting activities and demonstrations fall within high-cost acts. However, our approach does not allow us to make such clear-cut distinctions because we do not have information on how often people engage in any of the activities or on how intensely they prepared for any of the acts.

#### **10.4 Political Resource Environment for Migrants**

Along with expanding our understanding of political engagement by examining a wider range of repertoires, this chapter also aims to contribute to the question of migrants' resource environment. We try to identify resources that are conducive to activities of migrants. Although we align with arguments of shifting patterns of political behaviour, we accept that it is important to consider the political and social context within which this behaviour is or is not happening. The POS approach, which predominates in studying different aspect of migrants' integration, postulates that national and local institutional frameworks with their policy setups shape migrants' behaviour by either stimulating or inhibiting their mobilization (Fennema and Tillie 2004; Ireland 1994; Koopmans et al. 2005). Ireland (1994), as the first to apply this approach in the field of migration studies, actually compared France to Switzerland and established Swiss closed political opportunity structures as the main reason for the low level of political mobilization among migrants. Many studies, globally and in Switzerland, have followed the trend of investigating institutional environments and their role in migrants' actions (Eggert and Giugni 2010; Giugni et al. 2014; Giugni and Passy 2004; Manatschal and Stadelmann-Steffen 2014).

Nevertheless, Però (2008) considers a focus on political opportunity structures insufficient to explain migrants' political behaviour. In his account of Latin American mobilization in London, Però gives credit to the approach and confirms the importance of opportunity structures of the receiving country for channelling the Latino collective action. The limitations of the approach are found largely in terms of its narrow application of the institutional framework, failure to include a transnational dimension of opportunity structures and migrants' earlier political socialization. Recent studies in the case of Switzerland have improved on the first aspect. Giugni and Passy (2004) and Manatschal and Stadelmann-Steffen (2014) assume integration policy regimes to be more than only legal regulations. Manatschal (2012) has previously shown that cantonal integration policies correspond to local cultural notions of belonging. However, in her work, the notions of who belongs to specific communities and cultural understandings of migrants' entitlements are conceived as being expressed through integration policies. Wider cultural attitudes, such as "multiculturalism" in the United Kingdom (Però 2008) or a more migrantwelcoming culture in some Swiss cantons (Manatschal 2012), can also work towards encouraging mobilization through constituting incentives in different areas of daily life. The second main limitation of the POS approach, as viewed by Però, is that it adopts the perspective of the nation states as key referents and thus does not explain diverse and changing mobilizations within a stable institutional environment of a single country. Recent works on the interaction between migrant integration and transnationalism (Erdal 2013; Erdal and Oeppen 2013; Van Meeteren 2012) have at least partly addressed the concerns about the POS approach not being adapted to the transnational turn in migration studies. Most studies, however, stay with studying bi-national activities that bind migrants across their countries of settlement and countries of origin and do not allow for considering the rise of transnational and global political structures that are increasingly the targets of political actions (O'Toole and Gale 2013).

Considering these limitations of the predominant approach to the study of migrants' political mobilization, we are interested in adding to the discussions by expanding the concept of a "resource environment" (Levitt et al. 2017) and adapting it to the context of political participation. We initially want to observe how the new migrants mobilize and how the "resource environment" in which individuals are embedded influences their repertoires of actions. Similar to migration itself, migrant engagement depends upon the resources that individuals or groups of people are able to mobilize. We suggest that the individual's resource environment comprises a combination of the individual's social and human capital, migrants' background in terms of political socialization, their experiences in the receiving context and of cultural notions of belonging that are prevalent in resident societies. By combining all of these factors as the resource environment, we are able to analyse not only the role of the environment in structuring opportunities and obstacles for migrants but also the role of migrants' agency. Different individuals control different compositions of capital, which influences their social position in the receiving society and their abilities as actors. The literature on political participation in general and the literature focussing on migrant political participation in particular corroborate the importance of education as a critical resource for political participation (Eggert and Giugni 2010; Koopmans et al. 2005; Morales and Giugni 2011). Migrants' social esteem needed to engage in political activities can also come from their political biographies and from being socialized in political cultures that praise taking action. People with earlier experiences of engaging are expected to be more inclined towards recurrent mobilization in the receiving country. Political socialization starts at a young age, when people develop their political values and beliefs in interactions with the family, school, media and other agents of socialization (de Rooij 2012). We are not saying that people who were not active before migration would remain politically inactive in a new setting; rather, we postulate that experiences of migration could motivate some people to adopt new roles. The cause could be unfair working conditions, as in the case of many Latino migrants in London (Però 2008), which pushed them to organize, or assuming a new role as an accompanying spouse with more time for public involvement (Fechter 2016; Isaakyan and Triandafyllidou 2014; Trundle 2014). Finally, the possibilities to use individual resources nonetheless depend upon differential opportunities that are linked to cultural notions of belonging and to the attached predispositions for considering migrants' rights and obligations.

#### **10.5 Research Design and Methodology**

In Sect. 10.3, we have already described the dependent variable (overall political participation, direct contact activities, expressive and quiet forms of activities). Because the outcome variables are binary, the models used for analysis are binary logistic regression models calculating the probability of the outcome. All of the results presented in this paper are based on data weighted with normalized weights. In this section, we describe our main independent variables, which respond to the attempt to expand the concept of "resource environment" as described above. Independent variables can be divided into three groups.

The aspect of including a transnational lens and migrants' previous political experiences is addressed by the variables related to migration background. We introduced a new variable for democracy level in the country of birth, created based on the dataset in the Polity IV Project, Political Regime Characteristics and Transitions, 1800–2016. We used the variable that refers to the competitiveness of participation (PARCOMP) and captures "the extent to which alternative preferences for policy and leadership can be pursued in the political arena" (Marshall et al. 2017, p. 26). Countries are assessed each year on a scale of repressed, suppressed, factional, transitional and competitive. Because of yearly variations, we examined the average score of migrants' countries of birth for the last 5 years, that is, from 2011 to 2016. The variable of birth country is therefore not used to examine the relevance of ethnicity as a determining factor of migrants' engagement but rather as a proxy for migrants' political socialization in systems with varying degrees of participation competitiveness. Countries of birth are grouped into five categories of a scale from repressed to competitive, following the classification of the Polity IV Project. Although we agree that ethnicity itself is not a mobilizing factor as such, being politically educated and socialized within a more or less politically encouraging system can play a role in migrants' social esteem needed for participation. Given that the Swiss political system is ranked as competitive, we could also expect that immigrants from other countries with a competitive system will have higher rates of participation because of the similarities between political systems, as has been suggested by Lien (2004) and Bilodeau (2008). We add the number of times people have moved internationally and their duration of stay in Switzerland as two other variables covering migration background. We find it interesting to explore whether people who are often on the move develop different approaches to engaging and making sense of their belonging. Time since moving to Switzerland is expected to be related to political engagement because people had the possibility of expanding their networks and learning about particularistic political interests in their surroundings. Other studies have found a link between rootedness and immigrant volunteering (Manatschal and Stadelmann-Steffen 2014; Stadelmann-Steffen and Freitag 2011) and a strong link between the time spent in a new country of residence and patterns of immigrant political engagement (de Rooij 2012).

The second set of variables relates to specific migration experiences in the receiving context. Individual experiences of migration are wide-ranging, and although some migrants suffer significantly from loss of resources, others benefit more from the move. We included variables that examine the general satisfaction with moving to Switzerland, the feeling of discrimination and the level of interest in current events in Switzerland among respondents. The variable on discrimination indicates whether the respondent has experienced a situation in which she felt treated less favourably than other people were. Although the dataset includes information concerning on what basis the discrimination occurred (race, religion, immigrant background, gender, disability, age or sexual orientation), we prefer to use the overall feeling of discrimination. We believe that the experience of marginalization can work as a motivating factor to change the situation regardless of the assigned characteristics based on which people felt discriminated against. Latin migrants in London spoke of exploitation in the workplace as a push for their mobilization (Però 2008). In that sense, structural constraints rather than opportunity structures mobilized the migrants. However, if the feeling of marginalization were not accompanied

by other elements of the resource environment, it might also have a discouraging effect. We can easily think of examples of discrimination leading to social exclusion of groups of migrants (Negi 2013; Wray-Lake et al. 2008). Latino labourers in the informal market in the United States face discrimination with social isolation (Negi 2013), and Arab-American immigrants (Wray-Lake et al. 2008) are shown as doubting their role in society due to experiencing prejudice based on their identity. Riaño (2003) showed how skilled Latin American women face de-qualification and deemancipation after they have moved to Switzerland. Experiences of discrimination can therefore equally inhibit or foster civic engagement. We also account for political interest in current events in the country of residence and predict that it will positively influence political participation in general as was found in Nguyen Long (2016) and Giugni et al. (2014).

At the contextual level, we add the variable on the language region. It has been previously proven that dividing respondents according to the language spoken in the place of residence can help explain volunteering in Switzerland (Stadelmann-Steffen and Freitag 2011). The rationale for including a variable on language region in the analysis is based on different regional notions of belonging that work their approaches to inclusion and exclusion through cantonal integration policies and reveal themselves in other aspects of daily life (Manatschal and Stadelmann-Steffen 2014). The division of the variable into German-, French- and Italian-speaking regions is based on the postal code of the respondent's residence.

Demographic and social-economic variables are included as control variables. Age, gender, level of education, level of employment, language proficiency and region of origin have been viewed in previous research as important predictors of immigrants' voluntary engagement and political participation (Aleksynska 2008; Eggert and Giugni 2010; Giugni et al. 2014; Koopmans et al. 2005; Morales and Giugni 2011; Verba et al. 1995). The literature on political participation in general and the literature focussing on migrant political participation in particular corroborate the importance of education as a critical resource for political participation. Research specifically addressing participation of migrants in Switzerland (Giugni et al. 2014; Manatschal and Stadelmann-Steffen 2014) aligns with these results. Because the level of education is expected to positively influence mobilization on political issues, it is therefore more interesting to observe whether skill levels also affect forms of engagement. We mark people with only compulsory education as low-educated, those with higher secondary education and up to advanced technical or professional training as mid-educated and those with at least a bachelor's degree as high-educated.

We include level of employment in the analysis with the expectation that those who are employed full-time will tend to engage less often than will those engaged part-time or who are not employed. According to Verba et al. (1995), time is one of the most important resources for participation. Respondents in our survey are considered not employed if they are seeking a job, undergoing training, looking after their home or in any other non-employed situation.

Having a good command of the language of the place of residence, measured through speaking skills, is considered an important resource for engaging with the local environment. Lack of local language skills and the perceived lack of effort to learn local languages has been at the core of accusations towards new migrants in Switzerland. Lack of communication abilities can be considered a reason for social exclusion (Ersanilli and Koopmans 2010) but, obviously, also a threat to national cohesion as exemplified by the remarks of politicians in Switzerland.

We also control for the respondents' regions of birth and divide them into five regional groups according to their country of birth: Europe, Africa, South and Central America, Anglo-Saxon countries (USA, Canada, Australia, New Zealand) and Asia.

#### **10.6 Results**

#### *10.6.1 Descriptive Analysis*

Before delving into the different forms of participation, we initially wanted to determine whether recent migrants, who are at the centre of our attention for this study, differ from the general Swiss population in terms of political interest. We name all respondents of the survey recent migrants because arrival in Switzerland within the past 10 years was one of the conditions to be included in the survey. The information for the general Swiss population was obtained from the European Social Survey (ESS), which worked as a model for several questions included in the Migration-Mobility Survey. The question on how interested the respondents are in politics in Switzerland is identical in both surveys and actually shows more people who say they are either very or quite interested in politics in Switzerland among recent migrants in the Migration-Mobility Survey than among the Swiss citizens in the European Social Survey (66.7% vs. 62.8%). The high level of political interest among migrants arriving within the past 10 years would not be anticipated if we believed the perceptions of migrants in the public debate.

Turning our attention to political behaviour, the picture is quite different. Of the respondents, 38.6% have done at least one of the mentioned activities within the past 12 months. Table 10.2 shows that except for boycotting products and expressing political opinions and communicating about activities on the internet, recent migrants are much more absent from political participation compared with Swiss citizens. The fact that behaviour on the internet and consumption show a different tendency could possibly be explained by structural and institutional barriers prevalent in other spheres that discourage migrant political participation. However, research that explored differences in the pattern of political participation between immigrants and the majority population found that explanatory mechanisms for participation work differently for immigrants than for the majority (de Rooij 2012). We suggest that levels of resources in terms of education and professional status must be combined with individuals' migration background and experiences of migration to explain political mobilization.


**Table 10.2** Political activities of recent migrants and overall Swiss residents (percentage responding "yes")

Note: We do not report the responses on donations to political campaigns and political parties because they were not available for the ESS for this round

Sources: Migration-Mobility Survey 2016, weighted results, and European Social Survey, round 2016


**Table 10.3** Political activity by education (percentage responding "yes")

Source: Migration-Mobility Survey 2016. Weighted results

Based on the list of nine possible forms of engagement in the survey, we create a dummy variable measuring the overall political participation. A given respondent might have picked only one of the activities or a number of them because multiple responses were possible. To identify the attributes that distinguish those politically active in different forms of political engagement, we divide activities into three major groups: contacting, expressive and quiet forms of activity. Given the exploratory nature of this chapter, we initially study the prevalence of different types of political involvement among respondents and break it down by their educational level (Table 10.3). Other studies have established that being more educated provides crucial resources for political participation. Therefore, it is not surprising that our data also show that individuals with high education are more likely to become involved in all types of activities. Almost 45% of highly educated respondents have participated in at least one form of activity. Among the low-educated, only 19% were engaged in any sort of political activity. The difference between high-educated and the rest is particularly large within the quiet forms of political engagement (boycotting products and signing petitions). Quiet forms of engagement are the most common among mid- and high-educated migrants, whereas low-educated migrants more often choose expressive forms of engagement. Examining more closely the types of activities that are clustered under the expressive forms, we ascertain that more-common engagement in this form is largely due to internet communication about political events.

In the explanatory analysis, we aim to observe how far political mobilization can be explained by factors other than the level of education. We are interested in not only whether recent migrants participate but also *how* their resource environment, which considers the specific migration experience, influences patterns of participation.

#### *10.6.2 Explanatory Analysis*

Table 10.4 shows the results of the binary logistic regression models for the three types of activities and for the overall political participation. Considering the results for overall participation, the most interesting finding is that almost all variables related to migration background and migration experiences have a significant effect on involvement. Only the feeling of general satisfaction in Switzerland and residence in different language regions of Switzerland lacked a significant effect on overall participation.

We observe that the resources environment works unevenly for different groups of activities. Level of democracy in respondent's birth countries works as a determinant only for quiet activities. Coming from a place with more competitive possibilities of participation positively affects the use of quiet forms of engagement among recent migrants in Switzerland. Theories of political socialization stress the importance of early age for determining adult political behaviour (Verba et al. 1995). The use of petitions and boycotts might be more common in countries that are ranked as competitive in the Polity IV index, which would predispose migrants to engage through these forms.

The number of times people moved internationally is another variable displaying a strong effect. We find that people with rich international experiences, who had moved from country to country before migrating to Switzerland, can use their international experiences as a resource for political mobilization. We assumed that there would be a difference between those people who have moved a couple of times as opposed to those who have moved 10 or 20 times in their lives. Our results appear to support this assumption by showing that moving up to five times works (compared with only moving once to Switzerland) as a predictor for any type of activities and specifically for the quiet forms. In any case, this finding goes against the literature on conventional political participation, which considers international mobility a reason for low participation. Migrants are supposed to be less likely to be recruited by the mainstream social institutions and political organizations because they are difficult to reach (Ramakrishnan and Espenshade 2001). Our approach, which


**Table 10.4** Logistic regression for different types of activities, average marginal effects



Source: Migration-Mobility Survey 2016. Weighted results (normalized weights) \* p <0.1, \*\* p<0.05, \*\*\* p < 0.01

observes engagement at an individual level, allows us to go beyond restrictions imposed by mainstream institutions and manages to capture migrants' activities when they happen outside the arena prescribed for politics. Thus, international mobility experiences actually are a resource. However, the longer people stay in the country, the more likely they are to engage politically, which aligns with the theory on the role of residential stability and social connectedness. They become easier to reach, possibly create additional social networks and might become more interested in the local place through their rootedness.

The feeling of discrimination is a strong motivator for engagement. Immigration is a transformational experience, and going through it might activate a new political identity (Nguyen Long 2016). A perception of exclusion could motivate people to seek other people with shared experiences and act as a measure of group consciousness. Although discrimination could be expected to work in two opposite directions, possibly leading to disillusionment with their new society, the effect for recent migrants in Switzerland works in favour of mobilizing. If an average person feels discriminated against, the probability of engaging increases by almost 16%.

Display of interest in current events in Switzerland also has an effect on all forms of participation. Political interest is a predisposition for participation found in studies concerning political participation in general and migrant political participation in particular (Giugni et al. 2014; Verba et al. 1995). It is a relevant predictor for all types of acts.

To observe the importance of context, our models include a language region. Living in a French-speaking canton increases the probability of migrants to be engaged in expressive forms, whereas living in the Italian-speaking canton decreases the probability of quiet forms of engagement. Higher engagement in the Frenchspeaking cantons could be a response to less-restrictive integration policies and to a local population that is in general less sceptical towards migrants compared with the German-speaking cantons (Danaci 2009; Manatschal 2012).

We observe that respondents' demographic background works unevenly for different groups of activities. Age, for example, has a significant effect concerning respondents' probability of engaging in direct and expressive forms. Age does not

appear to play a role when people sign a petition or boycott products. However, the possibility of being active in any of the expressive forms decreases for older migrants. Women traditionally participate less than men do (Giugni et al. 2014; Klandermans et al. 2008; Nguyen Long 2016). It is therefore not surprising that we come to the same finding based on our data. Being male is a rather strong predictor for expressive forms of engagement. However, gender does not appear to matter for quiet activities; men and women are equally likely to engage in this form.

Level of education is found to have the strongest effect on political engagement. Having a university education increases the probability of any type of engagement by 26.4% compared with the low-educated migrants. The strength of association between education and engagement is markedly more pronounced for unconventional forms of engagement, both the expressive and quiet ones. This strength might be due to the social esteem needed to engage in non-regulated political activities.

The variable on employment distinguishes those with full-time employment from those working part-time or who are not employed. Full-time employed tend to engage less in conventional acts of directly engaging with politicians. This behaviour could be explained by time availability; all of the activities grouped under "contacting activities" are listed as high-cost acts in the typology by de Rooij (2012). Notemployed are more likely to be engaged in expressive forms of activities, but overall level of employment does not affect different types of activities, which, according to the rationale by de Rooij, could mean that they are less time-demanding.

Political engagement grows with improvement of language skills. Host country language proficiency provides communication abilities and enables access to information. Language skills, however, do not predict that someone will engage through expressive patterns. Wearing a badge, joining a protest or engaging politically on the internet appear to be unaffected by the command of local languages, which suggests that not speaking Swiss German, French or Italian does not prevent people from mobilizing in these forms. Most likely, they will in this case target institutions other than the local government.

People born in regions other than Europe are less likely to engage in contacting activities. Specifically, respondents born in any of the African countries are significantly less likely to engage in these forms compared with those born in European countries. In contrast, respondents born in North America or Australia are more likely to be engaged in expressive activities.

#### **10.7 Discussion and Conclusion**

This chapter explores the engagement practices of recent migrants and observes how the "resource environment" specific to migrants influences their practices. We observe migrants' political participation by drawing on the concept of "resource environment" specific to migrants. Responding to the observed inability of the prevailing approach of studying migrant political participation to explain diverse mobilizations within a stable institutional environment of a single country (Però 2008), we include the influence of varied migration backgrounds and experiences to add to the debate on what counts as a political resource.

Overall, our findings suggest that migration background and migration experiences have a significant effect on involvement. Furthermore, we observe that the resource environment works unevenly for different groups of activities. Activities that require delegates and representatives to act in the name of others do appear to be high-cost acts (de Rooij 2012). Our results show that one is more likely to be engaged in contacting activities if one has a good command of the local language, has stayed in Switzerland for a longer time and expresses a high level of political interest. Interestingly, the level of education does not increase the probability of acting in the representative political sphere by contacting, donating to or joining a political party or organization. Time spent in Switzerland and local language skills – likely related to the time spent in Switzerland – matter most for these types of activity. Expressive forms of engagement are more likely for younger men with university education. The variables on feeling discriminated and being politically interested are significant for all forms of activities. Language proficiency, however, does not affect these forms. People are not discouraged from joining demonstrations, expressing political matters on the internet or wearing a campaign badge. Finally, acting with the quiet forms increases with higher education, language proficiency, the number of times people have moved internationally, duration of stay and by a higher democracy level in the migrant's birth country. It is interesting that the last variable is significant only for this form of activity. Political socialization in a more competitive system in a birth country does not translate to contacting politicians in a country of residence; nor does it affect expressive forms. However, for the quiet activities, which could be more self-actualizing than demanding for the others to change, the socialized norms of engaging carry over to adult-years in a different country. Moreover, level of employment does not appear to matter, giving an indication that one can do any of these activities regardless of working full-time or of related time constraints.

Our results show that migrants are not necessarily eager to gain access to the representative political sphere but might participate in other areas of engagement. The level of education is a strong predictor for unconventional forms of engagement, expressive and quiet ones. Agencies of migrant engagement relate to their political socialization and experiences of moving internationally, in which the feeling of discrimination acts prominently as a mobilizing force. The social esteem required to act against discrimination through non-regulated political activities appears to be coming from a high level of education and international experiences. When one feels discriminated against, the probability of becoming engaged through quiet or expressive forms is greater than of engaging with politicians.

Our approach considers that political patterns are becoming more individualized, direct and expressive (O'Toole and Gale 2013). Finding a motivation to join organizations or associations might be particularly challenging for migrants who have changed locations not that long ago and live in increasingly diverse societies. Because of these limitations in the literature in relation to new migrants, we argue that before proclaiming them apathetic, there is a need to expand our understanding of political engagement.

However, our approach does not allow us to observe the motivations behind actions and stays within the investigation of forms of engagement. Another limitation with our approach is that we cannot predict what explains more or less participation. A given respondent might have picked only one of the activities or a number of them because multiple responses were possible in the question used for our outcome variable. We do not have information on how often people engage in any of the activities or how intensely they prepared for any of the acts. Moreover, although this question was specifically designed to address the possible options of engagement for migrants, we nonetheless cannot claim to cover the full repertoire of political patterns and should therefore acknowledge the possibility of underestimating migrants' mobilization.

To conclude, we wish to respond to some of the questions raised in the beginning of this chapter. Foreign residents are interested in the Swiss political system. They are making use of varied forms of engagement and are leaning towards forms that are not constrained by the representative political sphere within the borders of the nation-state. Moreover, it is indeed possible that the image of uninterested migrants is misplaced because attention is focussed more on the expectations of the receiving society than on what migrants actually do. Therefore, the notion of the Migration-Mobility nexus as observed through practices of political participation enables us to go beyond the traditional migration approach that limits migrants' actions to the perspective of the nation states. Our findings show that new, selective and temporary patterns of mobility lead to broadening of political action repertoires. People on the move choose different types of activities, but their experiences of international mobility function as a resource for engagement.

#### **References**


zech&utm\_medium=fix&utm\_campaign=wall\_abo&utm\_source=zeitde\_zonpme\_int&utm\_ content=premium\_packshot\_cover\_zech. Accessed 8 Feb 2018.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Part V Transnational Life and Future Migratory Intentions**

## **Chapter 11 How Transnational are Migrants in Switzerland? An Analysis of the Migration-Mobility-Transnationality Nexus**

**Eric Crettaz and Janine Dahinden**

#### **11.1 Introduction**

Since the early 1990s, a new debate has unfolded within the field of migration studies pointing to a new phenomenon – "transnationalism" – which, as scholars argued, had been neglected by migration studies. In particular, Glick Schiller and her colleagues argued that contemporary migrants should not be characterized as being "uprooted", as was generally assumed in the classical assimilation theory. Rather, they are *transmigrants*, becoming firmly rooted in their new country but maintaining simultaneously multiple linkages to their homeland. They bring together their societies of origin and settlement (Glick Schiller et al. 1992; Basch et al. 1994). In due course, transnational studies have proliferated, and the transnational perspective has become one of the fundamental means of understanding the contemporary practices of migrants occurring across national borders (Faist et al. 2013; Vertovec 2009); however, their modest theoretical essence and often descriptive contextdependent analyses have been criticized (Portes et al. 1999; Waldinger 2015). For this chapter, we will address three shortcomings of this field of research.

E. Crettaz (\*)

nccr-on the move, Neuchâtel, Switzerland e-mail: Eric.Crettaz@hesge.ch

J. Dahinden Laboratory for the Study of Social Processes, University of Neuchâtel, Neuchâtel, Switzerland

nccr-on the move, Neuchâtel, Switzerland e-mail: Janine.Dahinden@unine.ch

© The Author(s) 2019 267 I. Steiner, P. Wanner (eds.), *Migrants and Expats: The Swiss Migration and Mobility Nexus*, IMISCOE Research Series, https://doi.org/10.1007/978-3-030-05671-1\_11

School of Social Work, University of Applied Sciences and Arts of Western Switzerland, Geneva, Switzerland

First, most of the literature relies on qualitative case studies, whereas quantitative studies remain rare. The latter have shown that transnationality1 is far from being a "lifestyle" of migrants as it appeared in the 1990s. Not every migrant is "transnational", neither in terms of networks nor in terms of activities, practices or feelings of belonging (Guarnizo et al. 2003; Portes et al. 2002; Dahinden 2005, 2009). In line with this research strand, we ask: How transnational are the migrants living in Switzerland? How can we characterize those who are most transnational and those who are the least, and how can we explain these different degrees and forms of transnationality?

Second, most studies in this field focus on one (or a few) nationally defined group(s) and then elaborate in a descriptive way on the transnationality of this/these group(s). This approach is problematic under two aspects: First, it prevents the creation of an overall theoretical model. Second, it is problematic because of the national- or ethno-centred epistemology (Wimmer and Glick Schiller 2002; Dahinden 2016). Therefore, we propose here an analysis of migrant transnationality among a broad range of migrants living in Switzerland, thanks to the Mobility-Migration Survey, performed within the framework of the nccr – on the move, to elaborate explanatory models for transnationality going beyond nationality.

Finally, highlighting different patterns of transnationality requires a theoretical framework that captures not only migration, but also mobility more generally: For this reason, we apply a "mobility lens" (Urry 2007). The traditional migration framework often ignores that migrants are also mobile, in terms of pre-migration but also post-migration mobilities (Dahinden 2010b; Hui 2016; Moret 2018; Schapendonk 2015). In other words, our analysis contributes to this book by investigating the Migration-Mobility-Nexus with respect to transnationality, going beyond normative ideas by integrating them analytically into one model.

Section 11.2 introduces the assumptions and conceptual reflections we find in the literature concerning transnationality. In Sect. 11.3 we present the dataset and the indicators used, and descriptive statistics in Sect. 11.4. Because the main factors contributing to transnationality are correlated and because we aim at going beyond mere descriptions, we performed regression models in Sect. 11.5 and were able to identify the most and the least transnational groups among the studied migration population. Finally, we conducted a multiple correspondence analysis (see Sect. 11.6), allowing us to understand in detail how these various factors interact, leading to the identification of five ideal types of "transnational migrants". In a nutshell, our analysis confirms that transnationality can be linear, resource-dependent or reactive. In other words, transnationality can simultaneously be a sign not only of possessing high resources – in terms of legal capital, education and economic resources – but also of discrimination. Migration regimes and educational skills are of utmost importance concerning high or low degrees of transnationality – however, occasionally in unexpected combinations.

<sup>1</sup>Although the first vague references in the literature talked of "transnationalism", we use here – in line with other scholars (among others Faist et al. 2013; Glick Schiller 2004) – the term "transnationality". The latter term points to the dynamic, historical variable and constructed nature of the phenomenon, whereas terms with the ending –ism are often subjected to ideological connotations rather than analytical perspectives (such as capitalism, socialism, and multiculturalism).

#### **11.2 Some Assumptions and Hypotheses About Migrant Transnationality**

Pries (2008) called for a clear operationalization of the concept of "transnationality". For this reason, we define transnationality as being composed of three dimensions: *transnational mobilities* in transnational spaces, *transnational social relationships*, and *feelings of attachment towards the country of origin.* In other words, being transnational involves (a) a mode of *being mobile*, (b) *ways of being* in terms of acting and performing (i.e., building up and maintaining transnational social relationships) and (c) *ways of belonging* (see also Levitt and Glick Schiller 2004). Considering all three aspects of transnationalism allows us to identify a diverse range of transnational formations and to detect their underlying mechanisms.

*First Dimension: Transnational Mobilities* Migration research has been highly influenced by a sedentarist bias due to methodological nationalism and the reproduction of the nation-state logic (Dahinden 2016; Hui 2016; Chavel 2014; Wimmer and Glick Schiller 2002). From a traditional migration perspective, the absence of mobility appears to be the norm, if not the ideal, assuming that a period of displacement is automatically preceded by a phase of sedentarity and followed by a process of settlement and mobility is reduced to the migration journey (Dahinden 2010a, 2012). In a similar reading, mobility is occasionally reserved to Europeans or highly skilled persons, whereas migration is associated with unqualified people, reproducing the normative political categories of desired and non-desired migrants (critically see Faist 2013). More recently, scholars have pointed to the continuous mobility processes of individual migrants and to crucial pre- and post-migration mobilities (Schapendonk 2015; Moret 2015). From the perspective of the "mobility studies" (Urry 2007; Cresswell 2010), migration is only one form of mobility among others. We argue that to understand transnationality – and to investigate how mobile the migrants living in Switzerland have been and remain, and who are the most and who the least mobile – we must include pre- and post-migration mobilities.

*Second Dimension: Transnational Social Relationships* The idea of transnationality emerged from the realization that migration does not lead to cutting social relationships; in contrast, they are maintained, reinforced or reinvented across borders, before and after migration (Vertovec 2009). Hence, we propose to examine transnational social networks at two points in time: *before* and *after* arriving in Switzerland. These two perspectives relate to two different fields of research within the transnational migration literature. First, since the 1960s, researchers have emphasized the fact that personal networks can be understood as conduits of information and social and financial assistance, thereby shaping, sustaining and channelling migration (Boyd 1989; Massey et al. 1993). One crucial dimension is so-called "chain migration" (Massey et al. 1993); social networks are channelling migrants to places in

which they know persons who can support them when they arrive, concerning housing, jobs and other fields. Hence, we analyse transnational relationships that have channelled individuals to Switzerland. Second, we investigate the transnational relationships currently maintained by migrants, hence *after* immigration. We are interested not only in the extent of network transnationality but also in the question of who are the most transnational and who are the least transnational concerning networks. Here, we primarily focus on "core transnational relationships" (Levitt and Jaworsky 2007), meaning on the most durable social relationships, to the detriment of social relationships that are mobilized only occasionally.

*Third Dimension: Feelings of Attachment Towards the Country of Origin2* Although we know from the literature that durable and stable transnational relationships often decline with the duration of stay and between generations (Kivisto 2001), transnational forms of belonging are more durable over time (Gowricharn 2009; Wessendorf 2010). Consequently, we investigate how attached the population under study is to their country of origin and who are the most and who are the least attached.

*Linear, Resource-Dependent or Reactive Transnationality?* We find in the literature different hypotheses concerning the mechanisms underlying different degrees of transnationality. The first hypothesis maintains that transnationality is a logical and inevitable consequence of migration and hence that it is "linear": migration does not mean a rupture with the country of origin. Rather, immigrants maintain automatically and simultaneously ties with their countries of origin (or a third country) and develop multiple forms of belonging. Likewise, it is argued that the duration of stay affects this linear transnationality; the longer the migrants stay in the receiving country, the more transnationality fades away (Kivisto 2001).

A second set of hypotheses is based on the finding that not all migrants develop and maintain transnationality to the same extent because transnationality depends upon the economic, cultural and social resources of migrants (Al-Ali et al. 2001; Dahinden 2009), on degrees and forms of discrimination and marginalization (Itzigsohn and Gioguli Saucedo 2005; Dahinden 2005) and of nation-state migration regimes and border politics (Waldinger and Fitzgerald 2004). Ideas of solidarity, reciprocity and belonging are also crucial for the establishment and maintenance of transnationality (Dahinden 2010b; Faist 2000). In other words, transnationality might be "linear", but it might also be "resource-dependent" (Itzigsohn and Gioguli Saucedo 2005). Studies showed that it is often the best integrated into the labour market, the naturalized, and the best educated, who are the most transnational. Conversely, transnationality might also be the result of discrimination and marginalization and therefore "reactive" (Snel et al. 2016).

<sup>2</sup>See sect. 11.3 on methodology for data limitations concerning transnationality beyond countries of origin.

We will follow these ideas and assess, for all three dimensions of transnationality, the hypotheses of linear/source-dependent/reactive transnationality among the studied migrant population.

#### **11.3 Data and Methods**

The following analyses are based on the Migration-Mobility Survey, performed within the framework of the National Centre of Competence for Research nccr – on the move. Data collection was carried out in the fall of 2016. Persons who arrived in Switzerland over the past 10 years and who are foreign nationals born abroad were interviewed. The stratified sampling approach allows for a satisfactory representativeness of 11 groups defined by the country/region of origin and interviewed in six different languages and for a weighting scheme based on the selection probability, participation rate and adjusted for the size of the reference population. All results presented in this chapter are based on weighted data.3 Only individuals of working age (between 24 and 65 years old) were interviewed.

Here are the 12 variables we used to operationalize transnationality and its three main dimensions ('I' stands for indicator):

Concerning *mobility* itself:


Concerning *transnational relationships before* coming to Switzerland:

	- … and *after* immigration we include questions asking whether respondents:

Finally, for *attachment, interest and involvement*:


<sup>3</sup>For more details concerning the origin groups, interview languages, weighting procedure, and further statistical aspects, see Chap. 2 in this volume.


There is one important limit concerning the data of the Migration-Mobility Survey. We had to concentrate our analysis on the country of origin. There are no responses in the data concerning multipolar mobilities or feelings of attachment beyond the country of origin.

Concerning the explanatory variables, we focussed on the following factors:

The *region of origin*<sup>4</sup> as a proxy for migration regimes, border politics and legal capital (Glick Schiller and Salazar 2013; Moret 2015), which have an effect on mobility, migration and transnationality. Legal capital means having the "right passport" to be able to be mobile and not be touched by migration restrictions.5 Switzerland is characterized by a dual dynamic; due to the bilateral agreements with the European Union, European citizens have unlimited mobility, whereas for so-called third-country nationals, important legal barriers to entering the country and to being mobile exist. Hence, we assume that EU citizens are also transnationally more mobile.6 Also related to migration regimes is the *type of residence permit* (settlement (C), residence (B), short-term (L), diplomat/international civil servant (DFAE)/and their families (Ci)), each endowed with a certain set of rights.

*Educational level* (no post-compulsory education/secondary/tertiary degree) is an important variable in terms of resources because it is an asset that allows overcoming migration restrictions; being highly qualified can be an entry ticket to Switzerland (for non-European citizens).

Whether respondents feel they have been *discriminated against* in Switzerland, *gender* and the *labour market status* (in labour force/in training/out of the labour force) are further variables included in the analysis.

<sup>4</sup>We have used the country of birth to define this variable. In 94% of cases, the country of birth corresponds to the "country of origin" self-defined by the respondents.

<sup>5</sup>See the global passport power ranking, https://www.passportindex.org/byRank.php. Accessed 16 June 2018.

<sup>6</sup>We have further regrouped nationalities/origins as follows: Northern and Western Europe (Germany, France, Austria, Netherlands, Belgium, Luxembourg, UK, Ireland, Sweden, Denmark, Norway, Finland, and Iceland); Southern EU (Italy, Spain, Portugal, Malta, Greece, and Cyprus); Western Balkans and Turkey (Kosovo, Serbia, Macedonia, Bosnia-Herzegovina, Albania, Montenegro, and Turkey); Eastern and Central EU (Bulgaria, Baltic countries, Czech Republic, Croatia, Poland, Romania, Hungary, Slovakia, and Slovenia); and South and Central America (all American countries but the United States and Canada, classified as North America). In the cases of Africa and Asia, we used standard geographical divisions. The groupings are due to the fact that, for most nationalities, the number of cases is too restricted to draw reliable conclusions; in addition, the analysis has more to do with the type of countries rather than specific countries (EU vs. non-EU, high-income countries vs middle-income, politically stable region vs unstable, and the like).

Furthermore, we included variables that directly affect those resources, notably through processes of accumulation, such as respondent's *age* (split into ten-year brackets) and *length of the stay in Switzerland*.

We will start by displaying descriptive statistics in Sect. 11.4 pertaining to three main dimensions of transnationality – mobility, social networks and attachment. Then, we use regression models in Sect. 11.5 that allow us to determine which characteristics make people more – or less – transnational, that is, the impact of each factor when the others are controlled for. In a last but important step (Sect. 11.6), we use multiple correspondence analysis to identify combinations of transnationality patterns and of common sociodemographic characteristics to identify ideal-typical situations.

#### **11.4 Transnational Mobility, Networks and Feelings of Attachment**

#### *11.4.1 Pre- and Post-migration Transnational Mobility*

The first indicator focusses on pre-migration mobility (see Table 11.1). Most respondents have already spent at least 3 months in a country other than Switzerland or their country of origin before moving to Switzerland.7

Fifty-four percent of respondents have already moved once or twice internationally, 37% three to five times, and 7.5% more than five times. In other words, we find a type of generalized mobility among migrants in Switzerland, which is striking because there are no reasons to believe, in light of the methodological procedures used to select the sample, that very mobile persons should be overrepresented in the sample.

The second indicator pointing to transnational mobility, however to *postmigration mobility*, concerns the number of visits to the (self-defined) country of origin, which shows that most of this population is very mobile (see Table 11.1). Although only 7.7% never visited their country of origin, most respondents return one to two times (39%) or even three to six times (35%) per year. Eighteen percent circulate at least once per month.

In summary, we find a highly generalized mobility among migrants living in Switzerland, which points to two elements: first, migration is accompanied by multiple forms of mobility and, second, mobility is more generalized than is usually assumed in the literature, in which it is often reduced to the highly skilled population.

<sup>7</sup>*Before moving to Switzerland* refers to the question, "Except for Switzerland and your country of birth, in how many countries have you lived for 3 months or more?" This question is completed by "Is it the first time you have lived in Switzerland for three or more months". In the event of several stays, the analysis refers to the current stay in Switzerland.


**Table 11.1** Pre- and post-migration behaviour (in %)

#### *11.4.2 Transnational Networks*

Let us now move on to the second dimension, namely transnational social networks. We initially turn to the channelling and migration sustaining effect of transnational relationships *before* migration. One strong factor for this channelling effect is the fact of joining a partner/spouse already living in Switzerland (see Table 11.2).

One-third of respondents who were in a relationship when they moved to Switzerland state that they have moved to Switzerland together (32%), but it is more common that one partner moved before the other (40%). In addition, some have met their partner in Switzerland (15%), and others have their partner (still) living abroad (13%). The case we are interested in here is that of persons who joined a partner8 who moved first to Switzerland. This scenario represents 21% of the population under study.

Concerning the role of networks in sustaining migration, people other than the partner are also of obvious relevance. Respondents were asked whether they had relatives in Switzerland when they arrived, and we cross tabulate this information with that in Table 11.3. Three in ten had relatives in Switzerland; only approximately 10% followed a partner/spouse and had other relatives in Switzerland. These numbers appear rather low given the importance of "chain migration" in the literature.

Concerning today's transnational social relationships, important information pertains to respondents' partners' country of residence (see Table 11.4).

Surprisingly, we observe that only a small minority (14%) are involved in an intimate transnational relationship. Another indicator of network transnationality concerns *transnational parenthood*. Fifty-three percent of respondents have children, and they were asked whether their children live in Switzerland:

<sup>8</sup>This reason mentioned by respondents is self-reported and hence not linked to the residence permit.


**Table 11.3** Relatives and/or partner already living in Switzerland when respondents arrived (in %)


Source: Migration-Mobility Survey 2016. Weighted results

Approximately one in six (16.5%) migrants have children living abroad, again a low number that however corresponds to the result we presented concerning transnational intimate relationships.

Because preliminary analyses showed that the sociodemographic profiles linked to transnational intimate relationships and transnational parenting were similar, we created a variable "core family abroad". This category includes individuals whose partners and children live abroad (coded 2), who either have children or a partner living abroad (coded 1), whose partners live abroad and do not have children (also coded 1), or whose children live abroad and do not have a partner (also coded 1). Respondents who have neither children nor a partner are obviously excluded from the analysis. All other respondents are assigned a value of zero.

Thirty-one percent of respondents have their "core family" abroad; however, having both one's partner and children abroad is quite rare (only 5%). Twenty-six percent have either a partner abroad and children in Switzerland, or children abroad and a partner in Switzerland, or a partner abroad and no children at all, or children abroad but no partner at all.

Finally, another very important component of transnational social networks can consist of "good friends" (defined by respondents). Compared with the indicators concerning transnational family aspects we discussed earlier, *Transnational friendship* is measured as 53% of respondents reporting having all or most of their good friends abroad (see Table 11.4).

In summary, chain migration is rather low. Moreover, transnational intimate relationships and parenting represent a minority phenomenon. Conversely, transnational friendships are more widespread.


**Table 11.4** Transnational social relationships (in %)

#### *11.4.3 Transnational Feelings of Attachment, Interest and Involvement*

As mentioned above, three indicators give us a sense of how respondents feel attached to their country of origin (namely I8, I9, and I10). Because exploratory descriptive analyses showed that the three questions were very similar in terms of differences across the various sociodemographic categories, we created an attachment index that combines them, which required transforming the variable "do you feel at home when you visit your country of origin" into a 0–7 scale9 and then calculating the unweighted mean of these three 0–7 scales. The answers were then regrouped into brackets and distributed as displayed in Table 11.5.

<sup>9</sup>The answer "Feels at home from first day on" was assigned a value of 7, "At home after an adjustment" a value of 3.5, and "Like a tourist/visitor" a value of zero.


**Table 11.5** Transnational feelings of attachment, interest and involvement (in %)

Approximately one-fifth of respondents give the lowest answers (0–2), approximately two-fifths indicate a medium degree of attachment (3 and 4), and two-fifths a high degree of attachment (5–7). Overall then, the degree of attachment is quite strong, although a non-negligible minority appears to be quite detached, subjectively speaking, from their countries of origin.

Respondents were also asked to say how much they are interested in news and current events in their country of origin and to indicate their involvement in political activities in that country. Previous analyses showed again that I11 and I12 led to similar conclusions in sociodemographic terms; hence, we generated an Interest/ Involvement Index (hereafter III). In a first step, we identified stronger and more organized involvement (sign a petition, make a donation, participate in a demonstration, contact a politician, or work in an association, political party or action group). These answers were assigned a value of 7. The other activities receive a value of 3.5, whereas respondents who were not involved in any activity received a value of 0.

Approximately 40% of respondents express a low interest/involvement (answers 0–2), whereas one in seven expresses a high degree (answers 5–7), with 45% displaying intermediate levels.

To summarize, the dimension linked to feelings of belonging is clearly the one displaying the highest level of transnationality. Feelings of attachment to the country of origin are strong among migrants, although some are detached. Belonging, measured through the interest in news and current events or in terms of political engagement in the country of origin, is linked to a lower degree of transnationality; however, it remains higher than do other forms of transnationality analysed in this chapter.

Having now described the main dimensions of transnationality, we want to learn who among our respondents are the most and the least transnational and the reason(s) for this status, which is the purpose of the next section.

#### **11.5 Regression Models**

The goal of this section is to identify by means of linear regressions (OLS) the sociodemographic characteristics that make people more or less transnational. The transnationality indicators described above have been standardized to be measured on a scale from 0 to 10, with 0 always applied to the least transnational group and 10 to the most transnational one; for continuous variables, this approach is straightforward. For qualitative variables, we proceeded as follows. For a dichotomous variable, the values given are 0 and 10; for variables with three answers, the values are 0, 5 and 10; for four answers, 0, 3.333, 6.666, and 10; and for five answers, 0, 2.5, 5, 7.5, and 10.

If we focussed the analysis on respondents who had to give an answer to each question included in our indexes, we would lose five-sixths of the sample. For instance, some respondents did not have children or a partner for the reference period of the question, whereas others refused to answer one or several questions. Hence, we have decided to include:


An examination of the OLS regression residuals shows a distribution close to the normal distribution, and residuals do not appear to present a specific challenge.10 Although most sociodemographic variables are correlated with one another, this correlation does not pose a problem in terms of multicollinearity.11

We ran four regression models – one for each dimension of transnationality, and the fourth addresses overall transnationality (see Table 11.6). For most variables, the table shows the impact of having a characteristic other than that of the reference category. The year of birth has been entered in a quadratic form (year + year2 ) to detect non-linear evolutions over time, and the same has been done with the year of arrival in Switzerland.

The *region of origin* plays a major role overall, even when controlling for a large array of sociodemographic characteristics. Being from Southern Europe rather than from Northern/Western Europe (the reference group) implies an overall higher degree of transnationality; however, this point holds for the second and third dimensions, but not for the first one. In fact, no one is more mobile than Northern/ Western Europeans. Southern Europeans distinguish themselves as being particularly attached to their countries of origin while maintaining strong transnational networks.

<sup>10</sup>However, data tend to be heteroskedastic. Because this characteristic does not appear to be attributable to a gross misspecification of the models and is more likely related to the standardization method, we chose to use robust standard errors to obtain reliable significance tests.

<sup>11</sup>The highest variance inflation factors hardly reach a value of 3, which is clearly below the usual limits.


**Table 11.6** Determinants of the various forms of transnationality, effect on the 0–10 scales

Note: \*\*\* = significant at 1%-level, \*\* = 5%, \* = 10\*

a European Union Member States

Source: Migration-Mobility Survey 2016. Weighted results

This is interesting because, from a legal point of view, Southern, Eastern and Central Europeans are submitted to the same mobility regimes.

This situation does not hold, however, for transnational social networks because there is no difference between Northern/Western Europeans, Eastern/central Europeans and Asians, whereas people from other regions are less transnational on this dimension. On the third dimension (attachment, interest, and belonging),

Southern Europeans, North Americans and Latin Americans are more transnational than are Northern/Western Europeans, whereas Africans and Asians are equally attached and interested. Interestingly, people from the Western Balkans and Eastern and Central Europe are the least attached.

*Women* are more transnational than men are overall, but this aspect is due to the third dimension of transnationality (attachment, interest, and involvement); they appear to be less mobile than men are, but the difference is not statistically significant. This result is interesting because we can nuance the conclusions found in the literature. Women might be less mobile or less involved in transnational networks, but it is not being a woman *per se* that matters but rather other characteristics of female migrants.

The *highly educated* are more transnational than others are, globally speaking. However, this point is primarily true for social networks and for mobility to a lesser extent. There is little difference concerning the third dimension; if anything, having an intermediary educational level leads to less attachment/interest/involvement than does having a lower or higher educational level. In general, this variable shows less effect than one would expect from the literature, which primarily relies on descriptive evidence.

*Age* has no effect *per se* on mobility and hardly any on transnational networks*;* separate tests not shown here indicate that its effect disappears once we control for the fact of having children and a partner/spouse. However, age has an effect on the index of attachment/interest/involvement; this index increases with age, but less so when respondents are closer to retirement age.

*Duration of the stay* has a curvilinear effect; those who have just arrived and those who have spent the most years in Switzerland are the most transnational, all else being equal, and this point holds for the first (mobility) and the second (networks) dimensions. For the third dimension, the longer the stay in Switzerland, the higher is the subjective transnationality.

*Labour market status* affects the mobility of those in training; they are less mobile than are those who are employed. This factor also affects those who are out of the labour force; they are less transnational in terms of networks, which points to missing resources. Indeed, persons who are out of the labour force and/or have lower incomes in Switzerland have smaller social networks and less frequent contacts (Gazareth and Modetta 2006).

*Diplomats* and other persons having an *international status* (N = 106) are more transnational compared with residents holding a settlement permit (C permit). However, this point is largely due to the subjective dimension, and, surprisingly enough, not to mobility; the high average educational level of this category most likely explains this surprising result. There is one significant difference between B-permit holders (renewable five-year residence) and settlement permit holders (C permit), namely that the former are more transnational in terms of networks than the latter. Holders of short-term permits (L) stand out as being significantly less mobile, which is not surprising given their low "legal capital for mobility", and as being more transnational in terms of networks.

*Having children* reduces mobility and increases network transnationality, whereas *having a partner* affects networks and transnational belonging. Family reasons are therefore a strong predictor for transnationality.

*Feeling discriminated against* affects all three dimensions of transnationality; people are less mobile, and they are more involved in transnational networks and more attached to, and interested in, their country of origin, bearing in mind that we control for the region of origin (which can also be an indirect measure of "ethnicity").

What do these results tell us about the hypothesis of linear, resource-dependent and reactive transnationality?

First, concerning linear transnationality, we have evidence that in fact, among people who have been in Switzerland for 10 years at most, transnationality is generally a given, but on a rather low level (with the exception of transnational attachment). Additionally, the simple equation that the longer the stay, the less transnational, does not hold, bearing in mind that individuals who have been in Switzerland for more than a decade were not interviewed.

Second, we find elements sustaining the idea that transnationality is *resourcedependent*; concerning transnational mobility, *legal capital* in terms of having a European passport and having a settlement permit is indeed a predictor of a high degree of transnational mobility (the least mobile being people from Africa and Central America, and also those with a short-term permit). *Education* is a second element sustaining the hypothesis that transnational mobility is resource dependent, the highly skilled being the most mobile. Conversely, *having children* is a barrier to transnational mobility. A similar picture emerges concerning network transnationality; *education* is again a predictor for high network transnationality. Legal capital does however not appear on this dimension. Instead, *resources related to the socio-economic situation* appear to affect this dimension; being out of the labour force accompanies a low degree of network transnationality. Conversely, having children is a predictor for high network transnationality. Finally, concerning attachment to the country of origin, none of the following appears to have an effect: *legal capital, education, or the situation in the labour market.*

Finally, concerning the hypothesis of "reactive" transnationalism, we cannot identify higher degrees of transnationality in economically marginalized people; rather, they have a lower degree of transnationality as we described above. However, the *feeling of having been discriminated against* affects all three dimensions; it does not lead to higher mobility. In contrast, it leads to higher levels of network transnationality and attachment.

In Sect. 11.6, we will further deepen this analysis, showing how these different variables interact. This will give us a more in-depth insight into the mechanisms that lead to high or low degrees of transnational mobility, networks and attachments.

#### **11.6 Towards a Typology of the Mobility-Migration-Transnationality Nexus**

The following results are based on multiple correspondence analysis (MCA), and on the same indicators of transnationality as in the regression models. This inductive statistical method projects the answers to the above-mentioned questions on a plane that is defined by the main axes that best summarize the information contained in each variable. The dimensions mentioned in the figures do not correspond to the dimensions of transnationalism mentioned above (mobility, transnational networks, transnational attachment and belonging); they are the result of the MCA. In Fig. 11.1, we connect the modalities of various transnationality indicators to observe how they contribute to the two main dimensions of the MCA.

Figure 11.1 shows that mobility defines both axes; today's mobility is strongly linked to the first axis, whereas the mobility prior to the move to Switzerland strongly defines the second axis.

Social networks define two diagonals; the first one goes from the upper left quadrant to the lower right one, from those who followed a partner and/or who had other relatives in Switzerland and/or who have their "core family" in Switzerland, to the opposite combination. The second one goes from the lower left quadrant to the upper right one, from those who have most or all good friends living in Switzerland to those who have most or all good friends living abroad.

In terms of transnational attachment, one diagonal emerges. It goes from the lower left quadrant to the upper right quadrant, from those who have a low level of attachment to the country of origin, a low degree of interest and low involvement in politics, to the opposite combination. This diagonal is almost congruent to the "most good friends abroad vs. most in Switzerland" diagonal.

Figure 11.2 identifies ideal typical combinations of transnational features by identifying five groupings that we have circled. In Fig. 11.3, we retain the same axes based on transnationality indicators, but project so-called *supplementary variables*, here the sociodemographic characteristics used as explanatory variables in the regression models.

The *first ideal type* in the upper left quadrant includes persons who moved to Switzerland when they had relatives and a partner there. They were and are the least mobile, are more likely to have their core family in Switzerland but have most of their friends in their country of origin and to be quite attached to it. Figure 11.3 shows that this ideal type most typically consists of young Southern Europeans who have recently arrived in Switzerland and who feel discriminated against. These persons are more likely to be women. We call this ideal type *the immobile, one-time migrants with linear transnationality*. This type also reflects the deep economic recession that hit Southern Europe, causing huge youth unemployment rates in these countries. This ideal type is the one that appeared as the "norm" in the early literature on transnationality. Glick Schiller and her colleagues (1995) understood the phenomenon as a reflection of basic political and economic global structural transformations and of particular power constellations. However, it is only one among five types.

dimension 1 (37.0%)


Source: Migration-Mobility Survey 2016

dimension 1 (37.0%)

coordinates in standard normalization


**Fig. 11.2** Ideal typical transnationality patterns Source: Migration-Mobility Survey

#### dimension 1 (36.7%)

supplementary (passive) variables: woman RegionOB ExperienceDiscrimin NewEduc LMSitNow2 QuartilesAge YrsinCH coordinates in standard normalization


Source: Migration-Mobility Survey

The *second ideal type* is mostly found in the bottom left quadrant. It includes people who moved several times internationally before coming to Switzerland. These people are now far less mobile, most of their friends live in Switzerland, and they have a low degree of attachment to and of interest in their country of origin and of involvement in political activities in that country. They are more likely not to have had relatives and a partner in Switzerland when they arrived. As seen in Fig. 11.3, this second group is rather made up of highly educated men who are older than average, who do not feel discriminated against and who have been in Switzerland for at least 6 years. We could designate this type as *the former very mobile men, now settled, non-transnational integrated migrants*. This type essentially corresponds to the ideas of theories of integration maintaining that transnational networks and identifications would decline with duration of stay (Kivisto 2001).

The *third ideal type* is entirely on the right side of the figure. It includes persons who were not very mobile before coming to Switzerland but who are mobile now; their core family is abroad, and they have a strong attachment to, interest and involvement in their country of origin. This group typically contains women from Asia, Latin America, Asia, and North America to a lesser extent, who tend not to be in the labour force, have an intermediate educational level and are quite young to middle-aged, whereas Latin Americans are as close to the lowest educational level as they are to the intermediate educational level. We decided to call them *the onetime migrants who are now mobile and transnationally attached to their core family abroad*. This form of the Migration-Mobility-Transnationality Nexus is particularly gendered. It is primarily women who are implicated in transnational care chains (Hochschild 1983).

The *fourth ideal type* is on the bottom half of the figure, close to the first axis, and completely on the right of the vertical axis. This group includes the most mobile respondents, both before and after arriving in Switzerland, and they neither followed anyone to nor joined anyone in Switzerland. They are more likely to have their core family abroad and have an average level of attachment to, interest and involvement in their country of origin. This group most typically consists of persons from North America (and Asians are quite close, although we attributed them to the third type), who have an intermediate to high educational level, and belonging to this group is not a gendered phenomenon. We decided to call them *the very mobile and cosmopolitan, not very transnationally attached*. We can also find this group in the literature, in which it is called the "cosmopolite class" (Beck and Sznaider 2006) or the Eurostars (Favell 2008).

A *fifth ideal type* emerges. This type includes persons who are the most mobile now and were quite mobile previously. They are not particularly attached to their country of origin. They are typically from the Western Balkans and have a low educational level. They were more likely not to have followed their partner or to have joined relatives and are more likely to have their core family abroad now. Belonging to this group is not really gendered (although the answer 'women' is somewhat closer to this group than is "men"). We call them *the hyper-mobile. They are not particularly attached; they are the transnational outsiders*. They appear to

circulate but not to be attached anywhere and not to have any resources. We could even designate this type "reactive mobility".

#### **11.7 Discussion and Conclusion**

The aim of this paper was to identify patterns of transnationality among migrants living in Switzerland. Having had the opportunity not to be limited to one national group allowed us to measure transnationality, to understand cross-group variations, and finally to present a typology of the Migration-Mobility-Transnationality Nexus.

Descriptive statistics showed that migration is accompanied by multiple forms of mobility and that transnational mobility is more generalized than usually assumed in the literature, in which it is often reduced to the highly skilled population. Data showed also that chain migration and transnational core family relationships are of minor importance. Conversely, transnational friendships are more widespread as a phenomenon. These results are rather surprising and call into question common assumptions about migrant networks; in general, family networks are considered more important than are friendship networks, although some recent studies have started to investigate the role of friendship networks in migration processes (Ryan 2015; Herz and Olivier 2012). Finally, transnational attachment to the country of origin and interest and involvement in news and politics are rather strong – even when a minority is clearly detached.

The regression models showed that being North/West European, or having a tertiary education, being in the labour market, having a settlement resident permit and not having children are predictors of a high degree of transnational mobility. Being African, from Central or South America, in training, or having a short-term permit in contrast is a predictor for a low degree of transnational mobility. The most transnational in terms of transnational networks are Southern Europeans, people with tertiary education and people having children. The least transnational on this dimension are people from the Western Balkans, North America, Africa and South and Central America, and those out of the labour force. Finally, people from Southern Europe and from Northern America, women and diplomats display high degrees of transnational attachment, but people who have lived the longest in Switzerland also display high degrees of transnational attachment.

These results primarily support the hypothesis that transnationality is resourcedependent and is to a lesser extent linear and reactive. The most important resources concerning *transnational mobility* are legal capital (European passport or settlement resident permit) and education. Children are a barrier for mobility. *Transnational networks* are also resource dependent, the most important elements here being education and being in the labour force. Finally, concerning transnational attachment, this hypothesis cannot really be confirmed.

Economically vulnerable people do not develop a "reactive" transnationalism; in contrast, they have rather low degrees of transnationality. However, the feeling of being discriminated against leads to strong transnational networks and attachment.

Finally, multiple correspondence analysis yielded further insight into the Migration-Mobility-Transnationality Nexus. Five different ideal-typical configurations emerged, showing different combinations of the sociodemographic and economic variables; the first are the ideal-typical transmigrants of the first references in the literature, which we called *the immobile, one-time migrants with linear transnationality*. Second, an ideal type that fits the integration theory emerged, namely the *former very mobile men, now settled, non-transnational integrated migrants*. A gendered form of transnational care, *namely the one-time migrants who are mobile and transnationally attached to their core family abroad* is a third type. The "cosmopolites" constitute a forth type, namely the *very mobile and cosmopolitan, not very transnationally attached*. Finally, we identified transnational outsiders in which mobility appears reactive, *the hyper-mobile not particularly attached*.

Interestingly, there are two ideal types representing one-time migrants, whereas the three other types incorporate people who were mobile before coming to Switzerland. The one-time-migrant types are the most transnationally attached. In both cases, the probability is high that they are women and rather less educated. Here, this one-time and uni-linear migration occurred either in a situation of economic crisis (type 1) or in a situation of low legal capital (type 3, non-European women). These ideal types incorporate a form of either a linear or a reactive transnational attachment to the country of origin. The latter ideal type is interesting because it appears that post-migration mobility could contribute to maintain or reinforce transnational attachment.

The other three types incorporate much more mobile people, before and after migration. Ideal type 2 is interesting because we find here former "cosmopolites", highly mobile and educated men, who became settled, with most of their friends in Switzerland. Type 4 incorporates what we find in the literature described as "cosmopolites". Interestingly, both types show a low attachment to the country of origin. In other words, intensive circulation appears to loosen ties and feelings towards the country of origin. Something similar can be said for type 5, a very mobile group of less educated individuals; they are 'mobile non-attached', characteristics usually ascribed to highly educated individuals.

In other words, one-time migration that is the result of an economic crisis or low legal capital (no possibility to circulate) is linked to strong transnational attachment. Conversely, high mobility, when linked to high education and legal capital, leads to an attachment to Switzerland, whereas if it is linked to low education and legal capital, it leads to a type of non-attachment to anywhere. These results point to the fact that multiple factors affect the emergence and maintenance of transnationality and that they can be entangled with unexpected results.

Finally, our results not only point to the context dependency of transnationality but also highlight that studying a diverse migration population can occasionally reveal unexpected results and, specifically, allows nuancing a general hypothesis about transnationality.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Chapter 12 Immigrants' Intentions – Leaning Towards Remigration or Naturalization?**

**Ilka Steiner**

#### **12.1 Introduction**

As presented Chap. 1 on the Migration-Mobility Nexus, migration patterns and regimes have undergone considerable transformations in recent decades in Europe, particularly in Switzerland. Today's patterns of migration move on a continuum from long-term and permanent to increasingly more temporary and fluid. The situation is in part due to the establishment of the dual admission regime, with free movement of people applicable to the nationals of the EU/EFTA member states but with other rules for controlling the entry, admission and stays of third-country nationals. However, the increasingly international labour market has also given rise to a new, highly mobile class of young professionals and students – the so-called Eurostars (Favell 2008) –, that profit from these new opportunities. However, the country of destination has few means to prevent their re-emigration (Ette et al. 2016).

Although it is today central to study (highly) mobile immigrants, it is also important to consider other (less mobile) groups of migrants. This point applies in particular to "highly skilled migrants, who live permanently in their host countries" (Harvey 2009). For instance, 64% of EU15 movers were returning migrants (Favell and Recchi 2009). Although this number appears quite high, it suggests that among EU15 migrants, one of three settled more permanently in the host society. Swiss statistics draw a similar picture.1 Of the immigrants who arrived in Switzerland in 1998, the

I. Steiner (\*)

nccr – on the move, Neuchâtel, Switzerland e-mail: Ilka.Steiner@unige.ch

© The Author(s) 2019 293

<sup>1</sup>See Migration-Mobility Indicators of the nccr – on the move, Remigration, http://nccr-onthemove.ch/knowledge-transfer/migration-mobility-indicators/how-many-migrants-settle-in-switzerland/. Accessed 6 March 2018.

Institute of Demography and Socioeconomics IDESO, University of Geneva, Geneva, Switzerland

I. Steiner, P. Wanner (eds.), *Migrants and Expats: The Swiss Migration and Mobility Nexus*, IMISCOE Research Series, https://doi.org/10.1007/978-3-030-05671-1\_12

proportion of those who had left the country after 17 years is 49%. This proportion is extremely high (over 80%) for American citizens, intermediate for EU citizens (62% for Germans and French, 54% for Italians and 34% for Portuguese), and lowest (less than 10%) for those from countries of former Yugoslavia and Sri Lanka.

Studying less-mobile groups is of further importance because foreign nationals who remain in the destination country might eventually apply for and receive citizenship, thus giving them the opportunity to contribute much more sustainably to the destination country's society. When considering the same immigrant cohort of 1998 in Switzerland as mentioned above, 42% of the stayers had obtained Swiss citizenship after 17 years of residence.2 Again, there are important variations in naturalization patterns, depending upon the country of origin. Immigrants from non-EU/EFTA countries were much more likely to be naturalized than were other nationals: While for example less than 10% of Portuguese nationals obtained Swiss citizenship, one third of individuals from the main immigrant countries (France, Germany, Italy) and two-thirds of immigrants from India were naturalized.

The Swiss legislation proposes two different naturalization regimes. Foreigners who have resided for 12 years in Switzerland can apply for ordinary naturalization,3 whereas foreign spouses of Swiss nationals can apply for facilitated naturalization after 3 years of marriage, provided they have lived in Switzerland for a total of 5 years.4 In 2016, on average, 78% of all 43,000 naturalizations fell into the first regime and 22% into the second regime (Loretan and Wanner 2017).

As emphasized by van Dalen and Henkens (2013, p. 1), "In the current era of globalization, understanding the decisions behind international migration is of increasing importance". Research on future migratory projects – be it remigration, settlement or naturalization – will rely on either the revealed-preferences approach or the stated-preference approach. While the former focusses on actual behaviour, the latter, primarily used by social demographers, geographers and psychologists, privileges the understanding of the intentions to migrate or to settle.

Research has in fact shown that intentions can constitute a good predictor of future behaviour (e.g., De Jong 2000; van Dalen and Henkens 2008; Armitage and Conner 2001). However, some authors claim that emigration intentions do not qualify as a good predictor. In fact, social, economic and political constraints can prevent the actual return (Snel et al. 2015), and short-term stays can be prolonged or long-term stays can be interrupted (Engler et al. 2015), as shown by the "myth of return" of many former guest workers (Snel et al. 2015).

Nevertheless, the principal aim of the stated-preference approach is not necessarily to measure future migration behaviour. The approach shows the position of migrants on the continuum of remigration-settlement-naturalization intentions and

<sup>2</sup>See Migration-Mobility Indicators of the nccr – on the move, Naturalization, http://nccr-onthemove.ch/knowledge-transfer/migration-mobility-indicators/how-many-migrants-get-naturalizedover-time/. Accessed 6 March 2018.

<sup>3</sup>Ordinary naturalization criteria were modified as of January 1, 2018, lowering for example the duration of residence to 10 years.

<sup>4</sup> Information from https://www.sem.admin.ch/sem/en/home/themen/buergerrecht/einbuergerung/ erleichterte\_einbuergerung.html. Accessed 2 February 2018.

therefore summarizes the respondent's attitude towards the migration experience. Not least, immigrants' intentions can affect behaviour such as investments in social contacts and skills (de Haas and Fokkema 2011; Carling and Pettersen 2014). Nevertheless, and in contrast to remigration intentions, the decision to naturalize is not only a rational individual choice but also subject to the destination country's legal provisions and administrative practices.

Thus, this chapter aims at understanding the factors that trigger remigration and/ or naturalization intentions in a high-income country setting. Using data from the Migration-Mobility Survey, we consider four types of intentions: naturalization intentions, settlement intentions (neither naturalization nor remigration intentions), remigration intentions, and naturalization and remigration intentions in conjunction. Although settlement and remigration intentions are often studied, naturalization intentions are les considered in the international literature. To the author's knowledge, no paper has thus far investigated migratory projects in which both naturalization and remigration intentions are held in conjunction. Finally, we test in a multinomial logistic regression the effect of several explanatory factors that can be categorized into four groups – demographics, transnationalities, feasibility and preparedness, and integration in the host country – on the four types of intentions.

Section 12.2 presents the framework of this chapter by discussing the conceptual and theoretical considerations, the factors that were found in other studies to be determinant in explaining naturalization or remigration intentions, the main research question of the paper and the hypotheses. We then describe in Sect. 12.3 the data and methods that we used in this article. Section 12.4 presents and discusses the results of the descriptive findings and the regression analysis. A conclusion completes the chapter.

#### **12.2 Remigration and Naturalization Intentions5**

When studying stated migration preferences, divergent concepts are applied: the intention, the plans, the willingness, and the wish. The operationalization of each of these concepts is crucial to avoid misunderstanding when interpreting the results. For instance, the psychological and behavioural consequences of a "wish" are quite different from those of a "plan" (Kley 2011). Moreover, expressing a wish, desire or the willingness to move involves far less consideration of feasibility than expressing intentions to move, making plans, or having expectations (Lu 2011). Steiner (2018) showed for example that remigration intentions are explained by wishful thinking or a feeling of longing to live elsewhere. Planning an emigration in contrast is the result of concrete events, facts and opportunity differentials between the resident country and possible future destination country. Due to data availability, we will focus on the consideration and intentions to remigrate and/or naturalize rather than on the plan.

<sup>5</sup>A similar version of the literature review on remigration intentions will be published in Steiner (2018).

#### *12.2.1 Conceptual and Theoretical Considerations*

Although the literature on remigration intentions has considerably increased in recent years, naturalization intentions are rarely studied. Moreover, remigration and naturalization intentions are mostly considered separately. Two exceptions can however be mentioned: Leibold (2006) and Massey and Akresh (2006). The lack of literature is a paradox because remigration and naturalization can be seen as the opposite cornerstones of integration into the receiving society. Massey and Akresh (2006) showed for example that the plan to become a citizen and out-migration intentions are negatively correlated.

In line with neo-classical migration theory, which understands the decision to migrate as a rational individual choice, economically successful migrants settle permanently in the destination country, whereas emigration is a sign of "failed migration" (Constant and Massey 2002). This line of thought can also be applied to socio-cultural integration or the classical immigrant assimilation theory and to emigration, in which well-integrated migrants settle permanently in the host country.6 Socio-cultural integration refers to the "identification with the home country, social contacts with native citizens, participation in social institutions of the host country and speaking the language" (Snel et al. 2015) and is often approximated by the length of stay in the literature. Thus, the longer migrants stay, the more they become integrated into receiving societies, the more difficult it becomes to return, and the more they are inclined to settle (e.g., van Baalen and Müller 2009), even to naturalize. Thus, return migration is conceptualized as a cause and/or a consequence of "integration failure", or as de Haas and Fokkema (2011) stated, "While 'winners' settle, 'losers' return".

Nevertheless, this explanation falls short for three reasons. First, remigration or return migration can also be the result of a "success" of the migration project, as was proposed by the New Economics of Migration. In fact, initial migration being conceptualized as a strategy to improve the economic situation of the household in the origin country (Stark 1991), emigration, and particularly return migration, is an indicator of economic success (de Haas and Fokkema 2011). Even more so, in today's dynamic global market for human capital, the increased mobility of the highly skilled might be explained by a specific career step and thus an anticipated short duration of stay. As shown by Massey and Akresh (2006), "The bearers of skills, education, and abilities seek to maximize earnings in the short term while retaining little commitment to any particular society or national labour market over the longer term". Thus, the two theoretical approaches are not mutually exclusive or contradictory (Constant and Massey 2002), even less so in the context of highincome countries, such as Switzerland. Hence, the actual remigration results in either a "success", that is, the achievements of one's aspirations in the host country, or a "failure" to achieve these aspirations (Borjas and Bratsberg 1996).

<sup>6</sup>The causal relationship is however not clear. Are the well-integrated predominantly settling down, or do the settled-down integrate better?

Second, the absence of socio-cultural integration as defined above might not always be a factor that explains remigration intentions. In fact, there might be cases of non-integrated ethnic minorities who intend to remain in the country of destination. Moreover, the integration is not necessarily dependent upon the length of stay but rather on the pace of integration. Finally, structural integration and socio-cultural integration do not always go hand-in-hand, thus making it difficult to decide whether the overall migration is a "success" or a "failure".

Third, although remigration and naturalization intentions appear to be opposite outcomes of the migration trajectory, they might be held in conjunction. Acquiring Swiss citizenship when holding citizenship from a country outside of the EU/EFTA for example provides access to the European labour market and thus furthers mobility. In addition, holding both passports (that of the origin country citizenship and the Swiss one) guarantees the possibility of re-entering Switzerland and thus allows a more transnational lifestyle between both countries, even more so after retirement. Finally, and because we are not measuring behaviour but "only" intentions, undecided individuals who are neither satisfied nor dissatisfied with their life might declare fostering remigration and naturalization intentions as possible future options.

#### *12.2.2 Determinants*

What follows is a separate discussion of determinants of remigration and naturalization intentions.

**Remigration Intentions** Although most studies find that the absence of structural integration is more important in explaining remigration intentions than are sociocultural factors, some studies find the opposite (de Haas et al. 2015; Carling and Pettersen 2014). Thus, developing a "one-fits it all" explanation or theory for remigration intentions (Snel et al. 2015) is unlikely because remigration intentions diverge according to the migrant's profile, the contexts in which migration occurs and thus opportunities in both origin and destination country, and initial migration intentions (Güngör and Tansel 2008; Soon 2008; de Haas and Fokkema 2011).

Also, although most studies consider migrants altogether, some scholars focus on one specific type of migrant, referring either to the profile (e.g., highly skilled migrants) or to the stage in their life course (e.g., students and workers). Those studies demonstrate that the determinants influencing remigration intentions differ considerably according to the type of migrant or reason of migration; a business migrant or a student for example does not have the same migratory projects as a family migrant (Mak 1997). Thus, age, which is correlated with reason for immigration (student, work, and family), also presents an important role because it is negatively correlated with the intention to migrate (Coulter et al. 2011; Ette et al. 2016). Nevertheless, Waldorf (1995), analysing return intentions of guest-workers in Germany, noted an increase in the probability of intending to return in the period prior to retirement.

Gender differentials in migratory behaviour are well documented. In particular, concerning labour migration, women are more willing to follow their partner when emigrating than the other way around (Vandenbrande et al. 2006); the literature findings on international migratory intentions are often not conclusive (van Dalen and Henkens 2013). The family situation has been shown to be an important factor, in which singles and childless individuals are more likely to have emigration intentions due to their independence and higher flexibility. However, Waldorf (1995) emphasizes the importance of the spouse's place of residence.

In general, transnational ties affect migrants' intentions to leave the host country. Family encouragement and support in the home country is positively related to intentions to repatriate (Güngör and Tansel 2008; Tharenou and Caulfield 2010). Van Dalen and Henkens (2013) confirmed these results by analysing the role of the presence of a partner in the country of origin as a factor intervening in return intentions. Several studies even found that family and relationship ties in the country of origin play a more important role in return decisions than economic factors do (Harvey 2011).

The social, economic and legal conditions in the migrant's origin country, approximated by the migrant's citizenship, also have an effect (Massey and Akresh 2006). According to Carling and Pettersen (2014), the national origin contributes strongly to explaining differences in return intentions, even after controlling for other background variables. One explanation might be the differing "access" to onward mobility. In fact, whereas return migration is generally an option for voluntary migrants, onward migration might be limited due to the absence of bilateral or multilateral international agreements on the movement and settlement of persons. Thus, for onward migration, the migration policy of the anticipated destination country – which can provide specific conditions for sub-groups defined according to their education or profession for example – might strongly constrain the individual's capability to move and therefore structure further movements. Thus and concerning the labour market, remigration (return and onward migration) for successful and highly qualified migrants might be easier, not least because of their better access and exposure to information. Finally, Ette et al. (2016) found a significantly positive relationship between migrants' rights provided by the current country of residence and permanent settlement intentions. Newcomers who are provided permanent settlement rights or a transparent process towards a secure legal status invest more in the integration process, subsequently extending the intended duration of stay.

Another set of factors that plays an important role is the feasibility of and the preparedness to move (Tharenou and Caulfield 2010). This point primarily refers to the prior migration setting before leaving the country of origin. Previous trips abroad increase the likelihood to leave the destination country (Massey and Akresh 2006; de Haas and Fokkema 2011). Cassarino (2004) argued that the returnee's preparedness refers to a voluntary act that must be supported by the gathering of sufficient tangible (e.g., financial capital) and intangible resources (contacts, relationships, skills, and acquaintances). Such resources can have been brought by the migrant prior to leaving his/her country of origin (e.g., social capital), can be mobilized during the initial migration experience, or can refer to information about postreturn conditions in the country of origin. Moreover, social capital has an influence on migratory intentions (Cassarino 2004), in which highly skilled occupations often require spatial flexibility (van Ham et al. 2001). Several studies also showed a positive correlation between educational level and migration intentions, arguing that higher levels of education increase employment opportunities and access to information (Coulter et al. 2011; Sapeha 2017; Ette et al. 2016), thereby decreasing the relative cost of migration.

An important set of factors constitutes the migrant's embeddedness in and satisfaction with the host country. Therefore, expatriates who are less embedded in the destination country in their career and community have fewer barriers to return and lower costs arising from doing so (Tharenou and Caulfield 2010). Although duration of residence does not automatically determine a higher propensity to settle (di Belgiojoso and Ortensi 2013), a crucial role is played by other factors. Factors that actually have a negative effect on the migrant's intentions to remigrate include living with a partner of the host country (Pungas et al. 2012), having school-aged children (e.g. Khoo and Mak 2000; Massey and Akresh 2006), owning a family business in the current country of residence (de Haas and Fokkema 2011), being integrated into a social network (van Dalen and Henkens 2007, 2013) and speaking the local language of the host country (Steiner and Velling 1992; Dustmann 1999; Ette et al. 2016).

Similarly, migration intentions are also closely linked to satisfaction with life in the country of residence (Mara and Landesmann 2013; Ivlevs 2015; Hercog and Siddiqui 2014), although one must consider that those factors particularly depend upon the comparative advantage of the possible destination country and the residence country. Thus, according to Pungas et al. (2012), over-education in the labour market, and therefore job dissatisfaction, is associated with an elevated willingness to return. However, not only job and career situation and prospects (Sapeha 2017) but also housing situation (Waldorf 1995), personal life and family satisfaction (Khoo and Mak 2000; Jensen and Pedersen 2007), the subjective well-being associated with the stay (Steiner and Velling 1992), or an experience of racism and discrimination (Steinmann 2018; de Haas et al. 2015) are important factors for remigration intentions.

*Naturalization Intentions* As mentioned, the literature on naturalization intentions is much smaller than the one on remigration intentions. Existing studies suggest however that the intention to naturalize is primarily triggered by legal considerations, social and cultural integration and emotional identification with the host country society and less by structural integration (Hochman 2011; Leibold 2006).

In fact, the decision to naturalize is not only a rational individual choice but also subject to formal conditions. Reaching a naturalization decision is a complex process, involving not only the foreigner but also the origin and destination country's legal provisions (e.g., duration of residence) and their administrative practices (Özcan and Institut 2002; Wanner and Steiner 2012).

Mehrländer showed in the German context, for example, that EU citizens (Italians and Greek) present lower naturalization intentions than do Turkish citizens (Mehrländer et al. 1996, cited by Leibold 2006). The main reasons for the interest of Turkish nationals are legal advantages (right of residence, right to vote and freedom of travel in the EU), personal roots in Germany and a lowered level of attachment to the country of origin. Among the arguments against naturalization frequently mentioned, one can find the obligation to abandon one's own nationality, legal disadvantages in Turkey, the loss of right to return and an already secure right of residence in Germany (Sauer and Goldberg 2001; Mehrländer et al. 1996, both cited by Leibold 2006).

Diehl (2002) showed that a plan to acquire the German passport among Turkish citizens is largely explained by emotional identification with Germany, German as a lingua franca among friends, being born in Germany and low religiosity. Language skills, increased duration of residence, social networks or marrying members of the receiving society increase the immigrants' naturalization intentions (Diehl and Blohm 2008, cited by Hochman 2011; Leibold 2006; Zimmermann et al. 2009). In addition, discrimination and negative attitudes towards individuals with a migration background hindered intentions to naturalize (Portes and Curtis 1987; Hochman 2011).

Concerning structural integration, results are contradictory. Although home ownership is often found to increase naturalization intentions (Diehl and Blohm 2003; Portes and Curtis 1987), Massey and Akresh (2006) showed in the United States (US) that those with high earnings and property ownership are actually less likely to intend to naturalize. Although several authors (Portes and Curtis 1987; Yang 1994) find that higher education increases naturalization intentions, Massey and Akresh (2006) do not find significant differences between educational groups. One reason why structural integration might be more weakly associated with naturalization integration could be the fact that "similar conditions govern[…] the rights and obligations of legal permanent foreign residents and naturalized individuals" (Hochman 2011). Thus, due to the relative comfort achieved in connection with permanent residency status, naturalization is not expected to yield a higher utility for labour market integration (Hochman 2011).

Finally, results concerning family characteristics are somewhat inconclusive. Although the respondents' marital status in Hochman's analyses (2011) had no significant effect, Massey and Akresh (2006) showed that married individuals have a lower propensity to intend to naturalize.

#### *12.2.3 Research Question and Hypotheses*

This chapter aims at understanding the factors that trigger remigration and/or naturalization intentions in a high-income country setting. Based on the literature, four hypotheses guide our research.


#### **12.3 Methods and Data**

The analyses are based on the Migration-Mobility Survey 2016 (see Chap. 2). In contrast with other surveys conducted in Switzerland, it focusses on relatively recently arrived immigrants, with a duration of residence of up to 10 years, and thus captures the attitudes towards the migration experience based on a short(er) duration of residence in the destination country. Moreover, and most importantly for our study, it includes information on both remigration and naturalization intentions, two dimensions that are rarely covered in general population surveys or even in specific immigrant surveys.

Consequently, our dependent variable is the immigrants' intentions, based on the following questions (see Table 12.1). "How many more years would you like to stay in Switzerland?" If the response fell between 1–20 years, "How often have you considered emigrating from Switzerland in the last three months?" Concerning naturalization intentions, we consider the following question: "Do you intend to apply for the Swiss nationality in the future?" In addition, because the question on naturalization intentions was asked to the whole sample, we obtain a fourth category of people fostering both types of intentions. Table 12.1 summarizes the construction of our indicator and the four response categories.

The analyses are conducted including the entire sample, 5973 individuals, corresponding to 458,969 weighted observations. In addition to descriptive analyses of the sample and immigrants' intentions by origin, using a multiple correspondence analysis, we run a multinomial logistic regression (see Chap. 7 for model specifications). Based on the availability of a discrete dependant variable that represents four outcomes that do not have a natural ordering, we estimate the effect of different explanatory factors on immigrants' remigration intentions, naturalization intentions or remigration and naturalization intentions. The base outcome is settlement intentions (that is, not having any intentions), and the estimated coefficients were transformed to relative risk ratios.7

<sup>7</sup>RRR are the ratio of relative risks for the outcome versus base category (settlement intentions) for each given covariate compared with a reference category. A relative risk of 2 means twice the risk, a risk of 0.5 implies half the risk.


**Table 12.1** Questions on immigrants' remigration and naturalization intentions

Source: Migration-Mobility Survey 2016. Questionnaire.

Based on the literature and data availability, the following explanatory variables were categorized into four groups of factors. First, basic demographics include age (continuous, 24–64 years), gender, origin (11 groups of the survey: Germany, Austria, France, Italy, United Kingdom (UK), Spain, Portugal, US/Canada, India, South America and West Africa), having (or not having) children and the reasons for immigration. For the latter, a cluster analysis8 categorized the individuals into five groups based on the reasons for immigration they had declared. We thus distinguish between individuals who primarily came for professional reasons, for educational reasons, to accompany the family, to start a family, or for lifestyle reasons/to gain new experiences/for the social network. Second, transnational ties considers partner's place of residence (Switzerland, abroad or not having a partner) and friends' place of residence (Switzerland, abroad, both in Switzerland and abroad). Third, feasibility and preparedness are composed of number of prior international moves (continuous, from 0 to 20 times), holding (or not holding) tertiary education, and income (puts money aside vs. spends all income or more). Fourth, embeddedness in and satisfaction with the host country is composed of satisfaction with the decision to move to Switzerland (or not satisfied), duration of residence (continuous, from 0 to 10 years), experiences (or not experiences) of racism/discrimination in Switzerland and improved (or not improved) labour market situation compared with before migrating to Switzerland.

<sup>8</sup>We applied the K-means method, a partitioning method that iterates between computing K cluster centroids by minimizing the within cluster variance and updating cluster memberships (Hastie et al. 2009).

#### **12.4 Results**

#### *12.4.1 Immigrants' Intentions*

Most of the recently arrived migrants in Switzerland do not express any intentions to remigrate (see Table 12.2). In fact, 34% would even like to naturalize, whereas 35% have no intentions to leave Switzerland or to naturalize. One of four migrants (26%) intends to remigrate and 6% declare fostering both remigration and naturalization intentions.

When asked for the reasons for not wishing to acquire Swiss citizenship, almost one-half of the participants declared that they do not want to give up their current citizenship, 36% do not see any benefit in doing so, and 20% do not want to go through the process, which is too expensive, complicated and long (see Table 12.3). When asked for the reasons they intend to apply for Swiss citizenship in the future, more than one-half of the participants (56%) declared that they feel that they belong in Switzerland, 43% said that their spouse or partner and/or close family members are Swiss, and 32% believe that doing so will give them better professional opportunities.

Surprisingly, men more frequently foster naturalization intentions than women, whereas the opposite applies concerning remigration intentions. No important variations can be observed concerning mean age at immigration over the four categories, although migrants who intend to settle were slightly older at immigration with respect to the mean observed for the total sample; individuals who would like to naturalize and remigrate were the youngest at immigration. Moreover, the educational level appears to affect settlement and remigration intentions, in which tertiary-educated individuals more frequently intent to remigrate and less frequently intent to settle.

The reason for immigrating to Switzerland influences immigrants' intentions. The highest share of immigrants came for professional reasons to Switzerland. Their distribution across the three categories – settlement, naturalization and remigration – is the most similar (30%–35%), indicating a relatively homogeneous group of migrant workers. Nevertheless, they present the highest share of remigration intentions and the lowest share of naturalization intentions. Rather similar to this first group are migrants who declare having immigrated for educational reasons or to accompany the family. Additionally and unsurprisingly, almost one-half of all individuals who came to Switzerland to start a family foster naturalization intentions, whereas only 20% declare remigration intentions. Finally, lifestyle migrants present the lowest shares of remigration intentions (16%) and thus the highest share of individuals who would like to settle (with or without naturalization intentions).

Family status does not play an important role. Migrants with children (57% of the total sample) slightly more frequently present naturalization and settlement intentions and slightly less frequently remigration intentions than do those with no children. Although these distributions indicate that children render an international movement somewhat more challenging, the presence of a child in this particular population, in which 60% hold a tertiary degree, definitely does not present a barrier to remigration.


304


**Table 12.3** Reasons for and against naturalization

Note: Sum does not correspond to 100% because multiple answers could be given to both questions

Source: Migration-Mobility Survey 2016. Weighted results

One further result considers the differing strategies based on the origin. In general, whereas EU/EFTA migrants show remigration intentions more frequently than non-EU/EFTA citizens, the latter more frequently have naturalization intentions and remigration and naturalization intentions, thus supporting – at first sight – some of our research hypotheses (see Table 12.2). Figure 12.1 plots the results of multiple correspondence analyses, entering the origin group variable and an extension of the four types of intentions, thus distinguishing for remigration intentions the destination, that is, onward or return migration. Among the individuals who expressed an intention to remigrate, 65% wish to return to their country of origin.

The different intentions are nicely delimited in the four partitions of Fig. 12.1, and we detect origin-specific strategies. Germans, Portuguese and Austrians foster primarily settlement intentions. The last two are situated the furthest away from naturalization intentions. This point might be explained by the fact that the Portuguese declare having a very strong feeling of attachment to their home country – a feeling that might also explain their relative closeness to return migration in the graph – and Austria does not allow for dual citizenship.

Naturalization intentions are very strong among individuals from France, Italy and from the two non-industrialized non-EU/EFTA regions, that is, West Africa and South America. Compared with the other EU/EFTA origin groups, French citizens are in fact the only group that feels more strongly attached9 to Switzerland than to their country of origin. Individuals from South America often declare fostering nat-

<sup>9</sup>Attachment is measured through the two questions of the Migration-Mobility Survey: "On a scale from 0 (no feeling of attachment) to 7 (strong feeling of attachment), to what extent do you have a feeling of attachment to Switzerland/to your country of origin?"

**Fig. 12.1** MCA coordinate plot, immigrants' intentions and origin groups, 2016, Switzerland Note: 85.8% of the total inertia is explained by the first two axes Source: Migration-Mobility Survey 2016. Weighted results

uralization and onward migration intentions, confirming that holding both in conjunction might be a strategy to access the European labour market for migrants from outside the EU/EFTA area.

Concerning remigration intentions, whereas the Spanish are associated with return migration, British citizens are situated halfway between return and onward migration intentions. Indians are somewhere in between naturalization, settlement and return. Finally, US citizens and Canadians are, on the one hand, associated with onward migration intentions, intentions that are most likely due to their high level of qualification and thus to their professional mobility. On the other hand, this group is also clearly associated with naturalization and return intentions. This result might be explained by a certain discontent by US citizens with their home country and/or the taxes they must pay in their origin country, even when living abroad.

#### *12.4.2 Determinants*

Table 12.4 presents the results of the relative risk ratios of the multinomial regression. Remigration intentions are largely expressed by migrants who came for workor study-related reasons. Thus, considering the effect of unemployment and being a househusband/wife on remigration intentions, we find support for our first


**Table 12.4** Multinomial regression, relative risk ratios and standard errors compared with the intention to settle

Note: \*\*\* p < 0.01, \*\* p < 0.05, \* p < 0.1. Tested for collinearity, and variance inflation factor never greater than 1.24. 58 missing observations were deleted

Source: Migration-Mobility Survey 2016. Weighted results (normalized weights)

hypothesis. Remigration intentions are explained by weak integration on the labour market. Nevertheless, social integration also appears to be decisive because the absence of friends and the partner triggers remigration intentions. The fact that no or weak satisfaction with the migration to Switzerland compared with being satisfied is associated with the highest relative risk ratios for remigration intentions further supports these results. In any case, when not being well integrated, the qualification level might then act as a facilitator for remigration; a tertiary degree provides easier access to other countries' labour markets and thus to mobility.

US/Canadian and UK citizens, who are often involved in expat migratory strategies, show the highest risk ratios to intend to emigrate. The results concerning the link between lowered remigration intentions and origin are less clear; West Africans, Indians and Portuguese present the lowest risk ratios to intend to leave Switzerland. The former is the one group of the sample that originates from less developed countries in which they do not want or cannot return. It is also the group that declares most often having been subject to discrimination (52% of all West Africans declare having experienced situations of prejudice or discrimination in Switzerland in the last 24 months), explaining the lowered remigration intentions when having experienced discrimination.

Naturalization intentions are largely expressed by migrants who came to Switzerland to start a family or for lifestyle reasons and are thus involved in morepermanent migratory strategies. Thus, confirming our second hypothesis, these intentions are found largely explained by social integration (presence of friends and the partner in Switzerland), increasing duration of residence, general satisfaction with the migration experience and absence of discrimination. The last might be explained by either dissatisfaction with the current residence country's society and/ or the presentiment that the outcome of the naturalization procedure might also be influenced by the same anti-immigrant feelings.

Because the migratory strategy is more permanent, even unemployment does not hinder naturalization intentions. Finally and due to the restrictive migratory policy in place for non-EU/EFTA countries that control migrants' entry, admission and stay, citizens from non-industrialized, non-EU/EFTA countries, that is South America and West Africa, had to overcome the highest barriers to migrate to Switzerland and thus present the highest relative risk ratios to foster naturalization intentions compared with settlement intentions.

Thus, related to our third hypothesis, several factors explain remigration intentions and naturalization intentions but in opposing directions. Naturalization intentions are lowered, and remigration intentions significantly increased for most origin groups compared with the West Africans (Austria, France, UK, Spain and US/ Canada) when the partner lives abroad compared with him/her residing in Switzerland, when friends live abroad compared with them residing in Switzerland, and when dissatisfied with the migration to Switzerland. Finally, naturalization intentions are increased, and remigration intentions lowered for lifestyle migrants compared with individuals who arrived for professional reasons.

Interestingly, we find three cases in which the direction of the relative risk ratios is the same, thus contradicting our third hypothesis. First, experiencing discrimination lowers not only naturalization intentions but also remigration intentions. Second, and consistent with most studies on the relationship between educational level and both types of intentions, holding a tertiary degree of education (compared with Secondary I and II degrees) increases the intention to naturalize and remigrate, a result that emphasizes how high educational attainment fosters the migrant's agency. Third, being unemployed at the moment of the survey (compared with employed) increases both types of intentions.

We now turn to the last outcome, that is, the conjunction of naturalization and remigration intentions. First and in contrast to the first two outcomes, the partner's and friends' places of residence do not have a significant effect, whereas the number of prior international moves slightly does. These results might point to a strategy of international mobility in which Swiss citizenship acts as a facilitator. The fact that third-country nationals, except for Indians, present the highest relative risk ratios would then confirm this impression and our fourth hypothesis. Nevertheless, because most of the participants must wait several years to apply for Swiss citizenship, it is not certain that we can interpret these intentions as being strategic. More realistically, they are most likely an expression of the immigrant not being certain about the future, in which both remigration and naturalization continue to present actual options.

Finally, we note two final results. Surprisingly, whether they have children does not influence immigrants' intentions. Although when migrants have children, the relative risk ratios point towards an increase in naturalization intentions, conjunction of naturalization and remigration intensions, and a decrease in remigration intentions, none of the ratios are statistically significant. Additionally, income does not yield a significant effect on any of the three outcomes.

#### **12.5 Conclusion**

In this article, we aimed at obtaining insights into the factors explaining immigrants' intentions. Very generally, we found that most of the recently arrived migrants in Switzerland do not express any intentions to remigrate (69%); of these, 34% would even like to naturalize, most likely reflecting a rather positive impression of their migration experience. This result corresponds to other findings, in which 60% of immigrants intend to stay permanently (Geurts and Lubbers 2017, for immigrants in the Netherlands; Tezcan 2018, for Mexicans in the US).

In their article, Massey and Akresh (2006) view "emigration as a dynamic decision rooted in the migrant's objective circumstances and psychological orientation to the host country". In fact, studying an immigrant's intentions somehow summarizes the migrant's experiences in the destination country and thus might affect his/ her investment in the integration process. We were in fact able to unfold two different strategies. First, immigrants who are involved in a more permanent migratory strategy, and thus have expressed naturalization intentions, are often family or lifestyle migrants who are socially well integrated, have a higher duration of residence and are satisfied with their migration experience. Unemployment even fosters naturalization intentions, most likely because of the more long-term intentions of the stay.

Second, remigration intentions are largely expressed by immigrants who came for work or study reasons to Switzerland. Thus, weak labour market integration (unemployed or staying at home) causes remigration intentions. Primarily citizens from the US/Canada or the UK express remigration intentions; these groups are likely most often involved in highly mobile expat migratory trajectories. Nevertheless, social integration and satisfaction with the stay are also important determinants.

However, the relationship between the level of integration and immigrants' intentions is more complex. In fact, 6% of the sample declared fostering naturalization and remigration intentions in conjunction. Although one could argue that this outcome presents a strategy of international mobility in which Swiss citizenship acts as a facilitator, it is not certain that we can interpret these intentions as being strategic. More realistically, they might be an expression of the immigrant not being certain about the future, in which both remigration and naturalization continue to present an option.

Finally, we would like to emphasize the role of education. In fact, holding a tertiary degree increases all types of intentions by not only increasing the likelihood of obtaining Swiss citizenship but also facilitating access to mobility. This result emphasizes how high educational attainment fosters the migrant's agency to choose whatever migratory trajectory they desire to follow despite the more restrictive migration regime that Switzerland has introduced for non-EU/EFTA nationals.

Unfortunately, several determinants identified in the literature could not be considered in this research primarily due to data unavailability. We were, for instance, unable to integrate factors such as psychological resources (van Dalen and Henkens 2013; Canache et al. 2013; Cai et al. 2014) or societal dimensions (e.g., the political climate towards foreigners (Reitz 1998; Bommes 2012; Ette et al. 2016)).

Also, we also would like to emphasize the dynamic nature of immigrants' intentions. In fact, whatever mechanism triggers or hinders them, they can be altered during the migratory path. For example, sudden life course changes can trigger or hinder remigration intentions, such as starting a new job or having a baby (Kley 2011). In addition, Leibold (2006), citing different studies, mentions financial motives and the education of children as reasons for postponing the emigration decision. Thus, even though they might constitute a good predictor for future behaviour, they should not be taken as a definitive decision that will certainly be translated into reality.

#### **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

## **Part VI Conclusions**

## **Chapter 13 Conclusions**

**Philippe Wanner and Ilka Steiner**

#### **13.1 Introduction**

The changing realities of migration and mobility in recent decades have progressively placed migration in the spotlight of social science research, which has led to the development of new schools of thought. A prominent one, the theoretical and empirical Migration-Mobility Nexus, which sketches a continuum between the two poles of migration – more long-term or permanent movement from one place to another and mobility (multiple, temporary movements between different places) – has recently gained popularity. It has also served as the conceptual and overarching framework for this book. Switzerland presents an interesting case to study, because the country has been characterized (as have been many other European countries) since the end of the 1990s by the establishment of a dual regime of migration and mobility. On the one hand, the free movement of persons regime has been applicable since 2002 to the nationals of the EU/EFTA member states. On the other hand, strict(er) rules controlling entry, admission and stay apply to third-country nationals. Moreover, data are available to study in-depth migratory and integration processes.

This book thus aimed at testing the theoretical Migration-Mobility Nexus and aimed at clarifying *how economic drivers, societal factors and legal norms shape the migration and mobility patterns of today's different immigrant groups in Switzerland*. It investigated not only how permanent and/or temporary today's migration behaviour really is but also how Switzerland's selective regime of

P. Wanner (\*) · I. Steiner

Institute of Demography and Socioeconomics IDESO, University of Geneva, Geneva, Switzerland

nccr – on the move, Neuchâtel, Switzerland e-mail: Philippe.Wanner@unige.ch; Ilka.Steiner@unige.ch

© The Author(s) 2019 317

I. Steiner, P. Wanner (eds.), *Migrants and Expats: The Swiss Migration and Mobility Nexus*, IMISCOE Research Series, https://doi.org/10.1007/978-3-030-05671-1\_13

migration and mobility influences extant patterns of social and professional inclusion and exclusion.

By involving researchers from several disciplines (anthropology, demography, economy, political sciences, statistics and sociology), the book addresses the following central dimensions of arrival and life in the host country. First, mobility behaviour and intentions, from the arrival in Switzerland to settlement and remigration and expressed through transnationality, were addressed in the chapters by Sandoz/Santi (employer support in the migration process, Chap. 3), Zufferey (serial movers, Chap. 4), Dahinden/Crettaz (transnationalism, Chap. 11) and Steiner (intentions to naturalize, settle or remigrate, Chap. 12). Second, three chapters addressed the dimensions of labour market integration and exclusion: Wanner (reason for migration and integration, Chap. 5), Vidal (effect of the migration on the professional life course, Chap. 6) and Pecoraro/Wanner (education and skill mismatch, Chap. 7). Third, social integration was studied by Bennour/Manatschal (feeling of attachment to Switzerland, Chap. 8) and Auer/Ruedin (perceived discrimination, Chap. 9), and political participation was addressed by Hercog (political practices, Chap. 10).

New statistical data are required to precisely describe migration, mobility, and related phenomena. We observe the progressive development of the statistical infrastructure in a period in which new technologies significantly decrease the cost of data collection. In addition to applying the same conceptual framework, all chapters were based on the Migration-Mobility Survey data. The use of the same source of data guarantees a certain cohesion throughout the book. In fact, despite being one of the most important immigration countries worldwide in proportion to its population, and the increasing economic and social importance of recently arrived and highly skilled migrants, Switzerland is ill-equipped in terms of monitoring systems and databases for tracking individual migration and structural and cultural integration. Thus, at the end of 2016, the nccr – on the move performed the Migration-Mobility Survey, which covered 6000 foreign-born migrants, aged 18 years or older at the time of immigration and between 24 and 64 at the time of the survey. The persons interviewed are recent immigrants, with a duration of residence of up to 10 years in Switzerland. Individuals holding the nationality of one of the following 11 countries/regions of origin were surveyed: Germany, Austria, France, Italy, United Kingdom, Spain, Portugal, North America, India, South America, and West Africa.

The survey gathers not only factual information but also information on subjective dimensions, perceptions, intentions, attitudes, and challenges. It covers migratory history, citizenship intentions, education and employment history and current situation, family configuration and household situation, integration (language skills, personal network and transnational ties, leisure activities, and civic engagement), and life in Switzerland. As explained by Steiner/Landös in Chap. 2, the Migration-Mobility Survey joins surveys initiated in other European countries, providing innovative data on recently arrived migrants and thus original information to improve our comprehension of various dimensions related to migration and mobility – information that can be useful not only to researchers but also to policymakers.

In this conclusion, we provide in Sect. 13.2 an overview of the principal results with respect to the three key dimensions mentioned above. Section 13.3 discusses the implications that these results have for the Migration-Mobility Nexus framework, and Sect. 13.4 briefly addresses how these results obtained for the Swiss context apply to the European and international context. Finally, Sect. 13.5 enumerates the main consequences of the results for future migratory policies and research.

#### **13.2 Principal Results**

#### *13.2.1 Increasing Mobility and Transnationality?*

At the end of World War II, migration was largely organized on a temporary basis around seasonal jobs. Many countries introduced policies constantly renewing migratory flows, instituting high mobility regimes. Only with the progressive decrease in the importance of seasonal and temporary permits did settlement progressively gain in importance during the last quarter of the twentieth Century. This pattern appears particularly true for third-country nationals, whose entry, admission and stay are regulated by strict(er) rules compared with nationals of EU/EFTA member states, who profit from the free movement of persons regime. Nevertheless, globalization and delocalization, improvements in means of communication and decreases in flight prices led to an important increase in spatial mobility (the number of flight passengers, which was 1 billion in 1990, was multiplied by 3.5 in 25 years1 ).

Statistically speaking, little is known about individuals who have undertaken multiple international migrations, a pattern that can take different names in the literature, such as circular, repeated, secondary, onward, stepwise, or serial migrations. In this context, Zufferey documents this dimension by disentangling the relationships between multiple migrations and individual factors such as level of qualification, origin and family composition. His results demonstrate that multiple migrations are common among recent immigrants, because at least one of two migrants has lived in one or more countries before arriving in Switzerland. Nevertheless, Zufferey finds an important heterogeneity in mobility practices between origins that can be explained not only by individual factors, such as the level of education of the group, but also by geographic and cultural dimensions, such as proximity contributing to reduce the number of international moves. Thus, non-EU migrants tend to be particularly mobile and to undertake a stepwise trajectory to Switzerland.

Expatriates or "expats", a term that generally designates highly skilled workers living outside their origin country, either independently or sent abroad by their employers, generally for a professional purpose and not permanently, are an

<sup>1</sup>The World Bank, Air transport, Passengers carried, https://data.worldbank.org/indicator/IS.AIR. PSGR?view=chart. Accessed 26 June 2018.

emerging group of migrants. They are indeed considered not only very mobile but also "desired" migrants because they are demanded by employers, primarily multinational enterprises, enterprises with a high level of productivity or international organizations. In Switzerland and in particular in the largest agglomerations of Zurich (finance), Basel (pharmaceutical companies) or Geneva (international organizations), the numbers of job opportunities for expats are particularly high. Sandoz and Santi, using both the Migration-Mobility Survey and ethnographic data, analyse employer support at the time of migration. The authors demonstrate the existence of a system of inclusion and exclusion that enables certain groups to access significant corporate relocation support. They confirm the "civic stratification" concept, which suggests that according to some individual characteristics (such as the level of education, gender, and ethnic origin), access to other nation states varies. In particular, highly qualified men from Anglo-Saxon countries are actively given the possibility to become "expats", whereas people with similar levels of qualification and experience but with a different gender, nationality or background have fewer opportunities to access employer support and thus to migrate to Switzerland. In this sense, the very notion of "expat" is a construction that reflects power relations at a global level. However, the authors observe that categories and power relations can change over time, as is shown by the current high demand for highly qualified Indians, particularly those working in the information technology sector.

A further dimension of mobility that nevertheless does not necessarily imply an actual move relates to the transnationality of migrants. Crettaz and Dahinden focus their analyses on three aspects: pre- and post-migration transnational mobilities, transnational social networks, and feelings of attachment to the country of origin. The authors show that transnationality is resource dependent and to a lesser degree linear (with increasing duration of stay, transnationality fades) and reactive (to experiences of discrimination or marginalization). The most important resources concerning transnational mobility are the so-called "legal capital" (European passport or permanent resident permit) and education, whereas unsurprisingly, children are a barrier for mobility. Transnational networks also depend upon the resources available, the most important factors of such networks being holding a high level of education and being part of the labour force. Finally, although vulnerable people do not develop a "reactive" transnationalism, they display rather low degrees of transnationality. Nevertheless, the feeling of being discriminated against leads to strong transnational networks and attachment. Finally, five ideal, typical configurations emerged from their analyses, showing different combinations of the sociodemographic and economic characteristics: (1) the immobile, one-time migrants with linear transnationality; (2) the former very mobile men, now settled, nontransnational integrated migrants; (3) the one-time migrants who are mobile and transnationally attached to their core family abroad; (4) the very mobile and cosmopolitan, not very transnationally attached, and finally (5) the hyper-mobile migrant not particularly attached either to the host country or to the country of origin.

Focusing on a further dimension of mobility that does not involve an actual move, Steiner analysed future intentions concerning remigration, settlement and/or naturalization. Most of the recently arrived migrants in Switzerland do not express any intentions to leave Switzerland (69%). Of this percentage, 34% declare that they intend to apply for Swiss citizenship. Only 26% intend to remigrate. The author unfolded two opposing strategies. Immigrants who are involved in a more permanent migratory strategy are often family or lifestyle migrants who are socially well integrated, have a higher duration of residence and are satisfied with their migration experience. In contrast, remigration intentions are largely expressed by immigrants who moved to Switzerland for educational or professional reasons. Thus, an absence of integration in the labour market increases remigration intentions. Citizens from North America and the UK most often expressed the intention to leave Switzerland; these groups are likely to be involved in highly mobile (expat) migratory trajectories. Nevertheless, social integration and satisfaction with the stay in Switzerland also influence the intentions. The analysis also shows that 6% declare fostering both naturalization and remigration intentions in conjunction, which is most likely an expression of the migrants' uncertainty about the future, in which both remigration and naturalization continue to present an option. Finally, a high educational attainment fosters the migrant's agency to choose whatever migratory trajectory they desire to follow, despite the more restrictive migration regime that Switzerland has introduced for non-EU/EFTA nationals.

As shown by these four chapters, migration and mobility is not necessarily an expression of actual behaviour. Moreover, there is not only one migration or mobility pattern. One can observe a large diversity of situations, referring to the characteristics of the arrival, how migrants manage their ties with the host and/or origin country or future intentions. Factors such as level of education, gender and origin are systematically identified as affecting how the migration trajectory occurs. Sections 13.2.2 and 13.2.3, which are devoted to migrants' integration into the labour market and their social and political participation, will show whether the same factors play a role in how the migrants' inclusionary and exclusionary trajectories are shaped in Switzerland.

#### *13.2.2 Labour Market Participation and Integration*

The demand of the labour market for highly qualified skills, but which also depends upon the sector for low-skilled workers, is one of the most important drivers of today's mobility and of migration to and from Switzerland. The important relationship between the labour market and migration means that labour market demands profoundly affect the socioeconomic composition of the flows. In fact, the dual admission regime inter alia supplies the economy, which is increasingly oriented toward services, with the required highly skilled labour.

The multi-layered market mechanisms have become important drivers of not only the volume and direction of the flows but also the inclusion and exclusion of migrants in national and local labour markets. However, migrants consider themselves relatively well integrated into the labour market, as shown by Wanner; threequarters of the men and almost two-thirds of the women declare an improvement in their work conditions compared with their situation before the move. Additionally, their unemployment rate is low, fewer than 20% of migrants do short-term work or work without a contract, and approximately 60% declare using their skills at work. Objectively, the level of participation in the Swiss labour market greatly depends upon the reported reason for immigration. Not only family migrants but also migrants who arrived in Switzerland for other reasons present a lower probability of job satisfaction and a higher risk of underuse of skills compared with those arriving with a work contract. The author further revealed a gender gap; men are better integrated than women are, regardless of the indicator that is used.

Additionally, Vidal confirms important gender differentials in her analysis on immigrants' labour market trajectory throughout their settlement in Switzerland but also considers their employment situation in the country of origin and the characteristics of the family migration process. In general, although the probabilities of being employed are relatively low immediately after moving to Switzerland, particularly among low-skilled and family migrants, the author confirms an overall improvement in immigrants' employment probabilities during the process of settlement in Switzerland. However, in comparison with men, women are more likely to be unemployed whatever the duration of stay and to be economically inactive or to work part-time. In addition, post-migration employment is lower for tied migrants and family-motivated migrants compared with professional migrants. Moreover, family-motivated migration has only temporary effects on labour insertion of male migrants, whereas it harms employment prospects for women more permanently.

Disadvantages on the labour market are translated in different ways, one of them being situations of educational or skills mismatch. Even when it is challenging to measure mismatch, for different methodological reasons, the results obtained by Pecoraro and Wanner show a substantial risk of a person being overqualified compared with the skills required by a job position. This point is particularly true for migrants from EU28/EFTA not bordering Switzerland, i.e., those from Spain, Portugal and South America. Mismatch is explained by the imperfect international transferability of human capital and concerns, first and foremost, migrants that arrived after the end of their educational path. In fact, recent migrants face the lowest risk of educational and skill mismatches when holding a Swiss diploma or having obtained recognition of foreign credentials. Therefore, foreign diplomas do not secure access to jobs in Switzerland that correspond to the required skills. In this context, the recognition of foreign credentials can be considered a strategy to decrease the risks of over education and skills mismatch, particularly for regulated professions.

Overall, migrants declared themselves satisfied with their professional status in Switzerland, even when they face difficulties with integrating into the labour market and a substantial risk of skills mismatch. The apparent paradox between subjective and objective indicators of integration into the labour market might be explained by opportunity differentials between Switzerland and the origin country. When measuring the migrants' satisfaction, they not only compare their own situation with their expectations concerning the labour force integration in Switzerland but also compare their situation with the one prior to migration, in other words in the country

of origin. Therefore, even when the integration into the labour market is not optimal in Switzerland, migrants can find it positive compared with the anticipated situation in their origin country. Most of them declare an improvement in their working conditions, particularly those who explicitly declare having migrated for professional reasons. One can also link this relatively good level of satisfaction with the state of the Swiss labour market, which is characterized in the mid-2010s by a high level of professional opportunities, low rates of unemployment and economic growth in the areas in which migrants are localized.

#### *13.2.3 Political and Social Participation*

In a federal state such as Switzerland, norms of inclusion or exclusion can differ according to the place of residence. Previous works show in particular that integration policies and attitudes toward immigrants (xenophobia and right-wing voting) vary from one canton to another, which can affect immigrants' national identity in terms of their feelings of attachment to Switzerland. Bennour and Manatschal demonstrate that cantonal reception contexts matter, not directly but rather as catalysts. Moreover, in line with assimilation theory, non-citizens' feelings of attachment to Switzerland increase with time spent in Switzerland. Inclusive cantonal reception contexts and liberal cantonal integration policies in particular amplify this positive effect of years of residence on immigrants' national identification except in the most restrictive cantonal reception contexts. These results are challenging because they demonstrate that cantonal policies can influence the pace of social integration and the migrants' well-being. Therefore, one can wonder whether good practices of local and cantonal policies must be developed to avoid situations leading to a poor integration of migrants or even to social exclusion. This latter element, which is part of the Migration-Mobility Nexus, was investigated in more detail by Auer and Ruedin.

Shedding light on the mechanisms of perceived discrimination, Auer and Ruedin investigated who, among recent immigrants, is more likely to feel discriminated against and report it when asked in a survey. They examine not only the societal but also the economic sphere and the opposition between the dual regime of migration and mobility and individual contexts. In fact, to personally feel discriminated against, people must be aware of the differential treatment and perceive it as unjust. The authors show that discrimination reported by the interviewer in a survey depends substantially upon individual traits, including aspects that shape whether discrimination is accepted and whether immigrants feel attached to the host society. Although respondents report less discrimination if their job situation has improved after migration, people more likely report discrimination when they originate from countries in which the national legislature represents ethnic minority groups relatively well. Earlier difficulties related to the migration process and the lack of supporting networks continue to affect the perception of unfair treatment. Moreover, this result shows that individuals distinguish to a surprising degree between discrimination in and outside the work environment. For instance, when they are proficient in the local language, respondents often report discrimination in the workplace and less often in a public environment. This distinction between discrimination in the workplace and discrimination in public also depends strongly upon the immigrant's origin. By analysing the interdependency between the two spheres and the perceived discriminations, one can observe that the current policies focusing on objective discrimination fail to address social cohesion.

Finally, mechanisms of inclusion and exclusion very much depend upon the migrant's citizenship. In fact, migrants lack the same political rights that nonmigrants have unless they obtain Swiss citizenship. A few localities and cantons provide them with the right to vote, and some even offer them the right to be elected (the cantons of Neuchâtel, Jura, Vaud, Fribourg, and some communes of the cantons of Appenzell Outer-Rhodes, Basle-City and Grisons) but only at the communal level and not at the cantonal or federal one. In this context of limitation of political participation, Hercog studied whether foreign residents are making claims on the political system in other forms and the extent to which they are interested in the Swiss political system. By observing different approaches to residents' civic engagement, which are not restricted to the practices of full citizens of the state, one can observe that migrants are more likely to be engaged in contacting activities if they have a good command of the local language, have lived in Switzerland for a longer time and express a high level of political interest. In contrast, a high level of education does not increase the probability of acting in the representative political sphere by contacting, donating to or joining a political party or organization. Overall, there is an interest in Swiss politics and in politics in the origin country; these interests can differ according to the migrant's origin. This interest is not linked to the different forms of engagement but rather to the degree of openness toward Swiss and foreign societies.

#### **13.3 Implications for the Migration-Mobility Nexus**

As argued in Chap. 1, when examining the linkage between mobility and migration studies, researchers must begin by questioning some of the assumptions underlying the standard definitions of international migration that are located within the framework of the international state system. One example in the present book is concerned with the highly mobile, highly educated expat migration, which presents a category of "wanted" migrants with specific migratory strategies of its own. The dynamic interaction between economically driven mobility and state-driven migration thus reproduces in some cases older forms of inclusion and exclusion but also creates new ones, for instance in the case of Indians. Additionally, the existence of several categories of transnationality and thus of mobility that do not necessarily involve an actual further move call into question the current categorizations of migrants.

In general, studying immobility, whether through transnationality practices, future migratory intentions, or the level of attachment to the host country, was shown to be an important aspect of the Migration-Mobility Nexus – being mobile without moving or migrating without further mobility.

Moreover, several analyses in this book show that the narrow focus on the nationstate, which is predominant in classical migration studies, must be circumvented to obtain original further insights into the Nexus by, for instance, focusing on the subnational policy level or going beyond practices of political participation that are linked to the nation states, because these new, selective and temporary patterns of mobility lead to a broadening of political action repertoires.

However, the linkage also raises the issue of the agency of those who are or have been literally on the move. In fact, structural drivers of migration and mobility, such as economic, societal and legal norms and factors, are closely linked to the migrants' agency, that is, their socio-demographic and economic characteristics, which in turn can act as constraints or facilitators to migration and mobility. The analyses in the present book have all emphasized that the temporality of today's migration and mobility behaviour and the economic, social and political inclusion and exclusion largely depend upon an interplay of three individual factors: gender, level of education and origin.

The effect of gender is observed in different chapters of this book, systematically with a less favourable situation for women, clearly raising the issue of gender equality. For instance, in comparison with men, women are more likely to be unemployed, to be inactive or to work part-time. Women are disadvantaged not only in the labour market but also more generally in society. Gender is often linked to the motive of migration, translating thus to the position in the family, and refers to the complex of problems of tied migrants. In fact, women migrating for family reasons are frequently dependent upon a man – the male breadwinner – who in turn will be better integrated not only into the labour market but also socially compared with the woman.

In today's society, moving from one country to another for professional reasons provides the best foundation to reach the objectives of the migration project. Accompanying a partner (generally the male spouse) to a foreign country, in contrast, is occasionally challenging, particularly when one cannot reconcile family life with professional aspirations. In this situation, women not only face poor integration into the labour market compared with their own aspirations but also face difficulties in social integration and declare a lower level of satisfaction with the migration, compared with men.

Irrespective of gender, the level of education and the origin structure migratory flows through constraints imposed by the migration policy in place and thus must be situated at the structural level. In fact, the dual admission system guarantees free movement for citizens of the EU/EFTA member states, whereas stricter rules apply to third-country nationals, with the exception of highly qualified personnel from these origins.

However, the level of education, obviously referring to the skills available (human capital), was also shown to foster the migrants' agency to choose whatever migratory trajectory they desire to follow, despite the more restrictive migration regime that Switzerland has introduced for non-EU/EFTA nationals, and to improve their economic, social and political integration and participation. This point is particularly true for high-income European countries with economies oriented toward high value-adding activities. Switzerland, which attracts multinational companies, and which has been characterized in the last 50 years by a rapid specialization of the labour market, is a very good example of an economy not only requiring highly skilled migrants but also largely depending upon them. In that context, "elites" or "expats" are not only (more or less) accepted, as is true for the other groups of migrants but are even attracted and invited to move to Switzerland. Social and structural integration is for those highly skilled migrants "organized" or "negotiated" by either the employers (who assist them to that end) or the pairs (i.e., other migrants that provide them with information and support to understand the specificities of Swiss society). For the other groups of migrants, in particular the low-skilled ones, arriving in Switzerland for family reasons or through other non-professional networks rather than for professional reasons, the level of inclusion depends more upon their abilities to adapt themselves to the host country society and to find a job. Networks and family already living in Switzerland are then the main providers of support to achieve that purpose.

Finally, the origin also appears to be a dimension that interacts closely with inclusion or exclusion. Almost all the analyses show the diversity of situation according to the origin, particularly pointing to the differences in terms of integration/inclusion between migrants from Southern Europe and Western Europe, the latter being in a more favourable position. The analyses also point to the fact that migrants from the southern continents (West Africa and South America) are more frequently concerned by the difficulties of integration or inclusion. Differences were observed in terms of social participation, reporting of discrimination or objective position on the labour market.

Such results can translate to different realities. First, understanding how the host society functions and having the capacity to be included can depend closely upon socio-cultural proximity (in terms of language for instance) and on geographical proximity. Second, some origins (in particular non-EU/EFTA citizens) are subject to legal barriers to the labour market that can contribute to their exclusion. Different origins, in particular West Africans, also mention discrimination based on their skin colour or migrant origin, observed either in the labour market or more broadly in other domains of the society. By increasing the difficulties associated with obtaining access to not only an adequate job but also housing or healthcare for instance, discrimination based on origin leads to social exclusion.

The analyses in this book thus showed that all three factors represent a specific dimension that must be considered when analysing migrants' behaviour, integration or social and political participation.

#### **13.4 The (Inter)National Dimension**

The present study took the Swiss case as a laboratory to empirically study the Migration-Mobility Nexus and thus changes in an advanced post-industrial society. Switzerland surely presents a specific context, inter alia with respect to its geographical position in the middle of Europe, its high GDP per capita, high level of health and good quality of life in general, and the high share of highly skilled migrants in the immigration flows recorded during the last two decades. In addition, compared with other countries, Switzerland is an interesting case to study migration and in particular mobility behaviour because immigrants obtain a permit – and are registered in population registers – provided they are in possession of an employment contract valid for at least 3 months. Due to this registration, they can be included in samples and surveyed, as was done in the Migration-Mobility Survey. Thus, not all analyses might be replicable in other contexts, or when more mobile groups can be included, the sociodemographic and economic composition of immigration flows might yield differing results.

However, the country is also part of an integrating Europe, the Schengen/Dublin area, the European service economy and general globalization dynamism. In addition, in terms of the composition of migratory flows, Switzerland has several points in common with other countries, such as the diversity of origin countries, skills and particularly the dichotomy between professional and family migration, a distinction that very much influences current patterns of inclusion and exclusion in Switzerland. Even when not directly transferable to other countries, the obtained results can positively stimulate the international debate on the intervening factors of migrant inclusion and exclusion in Europe. In particular, the nature of the relationships among motive for migration, professional integration, and social integration that are observed for Switzerland is most likely akin to those of other countries. Because labour market participation can trigger integration, social integration in particular, non-professional migrants need specific attention because their professional participation was shown to be absent or weaker, a situation that might dampen their rhythm of social integration.

#### **13.5 Implications for Future Research**

Finally, the analyses present limitations. Three recommendations can also be advanced concerning further research. First and very specific to the data on which all analyses are based in the present book, it appears important to dispose of information that is representative of the entire immigrant population, the Migration-Mobility Survey being representative of 68%. Specifically, migrants from Eastern Europe (in particular Poland, which is an emerging community) and the Balkans are excluded. Thus, not only important source countries but also origins that occasionally present a specific educational and occupational composition are excluded. Such exclusion might trigger different results with respect to the inclusionary and exclusionary dynamics. Second, there is a need for longitudinal data to better comprehend the pace of integration of recently arrived immigrants. Concerning the Migration-Mobility Survey, both extensions were addressed in the second wave that was performed in fall 2018. Finally, and although presenting a major challenge, it appears important in a globalizing world to produce internationally comparable data at the European level. As discussed in Chap. 2, definitions of migrants can vary, hindering or at least impeding such conjoint data collections. Nevertheless, comparable data could help to question some of the assumptions underlying the standard definitions of international migration, located within the framework of the international state system. Moreover, from a comparative perspective, investigating the Migration-Mobility Nexus would help to gain further insights, namely, how national legal and economic contexts and societal factors structure migration and mobility patterns in an integrating Europe.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.