**Research for Policy**

Roel Jennissen Mark Bovens Godfried Engbersen Meike Bokhorst

# Migration Diversity and Social Cohesion

Reassessing the Dutch Policy Agenda

### **Research for Policy**

### Studies by the Netherlands Council for Government Policy

### **Series Editors**

J. E. J. (Corien) Prins, WRR, Scientifc Council for Government Policy, The Hague, Zuid-Holland, The Netherlands F. W. A. (Frans) Brom, WRR, Scientifc Council for Government Policy, The Hague, Zuid-Holland, The Netherlands

The Netherlands Scientifc Council for Government Policy (WRR) is an independent strategic advisory body for government policy in the Netherlands. It advises the Dutch government and Parliament on long-term strategic issues that are of great importance for society. The WRR provides science-based advice aimed at: opening up new perspectives and directions, changing problem defnitions, setting new policy goals, investigating new resources for problem-solving and enriching the public debate.

The studies of the WRR do not focus on one particular policy area, but on crosscutting issues that affect future policymaking in multiple domains. A long-term perspective complements the day-to-day policymaking, which often concentrates on issues that dominate today's policy agenda.

The WRR consists of a Council and an academic staff who work closely together in multidisciplinary project teams. Council members are appointed by the King, and hold academic chairs at universities, currently in felds as diverse as economics, sociology, law, public administration and governance, health and water management. The WRR determines its own work programme, as well as the content of its publications. All its work is externally reviewed before publication.

### **The Research for Policy Series**

In this series, we publish internationally relevant studies of the Netherlands Scientifc Council for Government Policy. Many of the cross-cutting issues that affect Dutch policymaking, also challenge other Western countries or international bodies. By publishing these studies in this international open access scientifc series, we hope that our analyses and insights can contribute to the policy debate in other countries.

### **About the Editors**

Corien Prins is Chair of the WRR and Professor of Law and Information Technology at Tilburg Law School (Tilburg University).

Frans Brom is Council secretary and director of the WRR offce. He also is Professor of Normativity of Scientifc Policy Advice at the Ethics Institute of Utrecht University.

Roel Jennissen • Mark Bovens Godfried Engbersen • Meike Bokhorst

## Migration Diversity and Social Cohesion

Reassessing the Dutch Policy Agenda

Roel Jennissen The Netherlands Scientifc Council for Government Policy (WRR) The Hague, The Netherlands

Godfried Engbersen The Netherlands Scientifc Council for Government Policy (WRR) The Hague, The Netherlands

Mark Bovens The Netherlands Scientifc Council for Government Policy (WRR) The Hague, The Netherlands

Meike Bokhorst The Netherlands Scientifc Council for Government Policy (WRR) The Hague, The Netherlands

This work was supported by Scientifc Council for Government Policy, WRR, 'S-GRAVENHAGE, Zuid-Holland, The Netherlands

ISSN 2662-3684 ISSN 2662-3692 (electronic) Research for Policy ISBN 978-3-031-14223-9 ISBN 978-3-031-14224-6 (eBook) https://doi.org/10.1007/978-3-031-14224-6

© The Editor(s) (if applicable) and The Author(s) 2023 **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 license and indicate if changes were made. . This book is an open access publication.

The images or other third party material in this book are included in the book's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the book's Creative Commons license 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 specifc 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, expressed 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 affliations.

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

### **Preface**

The Netherlands is now a dynamic migration society, attracting people from all parts of the world. As a result, its diversity by origin is increasing. In addition, the country has to deal with more and more transient migration: many immigrants who come to the Netherlands are just 'passing through' and so eventually leave again.

The government must pursue a more active policy to familiarize all new migrants with the Dutch society and to incorporate them into it as effectively as possible. That is the main message of this publication of the Netherlands Scientifc Council for Government Policy (wrr), which is a translation and adaption of the Dutch report 'Samenleven in verscheidenheid: Beleid voor de migratiesamenleving'.

Godfried Engbersen, Mark Bovens, Meike Bokhorst and Roel Jennissen wrote this book. Together they formed a project group. Other people who were involved with this project group included Paul van den Berg, Marjolein Bogaers, Anneke Brock, Emina Ćerimović, Dmitri Berkhout, Sümeyye Ekmekçi, Iris Glas, Anne Haarbrink, Ernst Hirsch Ballin, Hamza Kacha, Anouk Koekkoek, Hanneke Leeuwestein, Anna Sophie Lauwers, Suzanne de Leeuw, Marit van de Mortel, Laura Mulder, Hester Oorbeek-de Jong, Elmar Smid, Will Tiemeijer, Ellen Wiering and Magda de Wit.

The authors wish to thank the experts who were willing to review parts of the report: Mérove Gijsberts, Peter Scholten, Warda Belabas, Jeanine Klaver, Tesseltje de Lange, Tom van der Meer and Talitha Stam.

The Hague, The Netherlands Roel Jennissen

Mark Bovens Godfried Engbersen Meike Bokhorst

### **Contents**





#### Contents


### **About the Authors**

**Roel Jennissen** is researcher at the Netherlands Scientifc Council for Government Policy (wrr). He obtained a PhD in demography from the University of Groningen on the basis of the book *Macro-Economic Determinants of International Migration in Europe* (2004). Currently, his main research interests are crime, international migration and super-diversity.

**Mark Bovens** is a member of the Netherlands Scientifc Council for Government Policy. He is attached to the Utrecht University School of Governance (usg) as Professor of Public Administration. Central themes in his research include public accountability, democracy, the constitutional state and citizenship in the information society, political trust, and success and failure of policy.

**Godfried Engbersen** is a member of the Netherlands Scientifc Council for Government Policy. He is Professor of Sociology and the director of the Sociology Department at Erasmus University Rotterdam. His current research is focused on transnationalism, labour migration from Central and Eastern Europe, and urban inequality.

**Meike Bokhorst** is researcher at the Netherlands Scientifc Council for Government Policy. Her research is about legitimacy of inspectorates and checks and balances of semipublic organisations. She studied philosophy, journalism and political science. In 2014 she completed her doctoral thesis on the legitimacy of regulations.

### **Chapter 1 Introduction**

The Netherlands is a dynamic migration society. It has received more than 150,000 immigrants annually since 2010, and more than 200,000 from 2015 onwards. In fact, the immigration record was broken every year between 2006 and 2020, except in 2012. Many of these migrants leave again over time, but others stay. As a result, migration is currently the country's main source of population growth.

Most immigrants in the twentieth century came from a limited number of countries, primarily Turkey, Morocco, Suriname and the Dutch Caribbean.1 Those days are long past, however. Nowadays they come from all parts of the world and have wide-ranging levels of schooling and reasons for moving. There are also substantial differences in the lengths of their stays.

<sup>1</sup>By the 'Dutch Caribbean' we mean the six islands in the Caribbean Sea under Dutch sovereignty: Aruba, Bonaire, Curaçao, Sint Eustatius, Sint Maarten and Saba. Previously overseas territories, in 1954 these were incorporated as an autonomous political entity within the Kingdom of the Netherlands, the Netherlands Antilles. On 1 January 1986, Aruba seceded at its own request and was granted so-called 'separate status', but remained part of the Kingdom. This was done to distance it from what Arubans saw as the dominant position of Curaçao. On 10 October 2010, the Netherlands Antilles was dissolved as a polity in its own right. Like Aruba before them, Curaçao and Sint Maarten now became separate autonomous 'countries within the Kingdom of the Netherlands'. Bonaire, Saint Eustatius and Saba (known as the 'BES islands'), meanwhile, were reclassifed as so-called 'special municipalities' ('bijzondere gemeenten') of the Netherlands itself. At present, the Kingdom of the Netherlands thus comprises four countries: the Netherlands (the 'mainland' in Europe plus the BES islands), Aruba, Curaçao and Sint Maarten. We classify migration between these different entities as international, because we have no choice but to follow the approach taken by Statistics Netherlands in this respect. We realize that this may raise questions, since all citizens of the Kingdom share the same nationality and it goes without saying that ethnic classifcations are inappropriate. Justifcation for this approach can be found, however, in the fact that although the Kingdom of the Netherlands is a single sovereign nation, some restrictions do exist on freedom of movement between its component countries – although only to the Caribbean islands, not from them to the European Netherlands. Moreover, we apply the same terminology to moves between the European Netherlands and the BES islands, despite these are now offcially one 'country within the Kingdom'. For the sake of readability and in line with standard Dutch usage, we refer people of Dutch Caribbean origin as 'Antilleans'.

A few examples, from different parts of the Netherlands, illustrate this. In Eindhoven – the heart of the Dutch 'Silicon Valley' *–* primary schools are currently dealing with a large infux of children of highly skilled Asian expatriates.2 At one school, almost all pupils in the fnal year are of Dutch parentage but a third of those in the reception classes and as many as half in the nursery groups hail from abroad. By 2030 it is expected that Eindhoven alone will have almost 4000 schoolchildren from highly skilled migrant families. Similar patterns can be seen in other municipalities, too, such as Amstelveen and The Hague. Many of these children attend regular schools, either because international education is too expensive or has no room for them or as a deliberate choice by their parents. But their arrival presents schools with new challenges. For example, it is often unclear how long they and their parents will be staying in the Netherlands – some remain for only a few years, others for the whole of their primary schooling.

This new diversity comes on top of the existing ethnic diversity in Dutch society. Zorgpoli Haaglanden, in the Transvaal neighbourhood, is one of The Hague's largest health centres. Of its 18,000 patients, 90% have a migrant background (frst or second generation). In the past this usually meant Indo-Surinamese or Turkish, but these days the facility also has thousands of Polish patients who speak little or no Dutch. This causes frequent miscommunication and misunderstandings with medical staff, as a result of which the Haaglanden now employs a Polish-speaking GP, practice assistant, physiotherapist and pharmacy assistant as well as personnel with a Dutch, Indo-Surinamese or Turkish background. In all, its current employees speak fourteen different languages.3

How can governments manage the arrival of large numbers of very different migrants? That is the question at the heart of this study. Many different institutions are working hard to provide tailor-made education, healthcare and other services for all these different groups, but that requires a lot of extra effort on their part. People living in many Dutch neighbourhoods have long been used to the arrival of new migrants, but the great diversity of those now coming into the country is making conviviality increasingly complicated. In this study the concept of conviviality refers to the processes of living together and interaction between members of different groups that have become an ordinary feature of social life in a multicultural society.4

The main message of this report is that the government must pursue a more active and targeted policy to familiarize all new migrants with Dutch society and to facilitate conviviality between all these different groups. This is hardly a new theme, but we show that the changed migration patterns of recent decades require a thorough reassessment of the policy agenda.

Managing migration effectively is one of the major social issues of our time. And one with many dimensions to it. We do not pretend that this report deals with all of

<sup>2</sup>See KleinJan and Vissers (2019).

<sup>3</sup>See Landeweer (2018).

<sup>4</sup>See, for example: Gilroy (2004), Wessendorf (2014), and Geldof and Oosterlynck (2019).

them – the theme is too wide-ranging for that – but instead confne ourselves to three core issues the government needs to address.


In this introductory chapter we frst discuss the various forms of diversity related to migration and briefy elucidate the three core issues listed above. We then clarify the terminology used in this report, before explaining how it is structured.

### **1.1 Fourfold Increase in Migration Diversity**

The above examples from Eindhoven and The Hague show how much diversity by origin in the Netherlands has increased and evolved in recent years. These changes take at least four forms.

### *1.1.1 Greater Ethnic Diversity*

First of all, diversity by origin has increased. Most immigrants to the Netherlands in the last century came from neighbouring countries, from former Dutch colonies, from Morocco or from Turkey. Since the end of the Cold War, however, they have been arriving from all parts of the world. Figure 1.1 shows the top 15 countries of origin for net migration in the period 2008–2018. These no longer include the 'traditional' sources of twentieth century migration; in fact, net migration from them was actually negative during this decade – more people from these groups left the Netherlands than entered. As a result of these developments, the Dutch population at the beginning of 2017 comprised people from no fewer than 223 different countries of origin (see also Box 1.1).5

<sup>5</sup>There are more countries of origin than there are nations in the world today. This is because some countries no longer exist, such as the Soviet Union, Czechoslovakia and the Netherlands Antilles (see footnote 1).

**Fig. 1.1** Net migration: top 15 countries of origin, 2008–2018 (×1000; excludes migrants of Dutch origin)

\* Persons from Eritrea who were born before its formal secession from Ethiopia on 24 May 1993 may be recorded as originating in Ethiopia

© WRR (2020) | Source: Statistics Netherlands (CBS)

### **Box 1.1: Shift in Perspective**

In public debate and policy concerning people with a migrant background in the Netherlands, a 'traditional' system of classifcation is still widely used. However, this rough division into 'former guest workers' and 'immigrants from former colonies', with 'Western' and 'other non-Western' as residual categories, has long since ceased to provide a picture of the nation's ethnic diversity. Today, the vast majority of Dutch residents with a migrant background in fact fall into the 'Western' and 'other non-Western' groups. Figure 1.2 shows the extent of diversity by origin when the old 'postcolonial' and 'guest worker' perspective is abandoned.

### *1.1.2 Greater Diversity by Length of Stay*

Secondly, there is greater diversity in lengths of stay. Here, too, a shift has occurred. Most immigrants from the former Dutch East Indies (now Indonesia), Suriname, Turkey and Morocco settled permanently in the Netherlands – even though many did not initially intend to do so. In recent decades, however, we have seen a considerable increase in temporary migration. We refer to this using the term 'transience'. On average, moreover, immigrants in this category are now staying for a shorter period of time. Of all those settling in the three largest Dutch cities, Amsterdam, Rotterdam and The Hague, between 1995 and 1999, for example, almost 16% had

**Fig. 1.2** Two different perspectives on the composition of the population with a migrant background in the Netherlands, as of 1 January 2017

© Jennissen et al. (2018) | Source: Statistics Netherlands (CBS)

left again after 2 years. By 2006–2010 that fgure had risen to 27%.6 In other places the rate is even higher: major provincial cities saw almost 35% of new immigrants leave within 2 years and in so-called 'expat municipalities' (see below) the fgure was almost 38%.

### *1.1.3 Greater Diversity by Motivation and Status*

Thirdly, there is now more variety in people's reasons for migrating and their socioeconomic status. The 'traditional' immigrants from Morocco and Turkey were lowskilled workers, as are many of those now arriving from Poland, Bulgaria and Romania. But others from these countries and many from elsewhere in the European Union (EU) and from the United States, India and China are highly educated and frst come to the Netherlands as students or to conduct PhD research before fnding employment in information technology (IT) or the fnancial sector, or with one of the many international organizations and businesses based here. Others, such as Syrians, Somalis and Eritreans, arrive as refugees feeing civil war and oppression. In addition, there are differences in legal status, religion, age and a wide range of other characteristics signifcant for the social position of migrant groups.7

<sup>6</sup>Schmeets (2019).

<sup>7</sup>Vertovec (2007), Meissner and Vertovec (2015), and Engbersen and Scholten (2018). Vertovec (2007) coined the term 'superdiversity' to describe the character of current migration: the "multiple-origin, transnationally connected, socio-economically differentiated and legally stratifed nature of international migration". Vertovec's article generated a massive response, in both academia and the policy arena. It is one of the most cited migration articles of the past decade.

### *1.1.4 More Geographical Diversity*

One fnal aspect of migration-based diversity concerns the characteristics not of the migrants themselves, but of the places where they settle. Today's newcomers are not distributed evenly across the country. And the same also applies to the 'traditional' migrant groups and their offspring. As a result, Dutch towns and cities vary widely in their ethnic make-up. Some are highly diverse, others host mainly members of one particular migrant group and others still remain almost exclusively ethnically Dutch.

The diversity of a community can be measured from the probability that two random residents meeting in the street will come from a different background. On the one hand there are places where this chance is relatively small: about one third of the Dutch populations live in a municipality ('gemeente') where the chance is less than 30%. On the other side of the coin are districts – many of them in the Randstad, the conurbation in the western Netherlands – where the odds are relatively high. In The Hague, Amsterdam and Rotterdam, for example, they are greater than 70%. Today, less than half of the residents of these cities come from a Dutch background; the majority have roots in a wide range of other countries. The same also applies increasingly to the suburbs of these three cities, such as Diemen, Rijswijk and Capelle aan den IJssel.

Even when they have a substantial migrant population in common, no two communities are the same in terms of that population's composition. Horticultural districts such as Westland and Zundert, for instance, are home to a large numbers of migrant workers from central and eastern Europe. Former industrial towns such as Gouda, Leerdam and Hengelo, have many 'traditional' migrants from Turkey or Morocco. Highly skilled newcomers from Asia or the English-speaking world are concentrated in expat enclaves like Amstelveen, Wassenaar, Voorschoten and Wageningen. Border communities such as Kerkrade, Vaals and Baarle-Nassau are home to many Germans or Belgians. And some places – Urk, Grootegast and Staphorst, for instance – still have very few residents with a migrant background. These differences refect the many different faces of diversity in the Netherlands, in this report we distinguish different types of municipality along the lines just outlined.

### **1.2 Three Important Issues**

In this report we focus upon three core tasks for government. We discuss each in more detail in a separate chapter later, but below we elucidate them in brief.

### *1.2.1 Better Organization of Reception and Integration*

One of the central messages of this report is that the reception of new immigrants must be better organized and structured. This is largely a task for local authorities and other local actors, such as schools, healthcare institutions and employers. In recent decades these actors have incorporated migrants from Suriname, the Dutch Caribbean, Turkey and Morocco, often by trial and error and with lot of effort on the part of teachers, employment offcers, care providers, police offcers and civil servants. Now that a mode has been found for these groups, however, new ones have appeared – Poles, Indians, Syrians, Eritreans and others – each with its own language, beliefs and customs. For them, too, a way has to be found to provide good care, education, work and security. This demands fexibility and empathy from the community and its institutions. And we have seen above with the example of Zorgpoli Haaglanden in The Hague, that is possible.

Local authorities sometimes have only limited insight into the arrival, presence and departure of migrants in their areas. Some are rather late in developing adequate social policies in response, only doing so when a specifc group has settled locally. They then start organizing welfare, educational or housing services accordingly, but the reality of the new diversity means that those who tailor their provision to the migrant 'groupe du jour' run the risk of always lagging behind relevant developments. Various Dutch regions are experiencing a constant infux of ever-changing migrant groups, a phenomenon which requires more than an ad-hoc response.

Attention thus needs be paid to the *entire migration cycle*, from a point before arrival until the possible moment of departure. How can we ensure that migrants are properly received and given a good start when they frst settle in the Netherlands? What facilities are needed for those who are here only temporarily? What forms of civic integration, basic and advanced, are appropriate for people who settle for a defnite or an indefnite period of time? And how can governments facilitate smooth departures?

The new diversity comes in so many different forms and local variants that a national, one-size-fts-all response is just not going to work. The issues in border towns like Kerkrade and Vaals are simply too different from those in expat communities such as Wassenaar and Amstelveen. Provincial towns like Gouda and Tiel have to deal with groups quite unlike those coming to large cities like The Hague and Amsterdam. The situation in Utrecht or Eindhoven is barely comparable with that in Rotterdam or Almere. All this means that the reception and integration of newcomers have to be shaped primarily at the local level in order to align with the specifc nature and characteristics of the local migrant population.

### *1.2.2 A New Agenda for Conviviality*

Another key issue is conviviality. Large sections of the Dutch population are concerned that increased immigration is putting social cohesion under pressure. According to the Netherlands Institute for Social Research (Sociaal en Cultureel Planbureau, SCP,) 'living together' and 'immigration and integration' have for many years been two of the three principal national problems as perceived by the general public.

In the past, the policy challenge around migration was defned primarily as an issue of *integration*, the idea being that immigrants needed help to bridge their distance from Dutch society. In some places, such as medium-sized towns with one specifc group of migrants, this remains an important policy matter. We argue in this report, though, that in communities and neighbourhoods with a high degree of both diversity by origin and transience, issues of conviviality also need to be addressed. Unlike policies of the past, which focused upon deprivation and multiculturalism, this approach involves society as a whole and not just specifc groups of migrants. Issues of diversity and conviviality affect everyone, including economically successful migrant groups and also those with a Dutch background. Moreover, they are not a matter of restoring a lost unity but about seeking new forms of cohesion in a changing society.

Increasing diversity, by its very nature, reduces the risk of newcomers and established population groups forming disparate 'blocs'. When a society has overlapping 'cleavages' and ethnic, religious, geographical, social and political differences between groups start to coincide, that may endanger social cohesion. But as the diversity of migrants in the Netherlands increases in terms of their origin, length of stay and socioeconomic background, so the chance of such overlaps appearing diminishes. After all, greater diversity means that there are no longer a few relatively homogeneous population groups with a migrant background. This reduces the likelihood of clashing 'us and them' perspectives dominating the national stage and of differences between established populations and newcomers coinciding with differences in religion, language, regional identity and social status. For example, we fnd a huge variety of religions practised and languages spoken amongst 'new' migrants. Moreover, they vary widely in terms of their social background and do not form a homogeneous political bloc. The new diversity can thus mitigate some past concerns about segregation and polarization.

Nevertheless, this does not mean that conviviality will come about automatically. Rather than a risk of homogeneous bloc forming, there is instead the danger of fragmentation and polarization along ethnic lines. The presence of many different groups from wide-ranging backgrounds and cultures increases the chance of triggering sensitivities and mutual misunderstandings. A high degree of diversity by origin makes it more complicated to live together or side by side in neighbourhoods and communities. Earlier research we conducted shows that residents assess local community relations as less positive in neighbourhoods where diversity by origin is high.8 They feel less at home and are more likely to feel unsafe. These diversity effects come on top of existing problems around the integration of specifc groups

<sup>8</sup> Jennissen et al. (2018).

and are not related to the socio-economic composition of the neighbourhood; they are found in wealthier areas, too, and the greater feelings of discomfort and insecurity we identifed affect residents with a migrant background as well.

Conviviality is also more diffcult when there is a great variety in the length of time migrants reside in neighbourhoods and communities. When established residents constantly have new neighbours, this can cause isolation that weakens mutual relationships. International studies show that a higher degree of neighbourhood stability is associated with more local friendship networks and a lower degree with more crime.9 When there is a lot of transience, it is more diffcult to form local networks so that people tend to live 'in parallel', are less likely to call each other to account for antisocial behaviour and experience more social disorganization.

In short, increased diversity of origin and greater transience make it more diffcult for residents to recognize each other, to develop routines and to make contacts. These issues give rise to an important new policy challenge: how to ensure new forms of cohesion and intracommunity relations in diverse social environments. How do you make all residents feel at home in their neighbourhood and community? How do you equip housing providers, schools and voluntary associations to deal with a plurality of languages and cultures? And how do you maintain a vibrant civil society when its members come from many different backgrounds and, in many cases, move on so quickly?

### *1.2.3 Gearing Migration Policy to Social Cohesion*

Finally, we look at the migrants *of the future*. Patterns of migration are not natural phenomena, but result from an interplay of institutional frameworks, changing circumstances, personal decisions and social networks. And to a certain extent government infuence, through legislation and regulations. We therefore consider specifcally whether social cohesion should play a role in migration policy.

### **1.3 The Idiom of This Publication**

Few topics are so in need of precise defnitions as the subject of this report. For that reason, we devote this section to some of the key terms we use. Note that the terminology in this English version refects the specifc Dutch situation, policy framework and idiom, and as such may differ from that current in other countries.

Unless otherwise stated, we here use the term 'diversity' to refer specifcally to ethnic heterogeneity in the context of migration. In applying this restrictive defnition, we acknowledge that elsewhere the term is often used in a broader sense and

<sup>9</sup>See, for example: Sampson and Groves (1989), Bellair (1997), Sampson et al. (1997), Markowitz et al. (2001), and de Hart et al. (2002).

may there also encompass gender, sexual orientation or cultural preference, for instance. We also realize that it frequently has a strong normative orientation.

### *1.3.1 What Do We Mean by 'Immigration Society' and 'Migration Society'?*

In this report we use a demographic defnition of the term 'immigration society'. By that we mean a society in which a signifcant proportion of the population has a migrant background and where immigration is the main driver of population growth.

The Netherlands has met this demographic defnition since the beginning of the twenty-frst century. Nearly one person in four has a migrant background and their representation within the overall population will continue to increase in the coming years, reaching approximately one third in 2050.10 Migration is also having a major impact upon population growth; since 2015 net immigration has exceeded the birth surplus (the number of births per annum minus the number of deaths). In other respects, too, immigration is unmistakably the driving force behind population growth. This becomes apparent when we compare the respective birth surpluses of those with and without a migrant background: the former has been higher than the latter since 2002. And since 2015 there has actually been a mortality surplus in the 'indigenous' Dutch population: the number of deaths per annum has exceeded the number of births. In other words, without immigration the total population of the Netherlands would have shrunk. In the period 2020–2050, its size is expected to increase by almost two million people. But this growth will be confned solely to groups with a migrant background. We look at this in more detail in Chap. 2.

This does not mean that immigration is a feature of *all* parts of the Netherlands. As mentioned above, there is great diversity between – and sometimes also within – communities. A number of municipalities, especially in the north and east of the country, have hardly any residents with a migrant background.

We also speak of a 'migration society'. This is a slightly broader concept, as it also includes emigration. We show in Chap. 2 that transience amongst migrants has increased. For instance, it is believed that three-quarters of newcomers to the Netherlands in 2010 have since left. They have either returned to their country of origin or moved on to somewhere else. Moreover, there are also people with a Dutch background who leave temporarily or permanently. This combination of immigration and emigration creates a high turnover rate within the total population.

<sup>10</sup>See Fig. 2.11.

### *1.3.2 What Do We Mean by Migrants?*

The concept of migration is open to different interpretations. After all, people can migrate both within a country, between cities or regions, and internationally, from one country to another. In this report we only look at the latter.11

But what makes someone an international migrant? The answer to this question has both a geographical and a temporal dimension. Geographically, one has to cross an international border.12 But not everyone who does that is a migrant. Tourists, for example, clearly do not count.

In this report we adopt the defnition used by Statistics Netherlands (CBS). This body, the offcial Dutch statistical bureau, regards someone as a migrant if that person intends to stay in the Netherlands for more than four months. That is the temporal dimension of migration. Incidentally, this defnition excludes asylum seekers since they are not generally included in the national population register unless and until they have secured a residence permit and move out of reception facilities to a more permanent home. Only at this point do they become classifed as migrants, specifcally 'asylum migrants' (Box 1.2).

### **Box 1.2: Different Types of Migrant**

In this report we distinguish between various types of migrant based upon the legal categories used in the Netherlands, which in turn are derived from a person's formal motive for migrating. By that we mean the offcial reason why they decided to come to the country.


11As stated in footnote 1, we also regard moves between the European Netherlands and Bonaire, Sint Eustatius and Saba (the 'BES islands') as international migration, even though they are all offcially part of the same country. The actual motives prompting people to migrate do not always correspond with the grounds on which they are admitted to the Netherlands. In many cases they have multiple motives simultaneously or successive ones over the course of time.

<sup>12</sup> Jennissen (2011c).

### *1.3.3 What Do We Mean by Migrant Background?*

As well as 'migrants', in this report we also refer to people with or from a 'migrant background'. Again, we here use the standard defnition from Statistics Netherlands: someone with a *migrant background* is a person living in the Netherlands who has at least one parent born abroad. Most migrants in the Netherlands therefore have a migrant background, but many people with a migrant background are not themselves migrants. People with a migrant background who were themselves born abroad belong to the *frst generation*; those born in the Netherlands comprise the *second generation*.

Statistics Netherlands determines *country of origin* as follows. Persons whose parents were both born in the Netherlands are considered to be of Dutch origin, irrespective of their own country of birth. All other persons are classifed in the frst instance according to the country of their own birth. If this is the Netherlands, as is the case for people with a second-generation migrant background, the country of their mother's birth determines their country of origin. If that is also the Netherlands, the father's country of birth is decisive.

### *1.3.4 Why Do We Use the Term Migrant Background?*

In a previous publication, we examined whether it is useful and desirable to use origin-based classifcations in research and policy.13 Doing so may have the undesirable side effect of magnifying differences between groups. But we also argued that drawing such distinctions may still be legitimate, subject to certain conditions. For example, it can be useful in detecting differences in health, educational outcomes and labour-market participation which make it possible to implement targeted policies. It can also be used to identify discrimination. One of the necessary conditions is that the distinction should not only be informative, but also as neutral as possible.

### *1.3.5 Why the Second Generation?*

Is it useful and desirable to include the second generation? The argument against this is that the term 'migrant background' sets these individuals apart even though they were born in the Netherlands and therefore have a Dutch background in many respects. On the other hand, some of the detrimental effects of migration, such as linguistic or employment disadvantages, are by no means always limited to the frst generation. We did consider only including those inhabitants of the Netherlands whose both parents were both born abroad in this report, since the literature shows

<sup>13</sup>Bovens et al. (2016).

**Fig. 1.3** Top 10 countries of origin of persons with a migrant background, by generation (×1000) © WRR (2020) | Source: CBS

that language defciencies, for example, hardly ever occur if at least one parent has a Dutch background. This, however, would have made it too complicated to present the longitudinal trends in Chap. 2. Moreover, it would not have substantially changed our empirical fndings. We therefore decided to follow Statistics Netherlands on this point.

Also in line with Statistics Netherlands, we have confned ourselves to the frst and second generations. Figure 1.3 shows that the proportion of second-generation migrants is now quite large, particularly in the case of the traditional groups. The *third* generation – that is, the grandchildren of migrants, but with both parents born in the Netherlands – are thus counted as part of the population with a Dutch background.

### *1.3.6 How Do We Measure Diversity by Origin?*

How can diversity by origin be measured? Dutch researchers and policymakers often consider only the percentage of the population with a migrant background, the percentage with a non-Western background or even just the percentage with a Turkish or Moroccan background. These are all highly problematic indicators. In the frst and last cases, diversity is not in fact measured at all because they can also conceal a high degree of homogeneity. If a neighbourhood has a large number of residents with a migrant background from only one country of origin, then its diversity by origin is low.

We have therefore compiled a diversity index, the so-called Herfndahl-Hirschman Index (HHI), for all Dutch neighbourhoods, municipalities and regions. This produces a fgure, between 0 and 1, indicating the probability that two randomly selected persons in the area in question belong to different groups by country of origin. The higher the fgure, the greater that chance. A low HHI thus indicates considerable homogeneity, a high one considerable heterogeneity. The fgure for the whole of the Netherlands is 0.38. In other words, the chance that two randomly chosen Dutch residents are from different ethnic groups is approximately 40%. In the three largest cities, where approximately 12% of the Dutch population lives, that probability is about 70% (that is, their HHI ≈ 0.7).

We included eighteen different groups in the HHI, using a geolinguistic classifcation of countries (or groups of countries) of origin or groups of countries. They were: Anglosphere countries, German-speaking countries, Scandinavian countries, Mediterranean countries, central and eastern Europe, Arab countries, Latin America, sub-Saharan Africa, south Asia, central Asia, southeast Asia and the Pacifc, east Asia, the Dutch Caribbean and Suriname, Belgium, Indonesia (including the former Dutch East Indies), Morocco, Turkey and the Netherlands – the latter meaning that people of Dutch origin also count as a distinct geolinguistic group.

### **1.4 Outline**

This report is in two parts. The frst comprises Chaps. 2, 3, and 4 and provides a comprehensive description of the principal developments in migration to the Netherlands over the past half century and in policy towards it during that period.


The second part, Chaps. 5, 6, and 7, is more prescriptive in nature. It discusses the three major challenges identifed in Sect. 1.2 and what is known about each of them from the research, before then moving on to make recommendations in Chap. 8.


**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 license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license 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 The Netherlands as a Country of Immigration**

Migration is leaving its mark on Dutch society. The Netherlands may not be a 'nation of immigrants' like Australia, the United States or Canada, where the majority of people are descended from migrants or have a migrant background themselves. But it is a 'country of immigration'. Approximately one person in four was born abroad or has at least one parent who was. We begin this chapter by sketching four current trends in this respect.

First, the number of immigrants per annum in the past decade has been higher than at any time in the previous century.1 As of 2010 more than 150,000 immigrants were arriving in the Netherlands each year, and since 2015 that fgure has exceeded 200,000. Consequently, immigration is now the main source of population growth. Net immigration (immigration minus emigration) has outstripped the birth surplus (the number of births minus the number of deaths) since 2015. Whilst international migration is quite sensitive to the economic cycle and is therefore volatile,2 it is unmistakably a consistent driver of current population growth. This becomes apparent when we compare the birth surpluses of residents with and without a migrant background: the former has been higher than the latter since 2002. And since 2015 there has actually been a mortality surplus in the 'indigenous' Dutch population: the number of deaths per annum has exceeded the number of births, resulting in negative growth.

Secondly, in recent years not only has the scale of immigration changed but so too has its nature. Whereas large groups of migrants once came from a small number of countries, smaller groups are now arriving from all corners of the world. Together with the long-established migrant population and their offspring, this has resulted in a sharp increase in diversity by origin.

<sup>1</sup>That said, just over a century ago, in September 1914, almost a million Belgian refugees fed to the Netherlands following the German invasion. The vast majority stayed only briefy, though, returning home after the Battle of Antwerp ended on 10 October 1914, and are therefore not considered migrants. But approximately 100,000 Belgians would remain in the Netherlands throughout the First World War (see, for example, Lucassen & Lucassen, 2018).

<sup>2</sup>See, for example, Jennissen (2003).

R. Jennissen et al., *Migration Diversity and Social Cohesion*, Research for Policy, https://doi.org/10.1007/978-3-031-14224-6\_2

Thirdly, Dutch society has increasingly become one of 'transient' migration: many newcomers leave again after only a few years. On average, the duration of a migrant's stay in The Netherlands is now shorter than it was a few decades ago.

Finally, we observe that changing immigration patterns have resulted in greater variance between communities in the ethnic composition of the population.

In this chapter we explore these four trends in more detail, since they provide the empirical context for the core policy challenges we present later in this report.

### **2.1 Increasing Immigration**

In the wake of the Second World War, technology in the feld of transport and telecommunications advanced enormously.3 This led to a signifcant decrease in the cost of international, and even intercontinental, migration. Travel became faster and cheaper, and the rise of telecommunications made migration less psychologically punishing as migrants were now more easily able to keep in touch with family and friends in their countries of origin. This was one of the reasons why the scale of international migration increased so much.4

The Netherlands, with its relatively open economy and society, shared in this global development. As can be seen in Fig. 2.1, it experienced rising immigration from the 1950s onwards. Although numbers fuctuated from year to year (the blue line), the overall trend (the red line) was steadily and strongly upward. That consistent increase hides various form of migration, however.

**Fig. 2.1** Migration to the Netherlands, 1946–2018 (× 1000) © WRR (2020) | Source: Statistics Netherlands (CBS)

<sup>3</sup>Nierop (1995).

<sup>4</sup>Castles et al. (2014).

### *2.1.1 Postcolonial Migration*

The somewhat erratic pattern in the course of immigration on an annual basis was caused in part by the so-called 'postcolonial effect'.5 In the 1950s and early 1960s, for example, there was a fairly substantial infux resulting from the decolonization of the former Dutch East Indies (Indonesia). In the period 1974–1980, the independence of Suriname left its mark on immigration levels. And a third postcolonial fow of migrants has come from the Dutch Caribbean.

### *2.1.2 Labour Migration*

The British geographer Paul White states that western Europe as a whole faced several overlapping waves of migration following the Second World War. The frst consisted of low-skilled labour migrants from southern Europe, Turkey and the Maghreb. They flled the gaps at the bottom end of the jobs market created by the unprecedented economic growth of the post-war period.6 This wave ended with the recession of 1973, as a result of which labour shortages disappeared and western European countries introduced more restrictive controls. In the Netherlands, however, this frst wave ended relatively late. Whereas neighbouring countries stopped recruiting labour migrants in 1973, it was another 2 years before offcial Dutch efforts to do so more or less came to a standstill.7 In the meantime, moreover, a large number of illegal immigrants8 were granted residence status.9

After the 1973 recession, low-skilled labour migration to the Netherlands and the rest of western Europe remained relatively modest in scale. This only changed from the 1990s onwards, as more and more people from the former Eastern Bloc countries – in particular Poland – began to fnd their way into the Dutch labour market. Since 2007, fnding work has been the main motive cited by non-Dutch migrants for coming to the Netherlands. That was the year in which workers from the central and eastern European states which had joined the European Union (EU) in 2004 no longer required a work permit, and also when Bulgaria and Romania joined the EU. Even before 2004, however, there was substantial labour migration from Poland to the Netherlands. Those migrants were mainly so-called 'Aussiedler': Poles who also held a German passport.10

<sup>5</sup>For more information on postcolonial migration with the Netherlands as its destination, see Oostindie and Schoorl (2011).

<sup>6</sup>White (1993).

<sup>7</sup>Lakeman (1999) and van de Beek (2010).

<sup>8</sup>By 'illegal immigrants' we mean persons residing in a country without a legal status.

<sup>9</sup>Burgers and Engbersen (1999).

<sup>10</sup>See: Jennissen (2011a), Dagevos (2011).

### *2.1.3 Family Migration*

The frst wave of migration by low-skilled workers, which ended in the Netherlands in 1975, was followed almost immediately by a second wave fuelled by family formation and reunifcation. A substantial proportion of the so-called 'guest workers' from Turkey and Morocco decided to remain permanently and so brought their families to join them. Moreover, this wave of family migration in the wake of labour migration persisted for many years because many children of the original Turkish and Moroccan migrants found partners in their country of origin and brought them over as well. Between 1976 and 2005, family migration was the most common form of immigration in the Netherlands.11

### *2.1.4 Postindustrial Migration*

Starting in the 1980s, the Netherlands was also confronted with what White calls "postindustrial migration".12 This third wave, he explains, consisted largely of asylum seekers, highly skilled workers and irregular migrants. It is their arrival, frst and foremost, which changed the pattern of international migration in western Europe from the end of the Cold War onwards; in the words of Belgian sociolinguist Jan Blommaert, that went from "people from a small number of countries of origin to a small number of host countries" to "people from a very large number of countries of origin to a very large number of host countries".13

In the Netherlands, it was mainly asylum seekers who shaped the picture of immigration during this third wave. One signifcant causal factor was the political unrest in central and eastern Europe after the fall of Communism. The war in former Yugoslavia triggered high immigration fgures in the frst half of the 1990s, as did the Kosovo crisis in the fnal 2 years of that decade. In addition, the Netherlands had to deal with an infux of asylum seekers from Asia and Africa. Although the fgures fuctuated, in general they remained quite substantial into the frst years of the new millennium. Then, following a period of relative calm lasting about a decade, they again rose sharply from 2013 onwards. In 2014 in particular, large numbers of Eritreans registered at Dutch asylum reception centres. This was due not only to continuing instability in their homeland in the Horn of Africa, but probably also because Switzerland tightened its admission policy for this group at around that time. Many therefore seem to have chosen the Netherlands as an alternative.14

<sup>11</sup>That is, for those who did not hold Dutch nationality. For Dutch citizens coming to the country, the motives for their migration are not known.

<sup>12</sup>White (1993).

<sup>13</sup>See Blommaert (2011).

<sup>14</sup>Chotkowski et al. (2014).

Starting in 2014, Syrian asylum seekers also found their way to northwest Europe, the Netherlands included, in large numbers.15

### *2.1.5 Decline of 'Traditional' Migrant Groups and Increase in 'Smaller' Ones*

To illustrate Blommaert's observation, in Fig. 2.2 we have broken down immigration by origin into a few aggregated groups. People from Turkey, Morocco, Suriname and the Dutch Caribbean form the four large 'classic' non-European/Anglosphere origin groups. Since the 1980s, however, their share in this cohort has declined sharply, with the numbers coming from Turkey, Morocco and Suriname in particular falling steadily.16 Immigration from Suriname decreased sharply after 25 November 1980, the end of the 5-year period following the country's independence during which people there could choose between staying and thus automatically exchanging their Dutch nationality for Surinamese citizenship or remaining Dutch – but on condition that they emigrated to the Netherlands.17

**Fig. 2.2** Immigration by origin (excluding Dutch background), 1996–2018 (× 1000) © WRR (2020) | Source: Statistics Netherlands (CBS)

<sup>15</sup>This despite the fact that the vast majority of displaced Syrians stayed within their own country or went to neighbouring ones.

<sup>16</sup>Because citizens of the Dutch Caribbean (see footnote 1 in Chap. 1) are Dutch nationals, they can travel freely between that region and the European Netherlands. Partly because of this, their rate of immigration – unlike that of other three large 'traditional' non-European/Anglosphere origin groups – has not decreased since the 1980s.

<sup>17</sup> van den Broek (2002).

Meanwhile, immigration by Turks and Moroccans declined because of a decrease in international family formation by these two groups. Policy measures to limit marriage migration played a role in this,18 but anyway by this time the size of the Turkish and Moroccan communities, including a large second-generation population, meant that more potential marriage candidates were available closer to home and so there was less incentive to 'import' a bride or groom from the country of origin.

As mentioned above, arrivals from other non-European/Anglosphere countries rose during the postindustrial migration era, mainly as a result of increasing asylum migration. These newcomers came from a whole range of countries, including many which had not previously experienced much migration to the Netherlands.19 In addition, cheap holiday fights to far-fung destinations and the arrival of the internet caused many long-distance romances to blossom. These regularly resulted in marriage migration to the Netherlands and thus also contributed to the rise of the 'smaller' origin groups in immigration statistics.

Since the 1990s, highly skilled migrant workers have come to the Netherlands in fairly large numbers from all parts of the world. These talented individuals are not greatly hindered by the absence of a network of fellow migrants in their destination country, or by language barriers (see Box 2.1). They are often already assured of a job upon arrival and English is increasingly the 'lingua franca' in highly educated circles.

### **Box 2.1: High-Skills Migration as a Source of Diversity**

As of 2015, some 33% of academic staff at Dutch universities were foreign nationals. Of these, 58% were citizens of another EU member state, 6% had another European nationality and 36% were non-Europeans. The top fve countries of origin for this group were Germany (5% of all academic staff), Italy (3%), China (3%), Belgium (2%) and India (2%), followed by the United Kingdom, the United States, Spain, Iran and Greece.

Likewise, in highly specialized companies such as ASML, which produces machinery used in the manufacture of computer chips, we see that high-skills migration results in greater diversity. This frm needs a lot of people with very specialist technical knowledge. Because there are not enough suitable candidates in the Netherlands to fll all these positions, ASML also recruits abroad – mainly through secondment agencies and universities, but also via its own offces in Asia and the United States. In all, no fewer than 115 different nationalities are represented in the ASML workforce. Of the nearly 500 new staff who joined the company in the Netherlands in the third quarter of 2017, some 49% were non-Dutch.\*

\*Source: authors' conversation at ASML.

<sup>18</sup>Bonjour (2007) and Nicolaas et al. (2011).

<sup>19</sup>For a comprehensive overview of the refugee groups coming to the Netherlands, see, for example: Wijkhuijs et al. (2011) and Maliepaard et al. (2017).

Figure 2.2 also shows that immigration from the European/Anglosphere countries has changed in character, with the proportion of people with a German or Belgian background declining. This is due mainly to the sharp increase in labour migration from the former communist EU member states from 2003 onwards.20 Until 2006 this mainly meant immigrants from Poland. Since then, however, Bulgaria and Romania have also contributed substantially. In addition, the reasons already mentioned for the increase in immigration by 'smaller' non-European/ Anglosphere groups apply to this category as well.21 Finally, the relatively strong rise in house prices in the Netherlands has played its part in reducing the share of people with a German or Belgian background in the immigration fgures for the European/Anglosphere group.22

### *2.1.6 Immigration Is Likely to Remain High in the Future*

As Figs. 2.1, 2.2, and 2.3 show, the trend in immigration to the Netherlands is upward. As of the beginning of 2020, there were no indications that this would change in the short term and that immigration would start to decline.

However, there are signs of possible shifts in the types of migrant coming to the Netherlands. Immigration from central and eastern Europe may well fall due to the ageing and shrinking population there, for example, as well as the convergence of prosperity levels between the eastern and western parts of the EU.23

The number of asylum migrants fnding their way to the Netherlands is also likely to decrease slightly. It is seems improbable that the fgures of 2015 and 2016, which saw more initial asylum claims than ever before – with the exception of 1994 – will be reached again any time soon. Statistics Netherlands assumes in its current population forecast that asylum migration will remain at its current level for the next 10 years or so.24 Because conficts abroad are a strong determining factor when it comes to asylum migrant numbers, they fuctuate considerably. Moreover, it is not easy to predict which countries will generate international fows of refugees. It is therefore certainly conceivable that – just as with Eritreans and Syrians today, and with Somalis, Iraqis and Afghans in the 1990s – the future will see an infux from countries that have not previously experienced large-scale migration to the Netherlands.

<sup>20</sup>Black et al. (2010) and Fihel et al. (2015).

<sup>21</sup>This applies even to increased asylum migration, since many asylum seekers came from countries we classify as 'European/Anglosphere', including the former Soviet Union, the former Yugoslavia, Romania and Albania (see footnote 26).

<sup>22</sup>See, for example, van Agtmaal-Wobma et al. (2007).

<sup>23</sup> Jennissen (2011).

<sup>24</sup> van Duin et al. (2018).

**Fig. 2.3** Immigration to the Netherlands, 2000–2040 (× 1000) © WRR (2020) | Source: Statistics Netherlands (CBS)

Statistics Netherlands also assumes that labour and student migration from outside the EU will continue to increase in the future.25 Fuelling the latter, it expects the internationalization of universities to continue.

In short, both past experience and the population forecast from Statistics Netherlands indicate that continuing substantial immigration can be expected for the foreseeable future. Furthermore, it is quite possible that this will include 'new' groups of immigrants.

### **2.2 Increasing Ethnic Diversity**

Not only has the number of migrants to the Netherlands increased, their origins are also more and more disparate. In January 1972, some 9.2% of the population had a migrant background; as of the beginning of 2018, this proportion had risen to more than 23% (see Fig. 2.4). This is a frst indication that diversity by origin has increased signifcantly. Figure 2.5 also shows that the population with a migrant background has become increasingly varied; to demonstrate this, for the years 1972, 1995 and 2018 we divided that population into those with and without a European/ Anglosphere26 background, then broke down the latter group by the four 'classic' countries of origin (Turkey, Morocco, Suriname and the Dutch Caribbean) plus 'other'.

<sup>25</sup> van Duin et al. (2018).

<sup>26</sup>By European/Anglosphere countries, we mean the whole of Europe except Turkey but including Russia and the former Soviet republics in Asia, plus Australia, Canada, Israel, New Zealand, the United States, South Africa and (for historical reasons, see footnote 27) Indonesia/the former Dutch East Indies. See Jennissen et al. (2018) for more details.

**Fig. 2.4** Composition of the Dutch population by origin in 1972, 1995 and 2018 © WRR (2020) | Source: Statistics Netherlands (CBS)

In the early 1970s, the vast majority of those with a migrant background had their origins in Europe or the Anglosphere – predominantly either neighbouring countries or the former Dutch East Indies.27 And, as described in 2.2, by this time the

<sup>27</sup> Immediately prior to the Japanese occupation of the Dutch East Indies (1942–1945), approximately 300,000 Europeans lived in the archipelago. They included about 100,000 'totoks' (Dutch immigrants and their descendants) and 175,000 Indo-Europeans, also known as 'Indos' – the offspring of relationships between Europeans (or Indo-Europeans) and native Indonesians who had been legally acknowledged by the European parent – plus 20,000–30,000 non-Dutch Europeans (Oostindie, 2010). Between the outbreak of the Indonesian War of Independence in 1946 and the New Guinea crisis in 1962, large numbers of people were repatriated from Indonesia to the Netherlands. Since they were mainly 'totoks' and 'Indos', we count people with their roots in the Dutch East Indies or Indonesia having a European/Anglosphere background. This also applies to ethnically non-European migrants in the decades after World War II, such as South Moluccans, Peranakan Chinese and Papuans. From the data in the population register, we cannot distinguish these groups from Europeans and Indo-Europeans with an Indonesian/Dutch East Indies background.

**Fig. 2.5** Size and breakdown by origin of the Dutch population with a migrant background on 1 January 1998 and 2018

© WRR (2020) | Source: Statistics Netherlands (CBS)

Netherlands was already experiencing immigration by guest workers and their families from Turkey and Morocco as well. The share of the four 'traditional' non-European/Anglosphere groups in the population with a migrant background was approximately 10%. Continuing family migration by Turks and Moroccans, as well as postcolonial migration from Suriname – mainly in the period 1974–1980 – and then from the Dutch Caribbean from the second half of the 1980s, increased this share to 34.5% by 1994. From then on, however, that proportion stabilized. This was mainly because the group with another non-European/Anglosphere background grew considerably in size; in 1972 it accounted for just 2.8% of the total population with a migrant background, but by 1995 – fuelled by postindustrial migration – that fgure had risen fourfold. In the most recent year for which we have fgures, 2018, some 23.6% of persons with a migrant background living in the Netherlands were of this so-called 'other non-European/Anglosphere' origin.

Figure 2.5 shows the size and composition by origin of the population with a migrant background in 1998 and in 2018. This reiterates the fact that the Netherlands is a country of immigration, the total number of people falling into this category having increased by more than 1.3 million in those 20 years alone.

Diversity by origin has increased considerably, too. The proportions with roots in the main 'traditional' countries of origin, both non-European/Anglosphere (Turkey, Morocco, Suriname and the Dutch Caribbean) and European/Anglosphere (Indonesia/Dutch East Indies, Germany and Belgium), declined from 70% in 1998 to 54% in 2018. Also striking is that some of the most prominent groups in 2018 had been negligible just 20 years earlier. There were now more people with a Polish28

<sup>28</sup>There has long been a Polish presence in the Netherlands, but the size of this community remained relatively small until the beginning of the twenty-frst century. See, for example: Pool, 2011; Lucassen & Lucassen, 2018.

background than a Belgian or Dutch Caribbean one, for example, whilst the number originating in Syria was only marginally smaller than the fgure for Belgium.

### *2.2.1 Emigration*

Besides increasing immigration, other demographic factors such as emigration, births and deaths have also strongly infuenced the increasing diversity by origin in the Netherlands. We look frst at emigration. Year on year, this fuctuates far less than immigration. In fact, as Fig. 2.6 shows, its level was relatively stable through the second half of the twentieth century with the exception of the 1990s. At frst sight, no clear trend either upward or downward is discernible during this period; the only noteworthy point is a decrease in emigration in the 1950s.29

The Social and Economic Council of the Netherlands (Sociaal-Economische Raad, SER) cites increasing prosperity at home and economic tensions in the

**Fig. 2.6** Immigration, emigration and net migration in the Netherlands, 1946–2018 (× 1000)\* \*Emigration fgures are after administrative corrections. (More information on these administrative corrections can be found in an article by Nicolaas (2006)). For the sake of clarity, emigration is expressed in negative terms

© WRR (2020) | Source: Statistics Netherlands (CBS)

<sup>29</sup>Emigration peaked in 1952, at a level that would not be equalled again until 1995. In the early 1950s the Dutch government published two policy papers advocating net emigration. This was achieved consistently every year between 1951 and 1960. Since 1947, 1948 and 1949 had also seen net migration, however, it is questionable whether this phenomenon was attributable to the government policy of the day. Net migration was considered desirable as a means to alleviate the demographic and employment problems of the time (SER, 1985).

**Fig. 2.7** Emigration from the Netherlands per 1000 inhabitants, 1946–2018\* \*Emigration fgures are after administrative corrections © WRR (2020) | Source: Statistics Netherlands (CBS)

principal destination countries as the main reasons why emigration fell from the early 1950s onwards. This decline was structural in nature, although that is not really apparent from Fig. 2.6. The post-war baby boom caused the overall population to grow by an unprecedented 28% during this period, from 9.3 million in 1946 to 11.9 million in 1963. That led to the emigration rate – the number of emigrants per inhabitant – dropping signifcantly in the second half of the 1950s and early 1960s (see Fig. 2.7). Starting in 1963, it then rose again slightly. As well as Dutch nationals, these emigrants included 'guest workers' returning home. Mainly from Italy and Spain, their departures peaked during the 1967 recession when approximately 26,000 did not have their employment contracts renewed. Almost half of those affected left the Netherlands as a result.30

During the 1970s and 1980s, the scale of emigration from the Netherlands remained fairly stable. And the same applied to the emigration fgure per capita. In the 1990s, however, there was an upturn.31 Emigrant numbers then increased sharply at the beginning of the twenty-frst century, reaching a record so far of 154,200 in 2017. The composition of this most recent group differs signifcantly from that in the frst two decades following the Second World War. Whereas the post-war wave consisted almost exclusively of people with a Dutch background, about two-thirds leaving the country permanently in the frst years of the new millennium had a migrant background.32

<sup>30</sup>Lakeman (1999).

<sup>31</sup>The relatively high emigration fgure recorded in 1996 resulted from the linkage in that year of Municipal Personal Records Database (Gemeentelijk Basisadministratie Persoonsgegevens, GBA) and the Aliens Registration System (Vreemdelingen Administratie Systeem, VAS). This led to the removal from the GBA of approximately 10,000 people who had been erroneously assumed to be resident in the Netherlands (Nicolaas, 2006).

<sup>32</sup>Nicolaas (2006).

Another important difference compared with the 1950s and 1960s is that noneconomic factors nowadays seem to be a major driver of emigration. According to researchers at the Netherlands Interdisciplinary Demographic Institute (Nederlands Interdisciplinair Demografsch Instituut, NIDI) and Statistics Netherlands, a desire to emigrate is now often motivated by negative views of the quality of life in the Netherlands – due to 'overpopulation', pollution, noise and so on – or of Dutch society, often expressed through critical attitudes towards the 'national mentality', the crime rate or multiculturalism, for example.33 That said, economic factors do still play a part as well. Relatively low property prices in Belgium and Germany, for instance, have encouraged quite substantial emigration from the Dutch border areas to these two neighbouring countries.34

### *2.2.2 A Closer Look at Net Migration*

The immigration and emigration trends described above resulted in positive net migration (immigration exceeding emigration) almost every single year from 1961 to 2002 (see Fig. 2.6). The only exceptions were 1967, a year of recession, and 1982, at the nadir of the economic malaise which had begun in the mid-1970s. In 2003 the tide turned, with emigration outstripping immigration for a number of years, but the situation reversed again from 2008 onwards. The most recent fgures indicate that net migration remains positive, due mainly to record numbers of immigrants arriving in the past decade and a half. These include large groups of labour migrants from the central and eastern European countries which joined the EU on either 1 May 2004 or 1 January 2007. Even more recently, they have been joined by a relatively high number of asylum seekers.

As well as the scale of migration, the respective compositions of the immigrant and emigrant populations can be important in shaping the development of a country's diversity by origin. We illustrate this in Fig. 2.8, distinguishing between those with a Dutch and a migrant background in the net migration fgures for the period 1996–2018. Unfortunately, this data is not available for previous years. We do, however, have fgures for the nationalities of migrants going back to 1977. Since these provide some indication of the composition of the migrant population by origin, they are also included in Fig. 2.8.

An initial look at Fig. 2.8 shows that, from 1996 onwards, net migration by people with a migrant background was very clearly positive. In other words, signifcantly more of them came to the Netherlands than left. Numerically, the total

<sup>33</sup> van Dalen et al. (2008).

<sup>34</sup> van Dalen et al. (2008).

**Fig. 2.8** Net migration in the Netherlands by origin and nationality, 1977–2018 (× 1000)\* \*Emigration fgures are after administrative corrections © WRR (2020) | Source: Statistics Netherlands (CBS)

'surplus' in the period 1996–2018 was 916,000 persons – equivalent to approximately 6% of the total Dutch population in 1996. Interestingly, though, there was also a brief period of negative net migration by this group in 2005 and 2006. That is probably attributable to the strong economic growth in Turkey at the time, coinciding with a perceived negative discourse around migrants in the Netherlands.35

What generates less attention but has still had a considerable effect upon the diversity of the nation's population is that 247,000 people with a Dutch background left the Netherlands in the same period. If we go a little farther back in time and look at net migration broken down by nationality, we fnd that positive net migration of people with a migrant background and negative net migration of those with a Dutch background has been a persistent trend since at least 1977. The only time when there was a 'surplus' of Dutch nationals settling in the country was a brief period in the mid-1980s, probably accounted for by a relatively large number of Antilleans (who are Dutch citizens) migrating to the European Netherlands due to major social and economic setbacks on their home islands – in particular, the closures of the oil refneries on both Curaçao (Shell) and Aruba (Exxon) in 1985.36

<sup>35</sup>Winkelman (2008).

<sup>36</sup>Oostindie and Schoorl (2011). Whilst the economic malaise on Curaçao continued, Aruba managed to recover by successfully marketing itself as a tropical holiday paradise (Oostindie & Schoorl, 2011). In 1990, moreover, the island's oil refnery reopened in a slimmed-down form (Van Belle, 2001).

### *2.2.3 Differences in Natural Population Growth*

Differences in the rate of natural population growth – that is, the number of births versus the number of deaths, also known as the birth surplus (or defcit) – infuence diversity by origin as well. If this rate is higher amongst small origin groups than larger ones, diversity increases. And in the Netherlands it goes without saying that those small groups have migrant backgrounds. Moreover, migrant groups in general tend to have higher birth surpluses than 'indigenous' populations because they are younger overall. Consequently, their mortality rates tend to be lower than average and their birth rates higher.37

In order to compare the birth surpluses of the populations with a Dutch background and with a migrant one, in Fig. 2.9 we show both for the period 1996–2018. To further clarify the situation, we have divided the latter into those with European/ Anglosphere and non-European/Anglosphere backgrounds.

The birth surpluses of the 'Dutch' and 'non-European/Anglosphere' groups were about the same in the second half of the 1990s and into the new millennium. As the latter group was far smaller numerically, however, its relative surplus was much higher. This led to an increase in diversity by origin. To be more precise: in the period 1996–2018 the Dutch population rose by just over 620,000 because more people with a non-European/Anglosphere background were born than died. Over the past 20 years, in fact, natural population growth has accounted for a larger share of the increase in the non-European/Anglosphere population than net immigration.

**Fig. 2.9** Birth surplus by origin, 1996–2018 (× 1000) © WRR (2020) | Source: Statistics Netherlands (CBS)

<sup>37</sup>Nevertheless, the larger 'traditional' migrant groups are still much smaller than the group without a migrant background.

In other words, the second generation has played a greater part in this group's growth than the frst.

Figure 2.9 further reveals that the effect of the differences in birth and mortality rates between population groups is accelerating. Since 2002 the birth surplus in the 'non-European/Anglosphere' group has been higher in absolute terms than in the population with a Dutch background. Indeed, as of 2015 the latter group has been experiencing a birth defcit (also known as a mortality surplus) – the frst time since the Second World War that its natural population growth has been negative.

Meanwhile, the population with a European/Anglosphere migrant background had only a modest birth surplus in the period 1996–2018, and actually experienced mortality surpluses in 1996 and 1997. This was in part because a relatively large proportion of this group lived in border areas with an ageing and falling population, and in part because its largest component subgroup – those with a Dutch East Indies background – was also ageing considerably by this time.

### *2.2.4 Diversity by Origin Continues to Increase*

From the above analysis, it is evident that diversity by origin in the Netherlands is set to increase in the coming decades. This is a result not only of increasing immigration, but also of the differences in birth surpluses between those sections of the population with and without a migrant background. Even if the borders were to be closed completely and immigration were to cease, both the size of the group with a migrant background and its diversity by origin would continue to increase for decades to come. This is due to the high birth rates and low mortality rates in this still relatively young cohort. The Statistics Netherlands population forecast reproduced in Fig. 2.10 illustrates this clearly.38 All three 'migrant background' groups plotted display an upward trend in terms of their share of the total population. The percentage of people with a Dutch background is therefore declining, and according to the Statistics Netherlands will continue to do so until well into the twenty-frst century.39

<sup>38</sup> van Duin et al. (2015).

<sup>39</sup>Both endogenous and exogenous factors underlie these developments. By endogenous factors we primarily mean circumstances in the Netherlands that infuence the natural growth of the various origin groups. Although predicting their future fertility and mortality rates is certainly no easy task, demographers are reasonably well-equipped to do this. The exogenous factors – those infuencing the extent of international migration to the Netherlands – involve far greater uncertainty. They can range from political or economic crises abroad to changes in the admissions policies of our neighbouring countries, which may generate substitution effects in the distribution of migrants across north-western Europe (Jennissen & van Wissen, 2015). The emigration rates of the different origin groups in the Netherlands are infuenced by both endogenous and exogenous factors.

**Fig. 2.10** Share of persons with a migrant background in the Dutch population, 1996–2060 © WRR (2020) | Source: Statistics Netherlands (CBS)

### **2.3 Increasing Transience**

### *2.3.1 Migrants Stay Shorter*

Dutch society is becoming one of 'transient' migration. On average, those coming to live in the Netherlands are staying for shorter and shorter periods of time. This has signifcant consequences for institutions. Schools, companies and voluntary associations increasingly have to deal with the phenomenon of people leaving in the middle of a school year, a production process or, say, a sports season.

Figure 2.11 shows that, of the foreign-born immigrants who arrived in the Netherlands in 1995, almost 20% left again within 2 years. This proportion had increased to more than 35% by 2010, partly due to increased migration for work and studies. Figure 2.11 also includes the 1995 Statistics Netherlands forecast that more than 50% of immigrants to the Netherlands would eventually leave again, a prognosis it revised in 2010 to almost 75%.

This issue of temporary versus permanent migration appears to be particularly pertinent when it comes to recent labour migration from within the EU.40 Amongst Polish workers, for instance, there is considerable variety in this respect. A substantial number return home once they have accumulated suffcient resources, others commute back and forth on a regular basis, depending upon the availability of jobs, and others still move on to other European countries for work. Nevertheless, the number of eastern European labour migrants who settle permanently in the

<sup>40</sup>Scholten and van Ostaijen et al. (2018).

**Fig. 2.11** Departure rates (observed and forecast) of foreign-born immigrants by year of immigration and length of stay in the Netherlands Source: Statistics Netherlands (CBS)

Netherlands seems to be increasing. One important indicator here is the rise in the number of families sending their children to school in the Netherlands.

It also appears that more and more people are just 'passing through' the Netherlands on a multinational migration trajectory. They include Somalis who move on to the United Kingdom and Poles who work in various European countries. In this context, we use the term 'liquid migration'.41 The classic image of migration involves a person leaving their home country for a specifc destination elsewhere and settling there more or less permanently. The idea of 'liquid' mobility, by contrast, refects the emergence of a more 'fuid' or 'transient' migrant population, of whom some stay and put down roots, some eventually move on and others return home at some point.

### *2.3.2 Durations Unknown*

Another aspect of this increasing transience is that it is not clear in advance who is going to stay and who is going to leave again, or when. This is something local authorities, for example, would dearly like to know in order to tailor their

<sup>41</sup>Engbersen and Snel (2013).

integration policies accordingly. The general trends revealed by the fgures presented above point to increased transience, to an unprecedented degree. Consequently, we can never be sure in advance whether a particular newcomer will remain in the country permanently or only temporarily, making it hard to tell who to invest in with regard to social and civic integration and who not to.

In this section we examine whether any patterns related to transience can be identifed. Do certain groups of migrants, such as those who come to work or study, leave earlier than others – family migrants, for instance? Hans Schmeets of Statistics Netherlands has investigated these questions in more detail.42 Using data from his agency's System of Social Statistical Datasets (Stelsel van Sociaal-statistische Bestanden SSB), he tracked two cohorts: migrants who registered in the Netherlands in the period 1999–2005 and those who did so between 2006 and 2010. Whilst 36% of the former stayed in the country for less than 5 years, the fgure for the latter was 47%.

### *2.3.3 Motives for Migration*

The shorter average length of stay may be related to the motives people have for migrating. Students, for example, settle mainly in large cities and in general stay on for only a limited period after completing their studies. In order to analyse this factor in more detail, Fig. 2.12 shows the average length of stay by reason for migration for both cohorts.

© WRR (2020) | Source: Statistics Netherlands (CBS) (Schmeets, 2019)

<sup>42</sup>Schmeets (2019).

In the '99-05' cohort, over 60% of immigrants remained in the Netherlands for more than 5 years. The proportion was higher for asylum and family migrants, just exceeding 70%, and lower for student and labour migrants. When comparing the two cohorts, the general trend towards greater transience becomes quite apparent: the percentage staying less than 2 years increased from 16% to 24%. Only with asylum migrants is this pattern not observed. It is quite plausible that the country of origin plays a role here. After all, opportunities to return vary. Asylum migrants from former Yugoslavia, for example, are generally able go back there if they so wish, now that the region is at peace again. Which is not the case with places like Iraq and Afghanistan. The overall picture is that asylum migrants stay longest in the Netherlands, followed by family migrants.

Labour and student migrants, on the other hand, remain for shorter periods. Fewer asylum migrants arrived in the Netherlands between 2006 and 2010 than in the previous 5 years, but those who did stayed longer. By contrast, more students came and more of them stayed only relatively briefy. The motive behind a person's migration does therefore partly explain the increase in the proportion of 'short-stay' immigrants. We say partly because there are always exceptions in patterns of this kind, such as the many Somali refugees who moved on from the Netherlands to a third country between 2002 and 2007.43

### *2.3.4 Other Characteristics*

For fairly obvious reasons, this variation in duration of stay by reason for migration correlates quite closely with the variation by country of origin. For the majority of immigrants from Suriname, Morocco and Turkey in the two cohorts examined by Schmeets, the motive for coming to the Netherlands was family-related. Those with Anglosphere and Mediterranean backgrounds were more likely to come for work reasons, whilst asylum seekers tend to hail originally from Arab countries, sub-Saharan Africa and central Asia. And study is the main motive bringing migrants from east Asia and Indonesia to the Netherlands. That said, however, and even after correction to take into account motives for migrating, country of origin does also seem to have some autonomous effect with regard to duration of stay. Additional analyses reveal, for example, that about 20% of labour migrants from east Asia, Scandinavia and the Anglosphere stay longer than 10 years. Amongst their counterparts from Arab countries, central Asia and Morocco, the fgure is no more than 10%.44

Variations in duration of stay can thus be explained in part by motives for migration and country of origin, but other more individual characteristics may also play a

<sup>43</sup>Klaver et al. (2010).

<sup>44</sup>Schmeets (2019).

part. For example, being reunited with a partner or having children who become 'rooted' in the Netherlands may reduce the chance of return or onward migration. Researchers at the Ministry of Justice who examined whether the arrival of family members in the Netherlands affects how long labour migrants remain in the country45 found that those who remain without close relatives here are more likely to leave again than those who are reunited with their families.

Although at least some marriage migrants do entertain the idea of 'returning home' at some point, having school-age children in particular tends to cause the postponement of such plans.46 The main reasons for this are the youngsters' wellbeing and the often better educational opportunities in the Netherlands. Moreover, parents often want to spare their children the trauma of migration. After all, they already know what it is like to suffer a language defciency and to have to prove oneself in a different type of society. All things considered, such individual characteristics also play their part in personal choices about whether or not to stay permanently in the Netherlands.

### **2.4 Diversity Between Communities**

### *2.4.1 Considerable Diversity by Origin*

In an earlier publication47 we calculated the diversity by origin of Dutch neighbourhoods, municipalities and regions. For this exercise, we frst divided everyone living in the country into eighteen categories by country or region of origin. Naturally, the largest of these is the Netherlands itself. Almost 80% of the population fell into this group in 2015. The other seventeen included Turkey, Morocco, Anglosphere countries, Arab countries, sub-Saharan Africa, south Asia and so on. We then compiled a diversity index covering every Dutch neighbourhood, municipality and region. This was a version of the so-called Herfndahl-Hirschman Index (HHI), which is used to measure various forms of diversity. It produced a fgure, between 0 and 1, indicating the probability that two randomly selected persons from the area in question belong to different origin groups. The higher the fgure, the greater that chance. A low HHI score thus indicates considerable homogeneity, a high one considerable heterogeneity (what we refer to as 'diversity'). The average for the whole of the Netherlands in 2015 was 0.38.

As Fig. 2.13 shows, however, there are wide variations in HHIs across the country. More than two-thirds of the Dutch population live in a municipality where the

<sup>45</sup>Wijkhuijs and Jennissen (2010).

<sup>46</sup>Sterckx et al. (2014).

<sup>47</sup> Jennissen et al. (2018).

**Fig. 2.13** Herfndahl-Hirschman Indices by municipality, 1 January 2015 © Jennissen et al., (2018) | Source: Statistics Netherlands (CBS)

chance that two residents belong to different groups is about one in three or higher. In the nation's three largest cities, that rises to more than two in three. A high degree of diversity by origin is thus an everyday reality for large sections of Dutch society. However, the nature of that diversity also varies widely from place to place (see Box 2.2).

### **Box 2.2: Types of Municipality**

In addition to the **'average'** Dutch community' in terms of diversity by origin, we have also identifed eight particular distinctive types (see also Fig. 2.14):


### *2.4.2 Limited Variation in Duration of Stay*

Looking at the average durations of migrant stays in the distinctive types of community listed in Box 2.2, we fnd that these do not vary widely between them. The one exception is the expat communities, where stays are generally short; more than half of their incoming migrants leave the Netherlands again within 5 years.

**Fig. 2.14** Examples of Dutch communities with distinctive forms of diversity by origin © Bovens et al. (2020)

### **2.5 Conclusions: More Migration, More Diversity, More Transience**

The main fndings of this chapter are as follows


Dutch Caribbean background. New countries of origin account for an evergrowing proportion of those with a migrant background.


There are wide variations in terms of diversity by origin across the Netherlands, but less so when it comes to duration of stay. Expat municipalities have the highest rate of turnover, with more than half of new arrivals leaving the country within 5 years.

**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 license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license 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 3 Societal Challenges**

This chapter examines the challenges facing Dutch society once migrants have settled here. Their great diversity by origin makes it complicated to live side by side in neighbourhoods or communities, which can lead to feelings of loss, unease and insecurity. In this chapter we outline the most important fndings in this area from our previous empirical research, as well as presenting new ones concerning the local impact of high levels of transient migration. The nature and scale of these challenges, especially those around social cohesion and labour-market participation, differ substantially from place to place. We therefore pay particular attention to that variety.

### **3.1 Ethnic Diversity and Social Cohesion**

### *3.1.1 Diversity Hampers Conviviality*

Large parts of the Netherlands are very diverse in terms of national origin. What does this mean for social cohesion?1 Once a quarter the Netherlands Institute for Social Research (SCP) polls public perceptions of the main problems facing the country. Figure 3.1 shows that the subjects 'immigration and integration' and 'living together' have long been high on this national list, and in recent years have consistently topped it. Clearly, a lot of people have concerns about these issues.

However, this does not necessarily mean that those concerns are related to increased diversity. There are many dimensions to social cohesion and it is infuenced by a multitude of factors. For example, it is wise to distinguish between specifc issues of cohesion and integration related to particular groups and more

<sup>1</sup> 'Social cohesion' is a term used frequently by researchers; it is about what holds a community together and about people's ability to live and work side by side.

R. Jennissen et al., *Migration Diversity and Social Cohesion*, Research for Policy, https://doi.org/10.1007/978-3-031-14224-6\_3

**Fig. 3.1** Perceived national problems as reported by Dutch population aged 18 years and older, 2008–2019 (four largest categories in 2019 third-quarter survey only)\* \*Average of the 12 quarters in the 3 years covered © WRR 2020 | Source: SCP

generic contextual issues such as a neighbourhood's socio-economic composition or its degree of diversity. In recent decades a number of studies have been published about defcient integration by 'traditional' migrant groups and how this might affect various aspects of social cohesion. Those effects are *compositional* – that is, they relate to the specifc demographic make-up of a neighbourhood. In this publication, however, we focus solely upon *diversity effects*: how a high degree of diversity by origin in itself affects conviviality within Dutch society as a whole. We look at this in more detail in Chap. 6. Another topic beyond the scope of this study is interethnic social cohesion, including the question of whether increased diversity leads to greater mutual tensions between groups from different countries of origin.

This section thus centres exclusively on the issue of whether a high degree of diversity by origin has an impact upon social cohesion within the general population. The starting point for our analysis is an article published by the American political scientist Robert Putnam in 2007. In this he argued that residents of 'ethnically diverse' – his terminology – neighbourhoods tend to withdraw from the public space and display less mutual solidarity and commitment to civic affairs.2 Putnam's fndings sparked heated academic debate and a food of follow-up research. A number of review studies have since appeared, consolidating the latest insights in this domain. The general picture they paint is that greater diversity does indeed reduce social trust, and that effect is strongest at the local level. For example, all report that neighbourhoods with a high level of diversity by origin score relatively poorly when it comes to mutual trust between their residents: compared with people living in

<sup>2</sup>Putnam (2007).

more homogeneous communities, they have less contact with their neighbours, evaluate those contacts less positively and have a more negative opinion of their living environment.3

The review studies are less unambiguous about the effects of diversity upon the more general indicators of social cohesion. According to Tom van der Meer and Jochem Tolsma, residents of diverse neighbourhoods do not generally score lower than others when it comes to general trust, undertaking voluntary activities or acting as informal carers. On the other hand, a review by Danish researchers concludes that ethnic diversity does have some effect, albeit limited, upon more general social trust – that is, the extent to which people are generally inclined to trust others. Highly diverse neighbourhoods, moreover, are home to a relatively high proportion of less well-educated and unemployed people. These factors, too, have a major infuence upon neighbourhood relations.4 When they are adjusted for, however, the negative effects of diversity upon trust remain.5

What about the Netherlands? Many Dutch studies are based upon small research populations in a limited number of neighbourhoods, or alternatively are primarily qualitative in nature, which makes it diffcult to generalize the results. Furthermore, they tend to focus only upon one or two of the 'traditional' migrant groups rather than the 'new diversity' we have identifed in this report. All this makes it very diffcult to draw general conclusions about the relationship between diversity and cohesion; for example, because compositional effects may play a major role. Moreover, hardly any Dutch research has explored possible links between diversity and certain indicators of social cohesion, such as neighbourhood safety.

In our empirical research, therefore, we have built upon earlier studies to investigate the relationship between diversity of origin and social cohesion using largescale data analyses. This involved linking the System of Social Statistical Datasets (Stelsel van Sociaal-statistische Bestanden, SSB) maintained by Statistics Netherlands with the national Police Records System (Basisvoorziening Handhaving, BVH) and the 2014 Netherlands' Safety Monitor (Veiligheidsmonitor). The SSB contains data on more than 17 million residents of the Netherlands, whilst the BVH includes – amongst other things – details of everyone formally charged as a suspect in a crime. The Safety Monitor is an annual mass survey covering safety, liveability and victimhood. In 2014, more than 86,000 individuals in 403 municipalities and 8798 neighbourhoods were interviewed. Thanks to these comprehensive databases, our analysis covers *every person* between the ages of 12 and 60 registered as living in the Netherlands (n = 10,746,180) and we have been able to examine the relationship between diversity of origin and social cohesion in *all Dutch municipalities and neighbourhoods*.

We use the Herfndahl-Hirschman Index (HHI) as a measure of diversity and not, as is usual, the percentage of residents from one of the 'traditional' migrant groups

<sup>3</sup> van der Meer and Tolsma (2014), Koopmans et al. (2015), and Dinesen et al. (2020).

<sup>4</sup>See, for example: Letki (2008), Phan (2008), and Gijsberts et al. (2012).

<sup>5</sup>See Dinesen et al. (2020).

or the percentage of 'non-Western immigrants'. Eighteen different groups are included in our HHI analysis, including residents with a Dutch background. In this way we have avoided compositional effects and the like, and have been able to obtain far 'purer' measures of the degree of diversity by origin and its effects. So, for example, communities like Amstelveen and Wassenaar, which are home to a wide variety of expatriates from Western countries, have a high HHI and are thus revealed as being diverse. That would not have shown up had the 'classic' measures of diversity by origin been used.

Social cohesion is an umbrella term encompassing many different factors and indicators. It is thus not possible to say exactly what 'the' effect of diversity upon 'the' social cohesion is. Increasing diversity can affect different aspects of cohesion in different ways. In our analyses we have looked at three clusters of indicators with regard to that cohesion: social capital, feelings of loss and social safety. Social capital is about the ability of people to live and work side by side. Feelings of loss refect people's emotional reactions when they lose touch with the anchors in their life. And social safety frst and foremost requires that attention be paid to changes in social control. These three selected clusters are shown graphically in Fig. 3.2, and below we provide an overview of Dutch and international studies into each of them as well as presenting the results of our own analyses.

### *3.1.2 Social Capital*

Dutch research into ethnic diversity and social capital focuses upon three subthemes in this domain: (1) general trust; (2) neighbourhood relations; and (3) civic participation and voluntary activities. In the studies on the frst and third of these, the fndings are not clear-cut. In two relevant papers6 with somewhat similar results, the authors fnd that diversity at the neighbourhood level does not have any universally negative effect upon general trust or willingness to volunteer. And the same goes for informal care. Where negative effects are reported, that is usually in studies evaluating relationships within a neighbourhood. For example, residents of ethnically diverse environments state that their mutual contacts are less frequent and less positive than those reported between neighbours living in more homogeneous settings.7 Residents of diverse neighbourhoods are also less engaged with their local community.8

In our own research we have analysed the link between the degree of diversity and one particular key indicator with regard to relations at the local level: neighbourhood cohesion. This refers to the strength or weakness of contacts and ties between residents. Our analysis shows that the more diverse a neighbourhood is, the

<sup>6</sup>Tolsma et al. (2009) and Gijsberts et al. (2012).

<sup>7</sup>See: Völker et al. (2007), Lancee and Dronkers (2011), and Gijsberts et al. (2012).

<sup>8</sup>Snel et al. (2018).

**Fig. 3.2** Diversity by origin and aspects of social cohesion © Jennissen et al. (2018)

less cohesive its residents perceive it to be (see Fig. 3.3).9 This also applies to residents with a migrant background; they, too, tend to think that people hardly know each other and that there is little contact between them. They are also less likely to feel that their neighbours interact in a pleasant manner and that they live in a sociable community where people help each other and do things together.10 In conducting our analysis, we took the socio-economic characteristics of the neighbourhood and its residents into account. Besides diversity by origin, socio-economic deprivation also has a negative effect upon neighbourhood cohesion. But the former has by far the greatest impact.

<sup>9</sup>See also Glas et al. (2021).

<sup>10</sup>See also Jennissen and Glas (2020).

**Fig. 3.3** Diversity by origin versus neighbourhood cohesion © Jennissen et al. (2018)

### *3.1.3 Feelings of Loss*

Faced with the presence of numerous and varied migrants, the non-migrant population may experience alienation; that is, the impression that it is losing power and control to the newcomers.11 Researchers describe this using the term 'feelings of loss': the sense that valuable heritage from the past is being lost as a result of undesirable developments in the present. Migrants themselves may also have to contend with such feelings; for example, if they have had to leave their homes because of war or the fear of persecution, or because of homesickness, alienation and homelessness.12 Here, though, we are concerned only with feelings of loss related to the degree of diversity in a neighbourhood. We also look at the experiences of established migrant groups, insofar as their size is large enough to be statistically signifcant.

In the Netherlands, a substantial proportion of people with a Dutch background sense a loss of control due to the presence of groups with a migrant background. They also feel that they are losing their own culture and identity, especially because of the Muslim presence in the country.13 A third dimension to feelings of loss is 'loss of home'. With the arrival of migrants, physical and social landmarks such as churches, shops or schools may disappear or change, so that residents long settled in a neighbourhood no longer feel 'at home' there. One condition for feeling at ease

<sup>11</sup>Smeekes and Mulder (2016).

<sup>12</sup>Boccagni and Baldassar (2015), Ryan (2008), and Svašek (2010).

<sup>13</sup>Smeekes and Mulder (2016) and Dekker and den Ridder (2017).

**Fig. 3.4** Diversity by origin versus 'feeling at home' © Jennissen et al. (2018)

in a living environment is a certain degree of so-called 'public familiarity': residents need to be able to understand and predict other people's behaviour, so that they know what to expect from each other.14 This ability can be diffcult to develop in neighbourhoods with a high degree of diversity by origin.

In our earlier study we probed the relationship between diversity by origin and 'feeling at home' in greater depth. Our guiding question was this: how does the composition of the population in their immediate vicinity affect this feeling in those with a Dutch background? The results (see Fig. 3.4) show that the greater diversity by origin, the less residents feel comfortable with the other people living in their neighbourhood and the less satisfed they are with its demographic composition. The same also applies to some groups of residents with a migrant background. For example, those of Surinamese origin also experience feelings of loss.15

### *3.1.4 Social Safety*

Increased diversity can also have consequences for social safety. The academic literature shows that there may be a connection between the degree of diversity by origin in a neighbourhood and the rates of criminality there.16 The underlying

<sup>14</sup>Blokland and Nast (2014).

<sup>15</sup>See also Jennissen et al. (2018: 97–104).

<sup>16</sup>Nye (1958).

mechanism is that a wide variety of backgrounds makes mutual communication and co-ordination more diffcult. This reduces social control, making people more likely to commit crimes. Various international studies show that a higher degree of ethnic diversity is associated with higher crime rates.17 Empirical Dutch research into this relationship is very scarce, however.18

In our analysis we have examined whether or not the probability that a person is registered as an offender is related to the degree of diversity by origin in the municipality in which he or she lives.19 This exercise reveals that*,* other things being equal*,* that chance is approximately 40% higher in a very diverse community than in a very homogeneous one. Figure 3.5 shows our fndings for an average 20 year-old man with a Dutch background who lives alone in a densely urbanized area. For a woman aged 20, a man of 60 or someone from a rural area, the likelihood of being a registered offender is obviously lower but the curve will be more or less the same. There is a ceiling to this effect, however (see Fig. 3.5): above a certain level of diversity, the probability of registration as an offender levels off. Communities with a high level of diversity, such as Rotterdam and The Hague, thus differ very little in this respect from places with an average level like Gorinchem and Helmond. So it is not the case that a very high degree of diversity by origin goes hand in hand with a high level of criminality.

The degree of diversity in a neighbourhood is certainly not the only factor associated with the likelihood that people living there will commit offences. The criminological literature is replete with underlying variables cited as infuencing the occurrence of criminal behaviour, plus explanations as to why. We have therefore adjusted our fndings for age, gender, origin group, income, educational attainment, type of household and position in it, receipt of social security benefts (as principle source of income) and extent of urbanization of the municipality of residence. All of these variables were also all signifcantly associated with the risk of committing a crime.

The diversity by origin of a living environment may also result in a reduced sense of personal safety. Indeed, our own analysis – presented in Fig. 3.6 – appears to confrm this. The more diverse a neighbourhood, the greater the chance that its residents do not open the door at night, avoid parts of its they consider unsafe, feel unsafe when out at night or home alone and fear becoming a victim of crime. Again, this is an autonomous effect of diversity unrelated to the socio-economic or ethnic composition of the neighbourhood.

<sup>17</sup>See, for example: Osgood and Chambers (2000), Hipp (2007), and Quick and Law (2015).

<sup>18</sup>The only recent study on this topic is by Jacqueline van Beuningen and her colleagues at Statistics Netherlands (van Beuningen et al., 2013). They examined, at the municipality level, whether there is any relationship between ethnic diversity and criminality. Their study shows that the percentage of residents to have been suspected of a crime is relatively high in districts where a large proportion of the population has a non-European/Anglosphere background. This is a compositional effect, however, and not a diversity effect.

<sup>19</sup>See also Jennissen et al. (2018: 104–109).

**Fig. 3.5** Probability\* that an average\*\* male with a Dutch background, aged 20 and living alone in a densely urbanized area is registered as an offender, 2014

\*To be exact, odds ratios

\*\*Not receiving benefts, on an average income and with a secondary-level education © Jennissen et al. (2018)

**Fig. 3.6** Effect of diversity upon feeling unsafe in a neighbourhood\* \*For a 40-year-old working, married female without a migrant background, not previously a victim of crime and with an average level of educational attainment, average disposable income and children in the household © Jennissen et al. (2018)

### **3.2 Transience and Social Cohesion**

The average duration of migrants' stays in the Netherlands is steadily shortening. It is quite plausible that this factor, too, can negatively affect the social cohesion in a neighbourhood or community. When more and more residents are just 'passing through', and the people who do stay constantly have new neighbours, this can lead to isolation amongst those who have settled there permanently. Which can weaken mutual relationships. This applies equally whether the 'transient' population consists of short-term migrants or of other groups.20

Little empirical research has been conducted into the relationship between length of residence and social cohesion in the Netherlands. Tolsma and his colleagues are amongst the few exceptions,21 and they found hardly any evidence that 'moving mobility' (a high degree of transience) at the local level affects social cohesion. One possible reason for this is that their study also included 'transients' – those whose stay is only brief – themselves and it may be that their positive experiences offset more negative feelings with regard to social cohesion reported by long-term residents.

In order to exclude this possibility, in our own additional analyses we have looked only at intracommunity relations and lack of safety as perceived by people who have been living in a particular neighbourhood for more than 10 years. And we do indeed fnd that this group feels that neighbourhood relations deteriorate and that they themselves feel less safe when the average overall length of time people live in the neighbourhood is shorter.22

Problems attributable to increased transience seem likely to be felt even more acutely in social units smaller than neighbourhoods, such as individual streets and housing blocks, schools and local voluntary associations. Schools, for example, have to deal with high rates of pupil intake and outfow at irregular times. Associations come under pressure because they rely heavily upon volunteers and in areas of high population turnover it becomes more diffcult to retain such people for any length of time, never mind recruit them to positions of responsibility. We look at this in more detail in Chap. 6.

### **3.3 Considering Local Diversity**

As we have seen in Chap. 2, the nature of diversity by origin can differ greatly between communities. And the same applies to diversity by length of stay. In some places a large proportion of the migrant population settles for a relatively long time.

<sup>20</sup>Kornhauser (1978).

<sup>21</sup>Tolsma et al. (2009).

<sup>22</sup>This is further elaborated in Jennissen and Glas (2020).

But in a growing number of others this form of diversity is increasing: some immigrants move on quickly, whilst others remain more permanently.

All of this confronts local governments with a combination of general and specifc policy challenges. Whilst some have to deal primarily with less skilled migrant groups and so their main objective is to improve participation in the labour market, others are concerned primarily with ways to facilitate the harmonious conviviality of many different groups. The challenges at the local level are thus closely related to the nature of the diversity by origin and by length of stay in the community. We therefore explore them using the categories presented in Chap. 2 (see Box 2.2). It is important to note that this is a fairly generalized classifcation, whereby some municipalities may fall into several categories. Amstelveen, for example, is a metropolitan suburb of Amsterdam but also home to a large number of expats.

### *3.3.1 Majority-Minority Cities*

In Amsterdam, Rotterdam and The Hague, diversity by origin is the norm and no single group forms a majority. More than half of residents now have a migrant background. This transition represents an important symbolic watershed because it raises questions about the identity of cities and makes it even clearer how diverse many people's day-to-day surroundings are becoming. These cities are faced with the task of facilitating the conviviality of very many different origin groups and of mitigating potential incompatibilities between them.

The Hague and Rotterdam, in particular, are having to deal with the short-term presence of labour migrants from within the EU. This form of temporary migration constitutes a new challenge for public institutions, including schools (see Box 3.1).

### **Box 3.1: School as a Place of Transit**

Primary school De Kameleon in south Rotterdam has approximately 450 pupils, the majority with a central or eastern European migrant background. This means that it has to deal with a very high rate of irregular turnover. Each year a total of about 250 pupils join or leave the school. In some cases the composition of a class changes by as much as 70% over the course of a single year. There is also a lot of so-called 'commuter migration' in this part of Rotterdam – families who shuttle back and forth to their home country every few months. The temporary nature of these pupils presents the school with a major challenge. The repeated departure of classmates during the school year is emotionally stressful for the children, and the new arrivals require a lot of attention from the teachers.

### *3.3.2 Metropolitan Suburbs*

Suburbs often serve as home for people who work in the nearby city. Schiedam is a suburb of Rotterdam, for example, and Zoetermeer an equivalent for The Hague. Conversely, major employers may be located in suburbs whilst much of their workforce lives in the city. Many employees of ASML in Veldhoven, for example, live in Eindhoven.

Metropolitan suburbs or peripheral communities also serve as overfow areas for the large cities. As a result, 'big-city problems' frequently spill over into them. Their population turnover is high, too, which creates problems for local institutions such as schools. For their policy and facilities, suburbs are strongly dependent upon co-operation with their metropolitan neighbours.

### *3.3.3 Larger Provincial Cities*

Larger provincial cities like Utrecht, Haarlem, Dordrecht, Nijmegen and Enschede also have high levels of diversity. But unlike the majority-minority cities and their suburbs, people with a Dutch background are still very much in the majority. Nevertheless, they generally face policy challenges similar to those in the majorityminority cities.

### *3.3.4 Medium-Sized Towns with One Specifc Migrant Group*

Certain medium-sized Dutch towns are characterized by the presence of a single large non-European/Anglosphere minority group. This is often a result of historical recruitment of guest workers from a specifc country of origin, as with the Moroccan community in Gouda and the Turks living in Leerdam and Almelo. Den Helder and Delfzijl, meanwhile, are examples of places which have attracted relatively large groups of Antilleans. Since later trends in migration, including the infux of highly skilled professionals and intra-EU labour migrants, have largely bypassed these towns, the original group has often remained the only one of any size.

Because of their specifc demographic make-up, the risk of ethnic polarization is relatively high in these communities. In Gouda, for example, tensions have arisen in certain neighbourhoods due to their large Moroccan population. In the past, national integration policy specifcally targeted 'traditional' migrant groups of this kind, but those efforts have been scaled down in recent years because of their sometimes counterproductive effects: the emphasis upon specifc origin groups tended to stigmatize both them and the communities they lived in.

In many cases, the established minority groups in these medium-sized towns now consist largely of second-generation migrants. This has signifcant repercussions for local policy. Since they are not newcomers, it is important that the members of this generation be covered as much as possible by general policy and by universal services and amenities. This reduces 'them-and-us' thinking and promotes opportunities for contact. Nevertheless, knowledge about the specifc origin group remains important. And targeted local policy can still be relevant.

### *3.3.5 Expat Communities*

Expat communities have residents from a wide variety of countries of origin, but relatively few in the 'traditional' migrant categories. Typical examples are Amstelveen, Wassenaar, Voorschoten and Wageningen. Their diversity by origin is usually quite high, without that causing particular problems. In part this is due to the relatively self-suffcient nature of expatriates themselves: in general they are wellto-do and highly educated, and thus quite capable of taking care of themselves and contributing to the local economy in various ways.

Nevertheless, they do still throw up a number of specifc policy challenges. Some expats send their children to international schools, for instance, but many opt for regular Dutch-speaking ones. Such phenomena call for a more differentiated approach, one which both facilitates temporary residence and, where appropriate, enables more lasting integration into Dutch society. Research amongst expatriates themselves, moreover, shows that they would like more contact with 'locals' in order to learn the language and integrate better.23

### *3.3.6 Horticultural Districts*

The large intensive horticulture sector in districts such as Westland, Zundert, Zeewolde and Horst aan de Maas has attracted a lot of people with a Polish or Bulgarian background. The central and eastern European origin group is often the only one of any size in these communities, which used to be fairly homogeneous. They are also characterized by a relatively high degree of diversity by length of stay in their migrant populations.

One important challenge facing their local authorities, then, is how to cater for these labour migrants when no-one can be quite sure how long they will stay. In the past, policymakers usually assumed that their presence was purely temporary. This, for example, was the thinking behind the development of so-called 'Polish hotels' – basic hostel-like living accommodation in horticultural areas for workers on shortterm contracts. In reality, however, the situation has proven more complex: it is impossible to predict how long individual migrants will decide to remain, which

<sup>23</sup>See, for example, van Bochove et al. (2010).

often creates tensions between the policy requirements associated with transience and those linked to permanent settlement. To what extent should public resources be deployed to integrate newcomers who might or might not leave again at any time?

Even temporary migrants, moreover, make demands of certain local facilities, such as schools for their children. In horticultural districts, then, policy differentiation is crucial. In other words, one approach and set of facilities for migrants whose stays are truly temporary and another for those in need of a greater degree of integration into local society.

### *3.3.7 Border Communities*

The presence of a relatively large number of residents with a German background makes certain Dutch border communities highly diverse.24 For example, many Germans have settled in Vaals and Kerkrade in the province of Limburg. Many are international commuters: they come to live just inside the Netherlands, but continue to work in their homeland. Despite having opted for cheaper or more spacious living on the Dutch side of the border, their professional and social lives are still played out on the other side. As a result, they play little part in local society. This lack of community spirit is an important policy challenge for local authorities in the border regions. The key question here, then, is how to ensure that their various population groups do not turn their backs on each other.

### *3.3.8 Homogeneous Communities*

Despite the fact that diversity by origin has increased considerably right across the Netherlands in recent decades, in 2015 some 3.2 million people – 19% of the total population – still lived in a municipality where more than 90% of residents have a Dutch background. For these homogeneous communities, addressing diversity is obviously not a policy priority. Nevertheless, this is a challenge they may well face sooner or later – for example, if they are allocated a quota of asylum migrants in line with the national dispersal policy.

<sup>24</sup>We confne ourselves here to communities with a large German population. There are also relatively large numbers of people with a Belgian background living on the Dutch side of the border with Belgium, but the ethnic, linguistic and cultural distinctions along this frontier are not so clearcut. The dialects of Dutch spoken on each side are not that different, for example. Even in a municipality like Baarle-Nassau in Noord-Brabant, which has a very high percentage of residents with a Belgian background, there is hardly any ethnolinguistic heterogeneity. In fact, the village of Baarle is divided between Baarle-Nassau and the Belgian municipality of Baarle-Hertog and the national frontiers running through it are not cultural borders at all. Baronies, a subdialect of Brabantian, is the vernacular throughout the village (Swanenberg, 2010).

### **3.4 Conclusions**

The empirical analyses in Chap. 2 and this chapter reveal that the Netherlands is set to face major migration-related social challenges in the coming years. These are going to come on top of the existing ones associated with the integration of certain 'traditional' migrant groups. The new challenges are summarized below.


In formulating policy to address these social challenges, it is important to take differences between different parts of the country into account. This means that national policy needs to leave plenty of scope for variation at the local level.

**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 license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license 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 Successive National Policy Models**

How can the government manage the arrival of large numbers of very different immigrants? Before addressing this question in the next chapter, here we frst analyse how Dutch integration policy has developed since 1960. A number of national policy models have been introduced during this period, but for different reasons none is able to deal adequately with contemporary patterns of migration and integration.

We then look at the development of Dutch immigration policy. As with integration policy, this is characterized by a high degree of volatility. Moreover, there is little coherence between migration and integration policies. We end this chapter with a brief discussion of the inadequacy of past national policy models and then formulate the most important policy issues raised in this report.

### **4.1 Changing National Models for Integration Policy**

In Dutch post-war integration policy, four successive national models can be discerned. On one hand each was a response to the migration patterns of the time, whilst on the other they were rooted in specifc contemporary views of migration, integration and citizenship (see Table 4.1).1 The criteria we have used to distinguish these models are as follows.


<sup>1</sup>Snel and Scholten (2005).

R. Jennissen et al., *Migration Diversity and Social Cohesion*, Research for Policy, https://doi.org/10.1007/978-3-031-14224-6\_4


**Table 4.1** Dutch policy models for integration (1960-present)

To be clear, these are ideal-type distinctions in which we highlight the central features of the models concerned.2 In reality, their delineation is not so sharp and they overlap to a certain extent.

Each of the four models we have identifed was developed in the context of a specifc pattern of migration.3 For this reason we frst take a brief look at those patterns. The frst three – laissez-faire, multiculturalism and disadvantage-driven – arose in the period of postcolonial and guest-worker migration (1960–1995). This

<sup>2</sup> Ideal types are, in the famous words of Max Weber (2002: 55), "constructed concepts endowed with a degree of consistency seldom found in actual history".

<sup>3</sup>See also Snel and Scholten (2005).

was when most immigrants to the Netherlands came from a limited number of countries: the former Dutch East Indies, Suriname, the Dutch Caribbean, Turkey and Morocco.

The fourth model, focusing upon 'citizenization' within Dutch society, took shape around the turn of the millennium. During this period, groups of refugees from Iran, Iraq, Afghanistan, Somalia, the former Soviet Union, the former Yugoslavia and elsewhere were applying for asylum in the Netherlands in quite considerable numbers. Mobility within the European Union (EU) was also increasing, as were the numbers of highly skilled and student migrants coming from both within and outside the EU. These new trends resulted in a substantial growth of diversity by origin.

We should point out here that the names we have coined for the four models do not coincide with the policy idiom used by the Dutch government at the time. It never spoke explicitly of a 'laissez-faire' policy, for example, or of 'multiculturalism'. Our models are in fact analytical constructs highlighting the fact that concepts of 'integration' and the associated policy have been subject to shifting interpretations over the past few decades.4

These changes, over a relatively short period of 60 years, reveal a high degree of *policy volatility*. More than a decade and a half ago, in 2004, the Parliamentary Committee of Inquiry on Integration Policy noted that consistency in this domain had been very limited over the years. With the one exception of the lack of evidencebased policy. Instead, it has always been determined to a large extent by ideology.5 It is also striking that, since the 1990s, the political responsibility for integration policy has been assigned to no fewer than four different ministries – Internal Affairs, Justice, Housing and Social Affairs and Employment at different times.

This volatility can be explained in part by the learning process Dutch society and government have been through in shaping a migration and integration policy. For a long time the Netherlands was "a reluctant country of immigration".6 It was not until 1998 that its political establishment offcially recognized that it had, de facto, become a country of immigration.7 Another factor is unforeseen processes: many migrants did not return home, as they had been expected to, and unemployment amongst certain groups increased rapidly during the 1980s. A third explanation is changes in migration patterns, such as increased diversity by origin. These processes necessitate continuous policy adjustment, so each of the models we have identifed was or is to some extent a reaction to the perceived shortcomings of an earlier one.8

<sup>4</sup>Entzinger (2006), for example, has this to say about the multicultural model: "In the 1980s, the term multiculturalism was not as common as it is today. The Dutch government never really used it. In hindsight, however, Dutch Minorities' Policy of the 1980s certainly can be labelled as multiculturalist. In a number of spheres it provided institutional arrangements that ran parallel to existing mainstream arrangements. The special characteristics of the migrant cultures served to justify such forms of separateness. The parallel institutions were generously supported with public funds." 5Klaver et al. (2016).

<sup>6</sup>Entzinger (1984) and Cornelius et al. (1994).

<sup>7</sup>See WRR (2001: 35).

<sup>8</sup>Duyvendak and Rijkschroeff (2004) and Snel and Scholten (2005).

We briefy discuss the characteristics of each model below, also explaining why, in itself, none of the four provides an adequate response to the challenges arising out of increased diversity by origin and the shorter average duration of migrants' stays in the Netherlands.

### *4.1.1 Laissez-Faire*

The frst model is a 'laissez-faire' approach in which, as far as possible, responsibility for the integration of migrants is left to the market and civil society. Enduring participation in the labour force and the help of migrant networks eventually enabled the post-war 'guest workers' to secure a place in Dutch society. Until well into the 1970s, however, offcial policy towards them had the characteristics of a laissezfaire model. There were no concerted government efforts aimed at their integration, mainly because the idea that their stays were only temporary still predominated. As 'guests' they were supposed to remain for a limited period only and then return home once their work was done.9 This conviction that the Netherlands was not a country of immigration was enshrined in the basic principle underlying the 1976 Aliens Act (Vreemdelingenwet).

At the local level, there were some modest policy initiatives during this period intended to tackle the socio-economic disadvantages experienced by migrants, particularly with regard to housing, but these were still based upon the assumption that they were not going to become permanent residents. The pressing issues in this respect related mainly to overcrowded and unsafe boarding houses in the major Dutch cities (see Box 4.1).10

### **Box 4.1: Unsafe Accommodation11**

The availability and quality of housing for foreign workers became an urgent issue in the 1960s. The newcomers found themselves living in overcrowded hostels, guest houses or private boarding houses. Neighbours and other local residents complained about the nuisance these caused. Much of the accommodation was unlicensed and did not meet basic safety requirements. When an immigrant boarding house in Amsterdam burnt down in December 1970, nine residents were killed. Campaign groups were subsequently established in several cities to denounce the poor conditions immigrants were forced to live in.

(continued)

<sup>9</sup>Heijke (1979), Muus (2004), and van Amersfoort (1982).

<sup>10</sup>Tinnemans (1994).

<sup>11</sup>See Bovenkerk et al. (1985), Tinnemans (1994), Lucassen and Penninx (1994), and Jennissen (2013).

### **Box 4.1** (continued)

Successive Dutch governments were unwilling to build additional housing especially for labour migrants, however, at a time of widespread shortages. Many Dutch citizens had been on the waiting list for a home for years. The authorities held employers responsible for accommodating the foreign workers they recruited. But three-quarters of immigrants were arriving outside of formal recruitment procedures and employers felt no obligation to house either them or those whose contracts had expired. Once foreign workers started bringing in their families as well, the crisis in the housing market only worsened.

In the summer of 1972, riots broke out in one Rotterdam neighbourhood. Tensions between local residents and foreign workers had escalated in part because the city council intervened too late. The construction of large social housing developments and the exodus of 'indigenous' working-class families to the suburbs eventually eased the problem.

Following the enlargement of the EU in 2004 and again in 2007, history repeated itself. Large numbers of central and eastern Europeans came to the Netherlands to work, but adequate accommodation for them was in short supply. Just as in the 1960s and '70s, these labour migrants often found themselves dependent upon the informal private housing sector. Their situation was a core theme of the 2011 report of the Ad-Hoc Parliamentary Committee on Labour Migration, but that still failed to generate adequate solutions.12 The COVID-19 pandemic has again thrown this defciency into sharp relief.13

One major criticism of the laissez-faire policy was that it did not take into account the permanent settlement of large groups of migrants from Morocco and Turkey, or the integration-related challenges that posed in such areas as housing, education and labour-market participation. During the 1970s and '80s, it was fnally and reluctantly accepted that many 'guest workers' were never going to leave and so there was a growing need for government direction and intervention.14 In response, a 'minorities policy' was introduced to promote the integration of migrants into Dutch society.15

Laissez-faire policies remain relevant when dealing with the now increasing diversity of migration because some groups, in particular highly skilled professionals, enter and participate in the Dutch labour market without any great problem.

<sup>12</sup>Tijdelijke commissie lessen uit recente arbeidsmigratie (2011).

<sup>13</sup>Aanjaagteam bescherming arbeidsmigranten (2020).

<sup>14</sup>Entzinger (1975).

<sup>15</sup>See the 'Minorities report' (*Minderhedennota*) issued by the Ministry of the Interior in 1983 (Ministerie van Binnenlandse Zaken, 1983; see also Vermeulen & Penninx, 1994: 9). This set out two objectives: frstly, to combat and eliminate the disadvantage and deprivation affecting specifc migrant groups; and secondly, the pursuit of a tolerant multicultural society in which people with different ethnic origins and cultures are accepted and valued as equals.

Nevertheless, laissez-faire as a general principle is not appropriate. Many asylum and family migrants have great diffculty fnding jobs and need targeted government support.16 Furthermore, this type of policy pays little attention to social cohesion in the neighbourhoods and communities where many migrants settle. Our research shows that in localities where diversity by origin is high, residents have a less positive view of neighbourhood relations and feel less at home and safe.17 In other words: self-evident conviviality does not come about automatically where there is a great variety of origin groups. This issue requires that local governments play an active role.

### *4.1.2 Multiculturalism*

A second policy model is multiculturalism. This is a somewhat ambiguous term because it refers both to actual situations in which multiple cultures are found side by side and to a normative approach which accepts or even promotes the coexistence of different cultural traditions.18 Multiculturalism as a policy model pursues integration whilst at the same time allowing social groups to retain their own culture and identity. Its Dutch version assumes that the principal path towards the effective integration of immigrants is through their own communities and self-organization, with a certain degree of institutionalization when it comes to cultural pluralism seen as a precondition for emancipation and integration. In the Netherlands, the ethnic minorities policy of the 1980s and early 1990s to a great extent refected this multicultural model. It was based upon a group-focused approach aimed at preserving migrants' own cultures and identities.

One aspect of that policy was to subsidize and consult migrant organizations.19 At least in part, our multicultural model was in line with the long-standing Dutch tradition of 'pillarization' ('verzuiling'): the division of society into parallel communities according to religious and/or social convictions, with each operating its own civil society organizations – places of worship, schools, political parties, public broadcasters and even sports and social clubs. Following the same principle, the government facilitated education in migrants' own languages and cultures and subsidized their own cultural, political and social organizations.20 It also consulted representatives of the largest groups as part of its decision-making processes.

<sup>16</sup>SER (2019), CBS (2018), Engbersen et al. (2015) and Bakker (2016).

<sup>17</sup> Jennissen et al. (2018).

<sup>18</sup>Hirsch Ballin (2011: 11).

<sup>19</sup>See Entzinger (2014).

<sup>20</sup>Entzinger (2002, 2006).

In the 1980s and early 1990s this policy came under fre for failing to provide an adequate response to the evident socio-economic disadvantage suffered by many migrant groups. In fact, the critics claimed, the continuing pursuit of distinct sociocultural identities was maintaining or even reinforcing socio-economic disadvantage.21 Multiculturalism was also said to ignore certain problems and structural inequalities by adopting a culturally relativist attitude – for example, with regard to the position of women.22 It might also create an undesirable form of 'pillarization', in which separate schools, churches, newspapers, broadcasters and other organizations for each ethnic group enabled their own self-exclusion and so forestalled proper integration into the host society.23

Due to the persistent educational disadvantage and poor labour-market position of large migrant groups, the multicultural model fell out favour over the course of the 1990s. In 2003, the government decided to abolish programmes to teach migrants in their own languages (see Box 4.2). Nonetheless, one of the model's most important instruments was retained for much longer: the Minorities Policy Consultation Act (Wet Overleg Minderhedenbeleid, WOM), which came into effect in 1997, would function for another 16 years. This law provided for regular policy consultations between national migrant organizations and central government. It was not until June 2013 that parliament voted for its repeal.

How does multiculturalism relate to increased diversity? One relevant aspect is that it addresses issues such as intercultural competences and cultural sensitivity in the workplace. Such qualities are important for people and institutions having to deal with a high degree of diversity by origin.

Nevertheless, multiculturalism is insuffciently equipped to deal with increased diversity. First of all, for practical reasons alone this model is diffcult to apply in a society with a very large number of migrant groups, some with only a short average stay. A second critique is that multiculturalism centres on specifc groups with fxed traditions and customs, but essentialism of this kind fails to recognize that culture is always subject to change. Seeking to acknowledge cultural identity can also have the effect of magnifying differences between groups, making them more pronounced than they really are, whereas increased diversity in fact tends to blur intergroup boundaries and lead to mixing and hybridization.24 Mixed relationships are an indicator of this. Figures from Statistics Netherlands (Centraal Bureau voor de Statistiek, CBS), for example, show that 30% of all married or cohabiting Dutchborn people with a Surinamese background have a partner with a Dutch background. For those with a Dutch Caribbean background, the fgure is 49%. But at 6% and 8% respectively, the proportions for people with a Turkish or Moroccan background are

<sup>21</sup>Duyvendak and Rijkschroeff (2004).

<sup>22</sup>Bloemraad et al. (2008) and Kymlicka (2012).

<sup>23</sup>Koopmans (2002).

<sup>24</sup>See Burke (2009).

considerably lower.25 Finally, the multicultural model pays little heed to a shared framework of values and norms which enables people from many different countries of origin and staying in the country for different lengths of time to live together harmoniously.

### **Box 4.2: The Mother-Tongue Dilemma**

One particular characteristic of multiculturalism in the Netherlands was the facilitation of teaching in migrants' own languages. But after more than twenty years of experimentation with mother-tongue education in various forms, in 2003 the government decided to abolish these programmes. The principal reason cited at the time was lack of quality and results. The practical effect was that the government shifted responsibility for familiarizing new generations with the languages spoken by their forebears to the migrant communities themselves.26 As a result, mosques, community centres and other providers now offer extracurricular language and culture lessons to children with a migrant background. This approach has the disadvantage that there is little or no control over the quality and content of the lessons.

At many schools in the Netherlands it is common practice for pupils to be instructed to speak only Dutch at school and at home, even when that is not their frst language. One frequently heard argument for this approach is that multilingualism hinders learning of Dutch, and by extension the other subjects taught in Dutch. This assumption runs counter to scientifc insights into the interaction between learning a frst and a second (or even third) language in children, however. In fact, a good command of their mother tongue facilitates the learning of a new language because general linguistic skills and concepts already acquired can be 'transferred', as it were, to the second tongue.27 Various positive effects for cognitive, social and personal development have also been recorded in bilingual learners profcient in both languages.28

Given the current multilingual diversity in the classroom, on the other hand, it is very diffcult in practical terms to offer all children mother-tongue education. This does not mean, though, that monolingualism should be a strict norm.29 For example, students with the same mother tongue can help each other, parents can assist them and teachers can use translation apps and multilingual social robots.30

<sup>25</sup>A 'mixed' household here refers only to a person with a migrant background married to or cohabiting with someone from a Dutch background. Unfortunately, from the available Statistics Netherlands data it has not been possible to identify mixed households where both partners have a migrant background.

<sup>26</sup>Ministerie van Onderwijs, Cultuur en Wetenschap (2008).

<sup>27</sup>See, for example, Chin and Wigglesworth (2007).

<sup>28</sup>PO-Raad (2017).

<sup>29</sup>PO-Raad (2017), van Avermaet (2015), and Canagarajah (2011).

<sup>30</sup>See, for example, L2tor (2017).

### *4.1.3 Disadvantage-Driven Policy*

The third policy model focuses upon enhancing socio-economic participation by eliminating disadvantages in the felds of employment, housing and education. Just as the multicultural model was a reaction to laissez-faire thinking, so this policy was developed in response to multiculturalism's apparent shortcomings. Unlike its forerunner, disadvantage-driven policy targeted the migrant as an individual, not their community. Nevertheless, members of certain groups were still very much at its heart, in particular those with a Dutch Caribbean, Moroccan, Turkish or Surinamese background.

Achieving proportionate or full participation in the labour market is a core goal of policies centring on disadvantage. Their underlying assumption is that socioeconomic participation is a crucial precondition for integration. Only when immigrants can stand on their own two feet will they also fnd their way in society in other respects. Within this model there is a strong role for active government policies, especially in the felds of education, housing and employment. One infuential document in its design was the 1989 WRR report *Immigrant Policy* (*Allochtonenbeleid*), which warned of a growing social gap between people with and without a migrant background. It pointed out that unemployment amongst certain groups of migrants had risen to more than 40% of their active population and that those affected faced the risk of structural exclusion from the labour market which could be carried over into future generations. The WRR found that the government regarded migrant groups too much as a 'problem category', rather than offering them opportunities to achieve greater self-determination. Partly as a result of this, many members of these groups were over-reliant upon public provision in the form of benefts, other welfare assistance and social housing.

In line with these fndings, during the 1990s the Dutch government linked integration directly with an activation policy towards the labour market and the welfare state.31 Intervention in sociocultural domains was scaled back as much as possible, since culture and identity were now deemed matters of private concern. This decade also saw a return to universal policy, which came to take precedence over measures targeting specifc groups. The evolution of policy aimed at compensating schools for their pupils' educational disadvantage, for instance, shows how 'ethnicity' began to play less and less of a role. Eventually, the level of the parents' schooling was made the primary benchmark (see Box 4.3). However, this 'universal' criterion turned out to take insuffcient account of specifc educational challenges related to the arrival of intra-EU labour migrants and highly skilled professionals. For this reason there is now a new arrangement, in which the mother's country of origin is once again taken into consideration (see Box 4.3).32

<sup>31</sup>See WRR (1989).

<sup>32</sup>The WRR (1989) writes, "The frst and most elementary approach is that of *general policy*. To a signifcant extent, the problems facing immigrants are accentuations of those facing society in general. This means that policies specifcally tailored to the immigrant community, in which the overall problems are not addressed, will necessarily remain confned to the treatment of symptoms. General measures, extending to all residents, to promote employment, education opportunities and suitable housing form the basis for conducting *specifcally targeted* policies where these are required."

### **Box 4.3: Compensating for Educational Disadvantage and the Role of Origin**

There is a long-running debate about the role of ethnicity in determining educational disadvantage. Dutch primary schools receive fnancial compensation for pupils who have been unable to develop to their full ability due to circumstances at home. From 1985 this was calculated using a weighted combination of ethnicity and the schooling and occupations of the children's parents or guardians. The occupation factor was subsequently dropped, however, and later ethnicity as well. With effect from 2006, schools thus received no additional budget for migrant pupils whose parents had attended further or higher education.33

It has now become apparent that this arrangement can cause fnancial problems for some schools. Those with a large number of pupils with a Polish or Romanian migration background, for example, or attended by the children of highly skilled migrants. The current algorithm releases additional funding only for pupils with less well-educated parents, whereas both these groups tend to have a relatively good schooling. But that certainly does not mean that their children will be profcient in the Dutch language when they start attending a school here, making it inevitable that it will have to devote extra effort to teaching them Dutch so that they can participate fully in regular education. A lack of funding for this purpose is experienced as a problem by many primary head teachers.34

At the behest of the Ministry of Education, Culture and Science,35 Statistics Netherlands has therefore developed a proposal for a new weighting system. This uses fve key criteria to determine whether a child has an educational disadvantage:


Under this plan, origin – specifcally, the mother's country of origin – thus again becomes one of the factors determining educational disadvantage, together with how long the mother has lived in the country.36

<sup>33</sup>Fettelaar and Smeets (2013).

<sup>34</sup>See, for example, Vogels et al. (2014).

<sup>35</sup>CBS (2016) 'Herziening gewichtenregeling primair onderwijs' [Revised weighting system for primary education].

<sup>36</sup>Netherlands House of Representatives, 2017–2018 session, 27 020, no. 88.

The disadvantage-driven model has been criticized as one-sided by overemphasizing adversity and ignoring the overall improvement in the social position of migrant groups, particularly the second generation. It has also come under fre for reducing integration-related issues to personal problems facing individuals and thus ignoring structural changes in the economy which have particularly affected the position of low-skilled migrants.37 As a result, it is claimed, the resulting policy could be counterproductive since it promotes an image of migrants as being unable or even unwilling to participate in the labour market. In addition, it glosses over sociocultural themes such as traditional gender roles, which also handicap labourmarket participation. In fact, this model adopts a laissez-faire approach when it comes to sociocultural matters, fuelled by a strong belief that participation in the labour market leads more or less automatically to desirable outcomes in that domain.38

As the basis for a generic policy model, disadvantage is too limited a theme to cope with the increased diversity we are witnessing today. Many current immigrants do not suffer any socio-economic disadvantage and their labour-market participation is not a structural problem. Moreover, the social position of the second generation is improving. The children of asylum migrants and former guest workers are showing clear progress in their educational attainment and, in part at least, in their position in the labour market.39 Nevertheless, a disadvantage-driven policy is still benefcial for certain specifc groups more alienated from that market, such as many family and asylum migrants.

Finally, this model pays little attention to problems of social cohesion resulting from the increased diversity of the migrant population and its rate of turnover. It assumes that improving the social position of migrant groups will automatically lead to greater cohesion. But this is not the case. Our analyses show that, even in neighbourhoods populated by large numbers of migrants with a strong labourmarket position, conviviality is a complex issue.40

### *4.1.4 Citizenization*

Our fourth policy model is citizenization. This again adds a sociocultural dimension to integration policy and, albeit implicitly, also recognizes the increased diversity of international migration. In the Netherlands, this model is intertwined with the civic integration policy launched in the late 1990s and refects general changes in attitudes towards participation and citizenship. It places a stronger emphasis upon

<sup>37</sup> In particular, the transition from an industrial society with a high proportion of factory work and other low-skilled labour to a postindustrial society has had a major impact upon the labour-market position of many low-skilled immigrants.

<sup>38</sup>Dagevos (2001).

<sup>39</sup>Maliepaard et al. (2017) and de Mooij et al. (2018).

<sup>40</sup> Jennissen et al. (2018).

individual responsibility and upon the duty of people forming part of a community to contribute actively to it.41 This trend can be traced in a series of new of laws: the 1998 Civic Integration for Newcomers Act (Wet inburgering nieuwkomers), followed by the 2007 Civic Integration Act (Wet inburgering) and its 2013 and 2021 namesakes.

The act of 2013 was based upon one of the keys points in the coalition agreement establishing prime minister Mark Rutte's frst government in 2010 and its policy paper the following year on 'integration*,* cohesion and citizenship' (*Integratie, binding en burgerschap*): anyone who settles permanently in the Netherlands has a personal responsibility to integrate into Dutch society.42 This requirement is in turn derived from what the government views as a core Dutch value: it is up to every citizen, and likewise everyone living in the Netherlands, to contribute to society to the best of his or her ability and to be as self-reliant as possible in that respect. For many immigrants, civic integration is a precondition for long-term residency and the gateway to acquiring Dutch citizenship. The principal pillars of this process are learning the Dutch language and acquiring a thorough knowledge of Dutch society. Since 2017, newcomers obliged to participate in a civic integration programme have also been required to sign a 'participation declaration' ('participatieverklaring'), a legal document in which they acknowledge and accept the core values of Dutch society.43 Not all migrants coming to the Netherlands are subject to that obligation, however, and others (known colloquially as 'oldcomers') have been here since it before it was introduced.

The 2011 policy paper on integration*,* cohesion and citizenship was also instrumental in the repeal of WOM (see above). Henceforth, universal policy would be the rule. Particularly since Rutte has been premier, national policy towards the integration of migrants has shifted away from regarding this as a separate issue and incorporated it into more general domains such as education, social affairs, housing and home affairs. This trend is referred to as 'mainstreaming'. One exception is the specifc policy concerning civic integration, but this has seen its funding diminish due to a greater emphasis upon newcomers taking personal responsibility for their incorporation into Dutch society.44

Criticism of this model is directed at both civic integration policy and mainstreaming. The former, for instance, is said to 'moralize' or 'culturalize' integration because it places such a strong emphasis upon cultural knowledge of Dutch society.45 Sociocultural adaptation or assimilation, it is argued, is overexaggerated as a condition for successful participation.46 Especially since 2013, serious concerns have also been expressed about the policy's focus upon individual responsibility, the

<sup>41</sup>Schinkel and van Houdt (2010) use the term 'neoliberal communitarianism' for this.

<sup>42</sup>Ministerie van BZK (2011).

<sup>43</sup>Such concepts as freedom, equality and solidarity are central in this. The declaration is intended to welcome newcomers and to remind them of their rights and obligations and of 'Dutch values'.

<sup>44</sup>Scholten and van Breugel (2018) and Scholten (2018).

<sup>45</sup>Schinkel (2008, 2010) and Duyvendak et al. (2016).

<sup>46</sup>Entzinger (2006).

abolition of state funding for integration-related activities, the disappearance of a supporting knowledge infrastructure due to budget cuts, the marketization of the system and the poor quality of civic integration programmes. These factors have resulted in a decrease in the number of people required to undergo civic integration who actually complete their trajectory successfully within the prescribed period.47 The latest Civic Integration Act, which entered force in 2021, seeks to address a number of these issues. For example, it gives local authorities greater responsibility and more resources to shape the civic integration activities in their areas.

The main critique levelled against mainstreaming is that it risks diluting integration as a policy priority.48 If not accompanied by a coherent policy vision with harmonization across the various domains now tasked with a role in promoting integration, the issue is in danger of disappearing 'off the radar'.49

On the other hand, a universal policy of citizenization could well suit the increased diversity of the migrant population. When this varies very widely in terms of origin and length of stay, it is extremely diffcult to implement targeted policy for specifc groups. It is also appropriate that migrants not automatically be labelled as 'disadvantaged' but are instead addressed by the government as 'citizens' (whether or not they are actually Dutch nationals) with their own responsibility to participate fully in society. Nevertheless, specifc policy may still be required in the case of specifc problems affecting specifc groups, such as poor labour-market participation or discrimination.50

Another issue is how civic integration policy relates to increased diversity. In classic 'immigrant nations' like the United States and Canada, diversity is an important feature of the national self-image. By contrast, Dutch policy ignores not only this factor but also problems of declining cohesion at the local level under the infuence of increased diversity. Social cohesion is linked to sociocultural adaptation by migrants, but little attention is paid to facilitating their conviviality with non-migrants.

Finally, Dutch civic integration policy in its current form is ill-adapted to the short average stays of some of the 'new' migrant groups. Many EU labour migrants, as well as highly skilled professionals from elsewhere in the world, have no intention of remaining permanently in the Netherlands and no desire to become Dutch citizens. They would, however, beneft from facilities enabling them to learn the Dutch language quickly and to familiarize themselves with the country, its society and its culture. What participation and integration requirements are appropriate for newcomers who are only 'passing through'? Even the new Civic Integration Act fails to address this question adequately.

<sup>47</sup>Algemene Rekenkamer (2017) and Jongen et al. (2020). Sanctions such as fnes or non-discharge of the loan often cause refugee status holders, in particular, further problems. As such, this system can hinder their integration rather than facilitating it (de Lange et al., 2017; de Waal, 2017).

<sup>48</sup>Scholten and van Breugel (2018).

<sup>49</sup>Universal citizenization also implies a policy that is colour blind. This means, however, that it is diffcult to verify whether that policy is actually addressing the problems affecting people with a migrant background. There is no way of checking that it really does impact everyone equally.

<sup>50</sup>Specifc measures are sometimes needed to provide solutions to particular problems. Pragmatic considerations about what works and is effcient should prevail here (Dagevos & Grundel, 2013: 23).


**Table 4.2** Limitations of existing policy models with regard to increased diversity by origin and length of stay

### *4.1.5 Summary*


### **4.2 Little Coherence Between Migration and Integration Policies**

The prevailing policy model governing immigration to the Netherlands has changed on a number of occasions since 1960, but never has there been much coherence with integration policy. In this section we briefy discuss the different emphases within national migration policy over the years.

The leeway available to the Dutch government to regulate migration is limited by European policy and international treaties. Since the second half of the 1990s, for example, labour migration from central and eastern Europe has increased substantially.51 This is due to the enlargements of the EU in 2004 and 2007 and the principle of the free movement of labour within the Union. In addition, some aspects of the immigration of non-EU citizens fall outside the scope of Dutch admissions policy

<sup>51</sup> Jennissen (2011b).

because the Netherlands is committed to compliance with international treaty obligations. These include the provisions of the Geneva Refugee Convention and the European Convention for the Protection of Human Rights and Fundamental Freedoms (ECHR), as well EU regulations such as the Family Reunifcation Directive. These treaties are particularly relevant to asylum and family migration.

Despite these limitations, there remains scope for national policy in this area.52 Table 4.3 summarizes the changing accents in migration policy between 1960 and the present day. Again, these are ideal-type distinctions. The true situation is more diffuse because 'old' and 'new' measures are often in place side by side. Moreover, by defnition admissions policy is restrictive and selective. Nevertheless, differences


**Table 4.3** Dutch migration policy, 1960-present

a As early as 1948, an agreement was concluded with Italy to recruit miners for the Dutch collieries (Lakeman, 1999)

<sup>52</sup>See also WRR (2020).

in emphasis over time can be discerned. Table 4.3 shows that there has been a gradual shift from a recruitment-based policy aimed at low-skilled labour migrants to a more selective one prioritizing the highly skilled. The different emphases at different times were in large part a reaction to contemporary migration patterns and to evolving views concerning the function of migration policy for the Dutch economy and society.

The 1960s and '70s were characterized by the recruitment of low-skilled workers for the expanding Dutch economy. The government signed recruitment contracts with several Mediterranean countries, allowing employers to look actively for workers in them. Prospective migrants were also allowed to seek work in the mining, shipbuilding, metal and textile industries after entering the country on a tourist visa. The Foreign Labour Act (Wet Arbeidsvreemdelingen, WAV) of 1969 permitted the free movement of labour within the Benelux countries. This period of active recruitment coincided with the laissez-faire policy with regard to integration: no thought was given to the link between migration and integration as it was assumed that migrants would eventually return to their own countries.

The legal recruitment of labour migrants ended in 1975, in the wake of the economic recession of 1973.53 Irregular migration to the Netherlands continued, however, especially from Morocco and Turkey. The 1970s and '80s are known as the 'years of tolerance'. During this period it remained relatively easy to enter the country and obtain a tax number without holding an offcial residence permit and so work semi-legally in certain sectors of the Dutch economy.54 The migration policy of the time can be characterized as liberal – as refected, for example, in the easing of restrictions on family migrants and the Nationality Allocation Agreement ('Toescheidingsovereenkomst') between the Netherlands and Suriname. This provided that anyone born in the then colony but settled in the Netherlands by 25 November 1975, the date of Surinamese independence, retained their Dutch citizenship. And for the next fve years, until 25 November 1980, it remained relatively easy for Surinamese to acquire Dutch nationality. These arrangements and the economic malaise in the new republic prompted a mass exodus to the Netherlands: between 1974 and 1980, more than 110,000 people arrived from Suriname.55 Throughout these years the link between migration policy and integration issues remained a topic only rarely discussed. Nevertheless, the infux from Suriname went hand in hand with serious integration-related problems encompassing the labour market, education and housing.56

The period of liberal migration policies was followed by one of increasing restrictions. In particular, a series of measures was introduced to combat irregular migration. The conditions governing labour and family migration were also tightened and a new Aliens Act (Vreemdelingenwet) entered force in 2000 with the aim

<sup>53</sup>See Lucassen and Lucassen (2018).

<sup>54</sup>See Burgers and Engbersen (1996).

<sup>55</sup>See Oostindie and Schoorl (2011).

<sup>56</sup>See Vermeulen and Penninx (1994) and Engbersen (2003).

of making the Netherlands less attractive for asylum seekers. This was in part a response to the large numbers of refugees who had arrived in the 1990s, mainly from former Yugoslavia and from Afghanistan, Iran, Iraq and Somalia. After 2000 their numbers dipped sharply and did not start rising again until 2013, peaking in 2015.57 Still, though, there was no coherent link between migration and integration policies. Major concerns did arise regarding the socio-economic positions of both the traditional groups and the new asylum migrants, however. Many of the latter, especially, were in a very weak position in the labour market.58

After the turn of the millennium, the restrictive policy took on a more selective character. The Highly Skilled Migrants Scheme (Kennismigrantenregeling) launched in 2004 was an important element in this shift, as was the 2006 policy paper entitled 'Towards a Modern Migration Policy' (*Naar een modern migratiebeleid*).59 That document stated as one of its key pillars that policy should focus upon "the need for migrants existing in Dutch society as a whole", whilst "participation in society is expected from the migrant who chooses the Netherlands." Also at its heart was the point of a selective migration policy: "By stating clearly which migrants the Netherlands wishes to admit, the government is not abandoning its basic principle that our admissions policy is restrictive, but rather is combining that with the principle of *selectivity*. The contribution migrants are able to make to Dutch society should therefore play a much greater role in admissions policy than has hitherto been the case."60

The principle of selectivity comes to the fore primarily in the form of the Highly Skilled Migrants Scheme and a number of comparable arrangements designed to attract those with particular know-how and abilities to the Netherlands, or to retain them here, in order to strengthen the nation's economy. By contrast, it plays little or no part in asylum and family migration. In these domains, international treaties and European directives defne the extent to which nation states can be selective. For example, the European Court of Justice (ECJ) has ruled that a Dutch measure to make family migration more selective by raising the income threshold when bringing in a partner from abroad to 120% of the national minimum wage was in breach of the EU Family Reunifcation Directive. On the other hand, the government was able to introduce the Civic Integration Abroad Act (Wet inburgering buitenland, WIB) in 2006. This requires persons wanting to come to the Netherlands as family migrants to pass the basic civic integration examination before arriving in the

<sup>57</sup>See Engbersen et al. (2015).

<sup>58</sup>See Engbersen et al. (2015).

<sup>59</sup>Eligible for this scheme are non-EU/EEA nationals wishing to stay in the Netherlands for more than three months, working for a Dutch employer and earning more than a set amount (with a reduced threshold for those aged under 30). They can also bring members of their immediate family, who are granted free access to the Dutch labour market. Their residence permit is issued for fve years and the procedure runs via the employer, which must be recognized as a sponsor under the scheme.

<sup>60</sup>See *Kamerstukken II* (Dutch parliamentary papers) 2005/2006, 30 573, no. 1. Our italics.

country. Its enactment contributed towards a fall in immigration from Morocco and Turkey.61

The basic principles underlying this 'modern' migration policy align with the citizenization model of integration, in particular its emphasis upon active participation in Dutch society. More than ever before, efforts are now being made to link migration and integration policies. Yet they are still far from being coherent with one another. When it comes to migration, for instance, the main focus is strengthening the Dutch economy. Issues of conviviality arising out of increased diversity receive hardly any consideration, even though – as we have shown in Chap. 3 – the arrival of intra-EU labour migrants and highly skilled professionals who only stay in the country for a few years clearly impacts the social cohesion of neighbourhoods and communities.

In the light of the high net immigration fgures from 2010 onwards and then the refugee crisis of 2015, the ability of the Dutch society to absorb migrants in such numbers became a topic of serious debate. During the annual parliamentary policy debate of 2018, for example, the longer-term consequences for a wide range of policy areas, integration included, of the growth and changing composition of the Dutch population was raised on several occasions.62 The questions now being posed are new ones touching on the mutual coherence of migration and integration policies.

### *4.2.1 Summary*


<sup>61</sup>Nicolaas et al. (2011).

<sup>62</sup>This led to the adoption by the House of Representatives of motion 35 000, no.8, dd. 21 September 2018), tabled by Klaas Dijkhoff MP and others. In this motion the House requests that the government survey the consequences of demographic developments for policy areas such as housing, planning, infrastructure, energy supply, social security, care, education and integration. It also asks the government to develop a variety of demographic scenarios and to propose policy options for each of them.

### **4.3 Inadequate National Models and Local Variation**

The policy models for integration outlined above are national ones. But the substantial differences we have previously identifed between local communities concerning the extent to which they have to deal with increased diversity make it clear that a 'one-size-fts-all' model is not going to suffce.

In the recent past, local policies have often diverged from the national models. In particular, the citizenization policy dominant since the turn of the millennium has not always been followed by local authorities. Some, Amsterdam amongst them, held onto elements of multiculturalism – most notably co-operation and consultations with migrant organizations. Caelesta Poppelaars and Peter Scholten go so far as to describe national and local policies as "two worlds apart".63

In Chaps. 2 and 3 we have identifed eight specifc types of community in the Netherlands, in addition to the 'average' one. This, once again, is an ideal-type classifcation: some places fall into two of more categories. It is especially important to realize that Dutch towns and cities differ widely in the extent to which they have to deal with the consequences of migration. Diversity by origin is very high in the three biggest cities, whereas there are communities in the north of the country, in particular, with virtually no migrant residents. There are differences in migrants' origins, too, and in the length of their stays in the Netherlands. Horticultural districts have to cope mainly with temporary labour migrants from central and eastern Europe, border communities with the presence of Germans who often still work in their own country and expat communities with highly skilled but short-term migrants from Europe and elsewhere in the world, whilst the three major cities play host to almost every migrant group, including many international students.

In Chap. 3 we established that many Dutch local authorities, with the sole exception of those with homogeneous populations, face two major social policy tasks: (1) organizing the reception and integration of newcomers; and (2) strengthening social cohesion for all these residents. How these duties are interpreted, however, depends very much upon the local context.

### *4.3.1 Summary*

National policy models for integration are inadequate because there are so many differences between communities in terms of the origins of their migrant residents and the durations of their stays. Scope for local variations in the reception and integration of newcomers, efforts to foster social cohesion and the promotion of labourmarket participation are therefore essential.

<sup>63</sup>Poppelaars and Scholten (2008).

### **4.4 Conclusion: A Reassessment of the Policy Agenda**

The Netherlands is a country of immigration, although political recognition of this fact has come about only reluctantly. Partly because of that, integration policy in recent decades has been highly reactive and ideologically determined. Whenever views on participation and integration have changed, new policy models have been developed. And the same applies to migration policy: its emphasis has shifted repeatedly in response to evolving patterns of migration and changing views concerning its function for the Dutch economy and society. Furthermore, the new diversity is set to have very different impacts in different communities, with the resulting policy challenges quite possibly varying from one place to another – and sometimes even between neighbourhoods. This means that there can be no 'one-size-fts-all' model for the integration of migrants.

For the future of Dutch society, it is important that policy become less volatile and ad hoc. Instead, it needs to be given a solid and permanent foundation. But one with scope for local variations and bearing in mind the necessity to facilitate conviviality of all groups in society. In making this reassessment of integration and migration policy, there are therefore three key starting points.


These points of departure lead us into the next three chapters, in which we deal in turn with the three policy challenges they raise: organizing reception and integration of all (Chap. 5); strengthening the social cohesion of communities (Chap. 6); and making migration and integration policies more coherent with one another (Chap. 7).

**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 license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license 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 5 Better Local Reception and Civic Integration**

Each year a wide variety of immigrants arrive in the Netherlands, from all parts of the world. Some move on after a short time, others shuttle back and forth to their homeland and others still stay permanently. All these new residents make a home somewhere in the country. For most labour migrants, that is a free choice. Family migrants usually move in with a partner, and asylum migrants are assigned permanent accommodation in a particular municipality once they have been granted offcial refugee status. For all, however, the local government is the primary authority charged with helping them fnd their way in Dutch society.

At present, most local authorities are aware of the settlement of refugee status holders. This is because, under national dispersal agreements, they must arrange accommodation for these newcomers. They also have some insight into those groups required to participate in civic integration programmes.1 Many communities, however, know little about the arrival, stay and departure of other migrants. Consequently, they sometimes only make ad-hoc policies to cater for this infux when it turns out after some time that a specifc group has settled in their area. But as shown in Chap. 3, some communities are seeing new migrant groups arriving on constant basis. More is therefore needed than ad-hoc facilities for the 'groupe du jour'.

The core message of this chapter is that government should organize the reception and integration of new immigrants in a more systematic way. Local authorities need to think about how they can organize *structural* facilities that will help *all groups of migrants* familiarize themselves with Dutch society. Consideration should be given to the *entire migration cycle*, too, from the moment of entry up to and including the moment of departure. In this chapter we discuss several links in the migration chain, and in so doing look at the options available to local authorities and other organizations to promote social cohesion and labour-market participation. Where possible, we back this with research fndings. Unfortunately, though, our

<sup>1</sup>The Civic Integration Act 2007 devolved responsibility for this task to local authorities, but its 2013 successor reversed that change. Under the new system, effective from 2021, they have regained their directorial role.

substantiation is often limited to case studies and organizations' own reports because few scientifc impact studies and policy evaluations have been conducted in this feld.

How can local authorities exert more infuence over the types of migrant who settle in their area? And how can they organize a good reception for all these different groups? We address these questions in turn below, paying particular attention to temporary migration: how can housing and education be better adapted to migrants staying for only a short period of time? What civic integration facilities are required for them and for those who want to remain more permanently? And fnally, how can local authorities facilitate the departure of people who cannot or do not want to stay any longer?

### **5.1 Before Arrival: Infuencing the Nature and Extent of Migration**

How can local authorities dealing with high degrees of both diversity and transience in their migrant populations promote social cohesion and labour-market participation? To start with, they can try to infuence which migrants settle in their area. Current political debate in the Netherlands centres mainly on the issue of the *numbers* of migrants municipalities have to accept – asylum migrants, especially – but there is also a valid discussion to be had about the *types* of migrant suited to a particular community. The huge differences we see at the moment between municipalities in terms of their diversity (see Chap. 2) argue in favour of them being better matched with the place they settle.

This requires that local authorities think more consciously about which migrants will most beneft the economy and society in their areas. Which groups are already present and to what extent can they help strengthen social cohesion or labour-market participation? It also means that national government must be prepared to allow its local counterparts scope to infuence the type of migrants settling in their areas.

The existing legal frameworks offer only limited room for manoeuvre in this respect. The arrival and reception of migrants in the Netherlands are largely matters of national policy, European legal standards, and international treaties. It is, however, possible to exercise some control within these frameworks; for instance, by drawing up agreements between central and local government about the numbers and backgrounds of asylum migrants allocated to individual municipalities.2 Local authorities can exercise indirect infuence, too, by creating favourable conditions for the establishment of international companies in their area and by developing highquality housing and a living climate attractive to specifc groups, such as highly skilled professionals. They can also reach agreements with employers about accommodation and other facilities for temporary workers in agriculture and horticulture.

<sup>2</sup>Cf. Oomen (2019).

### *5.1.1 Attracting Labour Migrants*

Different types of municipality require different types of labour migrant. In horticultural districts such as Westland, Zeewolde, and Horst aan de Maas, for example, these are mainly production workers. Because this workforce will probably increasingly have to come from outside the European Union (EU) in the future, which regions of the world companies recruit from is relevant to local cohesion. For example, would a linguistic or religious 'ft' with migrant groups already present locally be benefcial? Or, conversely, is there a need for greater diversity to prevent concentration and segregation? Based upon the answers to such questions, it may be wise for local authorities to seek to 'steer' employment practices by, for instance, entering into agreements with recruitment agencies and employers or providing specifc facilities and conditions in the felds of housing and education (see Box 5.1). University cities and expat communities, together with companies, academic institutions and central government, already try to make themselves attractive places for highly skilled migrants3 by facilitating a wide variety of international educational provision and cultural facilities and by setting up so-called 'international centres'.

The political question here is how far local governments should go in their facilitating role for companies employing foreign workers. From the point of view of labour-market participation, they can legitimately challenge the 'easy' choice to recruit temporary migrant workers in certain sectors and the role international employment agencies play in this phenomenon. With a view to social cohesion, they could also consider whether forms of migration have displacement effects for existing residents – for example, distortion of the jobs market in their area or an impact upon the local housing market due to longer waiting lists for social housing or rising rents and property prices. In this respect, co-ordination within local authorities between their economic and social affairs departments is of great importance, since the effects of migration differ by policy domain.

<sup>3</sup>*Kabinetsbeleid ter vergroting aantrekkelijkheid NL als vestigingsland voor kennismigranten* ('Government policy to increase the attractiveness of the Netherlands as a country of settlement for highly skilled migrants'); see *Kamerstukken II 2018/2019* (Dutch parliamentary papers), 30573, no. 171, available at: https://zoek.offcielebekendmakingen.nl/kst-30573-171.html. See also Buers et al. (2018).

### **Box 5.1: East Asian Diversity in Amstelveen**

Suitable educational facilities for the children of their employees play a major role for international companies when deciding where to locate.4 A high concentration of east Asian immigrants live in and around south Amsterdam, especially in the Buitenveldert neighbourhood and the adjacent municipality of Amstelveen. They work for companies in information technology, the pharmaceutical and medical industries, and the fnancial sector. The Indian community in Amstelveen, for instance, has grown signifcantly in recent years: from just under 1600 in 2012 to more than 4800 in January 2019. Of the 8698 Japanese nationals resident in the Netherlands, 1687 live in Amstelveen. The town also appears to be an attractive location for companies and employees from China and South Korea. About 19% of its residents hold a non-Dutch passport and people of 134 nationalities currently live there.5

In response to this diversity, Amstelveen offers dedicated municipal facilities such as a Japanese nursery school and an Indian international primary school.6 The local Amstelland Hospital has a physical 'Japan Desk' and a digital 'India Desk', and there are a number Asian dentists and GPs practising in the area. There is also an annual Japan Festival and a Cherry Blossom Festival for Japanese businesses.

Sources: Municipality of Amstelveen; Statistics Netherlands (CBS).

### *5.1.2 Matching Asylum Migrants*

When it comes to asylum migrants, the legal frameworks allow greater scope for management of their geographical distribution across the Netherlands. The Central Agency for the Reception of Asylum Seekers (Centraal Orgaan Opvang Asielzoekers, COA) uses a national dispersal formula to allocate those granted refugee status on a proportionate basis to municipalities throughout the country. The local authority is then obliged to house those it is assigned. Within this system, room could be created for agreements with the COA to 'match' placements to local labour-market needs based upon the backgrounds of the individuals concerned.

In addition, compatibility with existing migrant communities could be a relevant consideration when placing asylum migrants. This might help them fnd their feet more quickly. Under such arrangements, however, central government (and/or the COA) would have to make sure that municipalities do not 'cherry-pick' the most talented or easiest-to-place asylum seekers but also take in their fair share of those harder to deal with. On top of this, the balanced distribution of different types of asylum migrant within municipalities is also important, bearing in mind the ability of

<sup>4</sup>WRR (2013).

<sup>5</sup>Source: Statistics Netherlands.

<sup>6</sup>Andersen (2017).

individual neighbourhoods to 'absorb' them.7 Local authorities need to be careful not to overburden already vulnerable communities with new vulnerable groups.8 A survey of predominantly Syrian refugee status holders, combined with recorded data on ethnic diversity and the Rotterdam District Profle (Wijkprofel Rotterdam) monitoring tool, shows that a positive local social climate facilitates the integration of asylum migrants in reasonably homogenous, predominantly white neighbourhoods, but not in ethnically diverse ones.9 The Netherlands can also learn here from experiences in Australia and Canada with the dispersal and settlement of migrants (see Box 5.2).

### **Box 5.2: Dispersal of Migrants in Australia and Canada**

In Australia, it is not uncommon for cities and regions to reach agreement with the Department of Home Affairs about the backgrounds of the asylum migrants they are allocated. For example, rural areas may ask for asylum migrants with a rural background. The existence in the region of established ethnic communities able to provide newcomers with social support is also considered.

The distribution of migrants across different parts of the country is also an important aspect of Canadian policy. There, newcomers are aided through the Resettlement Assistance Program. This provides income support and help with essential services, including temporary accommodation, life-skills training, and fnancial orientation.10 In all, Canada has more than 60 programmes to facilitate different categories of migrant with their entry and integration into society.

The government department Immigration, Refugees and Citizenship Canada sets long-term targets for the number of immigrants it wants to welcome to the country. This number is broken down by type of migrant and region of settlement. Most newcomers are labour migrants and, since the 2015 refugee crisis, invited asylum migrants. The plans for their reception are based upon demographic trends, operational capacity, community support, political choices, advisory reports, public consultations and society's expected immigrant absorption capacity.11 There are numerous programmes and pilots for economic migrants, including some specifcally aimed at recruiting people to work in the country's more sparsely populated regions, such as the Rural and Northern Immigration Pilot and the Provincial Nominee Program.12 The Quebec Skilled Worker Program focuses upon recruiting talent for the province of Quebec.13

<sup>7</sup>Dekker and Bokhorst (2020).

<sup>8</sup>Leidelmeijer et al. (2020).

<sup>9</sup>Van der Linden (2020). See also Damen et al. (2019).

<sup>10</sup>Government of Canada (2019c).

<sup>11</sup> IRCC (2018a, b).

<sup>12</sup>Government of Canada (2019a).

<sup>13</sup>Government of Canada (2019b).

### *5.1.3 Summary*


### **5.2 Arrival: One Reception for All**

The moment of arrival is an important one for both newcomers and their host communities. For migrants, moving to another country is a source of challenges and uncertainties. This applies not only to those seeking asylum, but also to labour, student, and family migrants. Many do not speak the Dutch language well, if at all, know little or nothing about local amenities, and lack social contacts outside their own immediate circle. A good reception can help them fnd their feet in local society faster and more effectively. It is also important for local authorities to have a clear picture of newcomers from the moment they arrive, to help them start participating in Dutch economic, social and cultural life as smoothly as possible. A well thoughtout 'reception and guidance' policy for all migrants is therefore hugely important.

Current policy in this respect, however, varies widely according to the type of migrant concerned.


<sup>14</sup>Despite their relatively good reception, highly skilled migrants are critical about opportunities to learn the Dutch language and integrate into Dutch society. See: Buers et al. (2018) and van Bochove et al. (2010).

<sup>15</sup>For mediation and services provided by employment agencies for migrant workers from central and eastern Europe, see Strockmeijer (2020).

<sup>16</sup>For the problems Polish migrant workers encounter due to the complexity of Dutch provision and regulations, see Nijhoff (2015).


Reception facilities of various kinds do exist, then, but for some groups these are very limited. Consequently, a fair proportion of new immigrants make a 'false start' when they frst arrive. Basic general services accessible to them all, irrespective of their legal status, motive and intended length of stay, could help them make a better start. A physical local information and registration point at which everyone is welcome would be very helpful in this respect. From here, individuals can be referred to services tailored to their own situation and needs.

Local authorities and institutions would also beneft greatly from such an arrangement, as it can help them learn at an early stage who is having problems or is likely to require further assistance. Particularly when it comes to vulnerable groups such as less well-educated family, labour and asylum migrants, local authorities indicate that they often only gain a clear picture of their needs when they apply for benefts or other social provision. It is precisely these vulnerable migrants, they say, who are most diffcult to reach.21 Identifying them at an earlier stage, at some form of initial reception facility, would enable local authorities to smooth their entry into the jobs market and their general socialization. Such early face-to-face contact

<sup>17</sup>For the problems experienced by international students in participating in Dutch society, see Nuffc (2017).

<sup>18</sup>See Krieg and van Rooijen (2018).

<sup>19</sup> https://www.coa.nl/en/work

<sup>20</sup>One example of how things can be done better is the Welcome Offces (Onthaalbureaus) in Flanders. These are recognized and subsidized by the Flemish government. There are eight offces in all, one in each Flemish province and one each in the cities of Antwerp, Brussels and Ghent. They co-ordinate and supervise civic integration programmes and also work with various other organizations. The aim of their primary civic integration programme is to enable the newcomer to function more independently in Belgian society. Sources: http://agii.be/lexicon/onthaalbureau-0; De Cuyper et al. (2010).

<sup>21</sup>Razenberg (2015).

makes it easier to assess participation opportunities and conviviality risks, too, and so adjust local service provision accordingly.

Our proposal, therefore, is that a network of 'one-stop-shop' reception centres be created for all new migrants settling legally in the Netherlands. The existing reception infrastructure for expats can serve as a basis for these. They should be established at the municipal level in large cities and on a regional basis elsewhere. Such centres would prevent local authorities from constantly having to develop ad-hoc facilities, as when they were faced with the arrival of large numbers of Syrian refugees or EU labour migrants. For asylum migrants, the centres would offer their services in conjunction in with the facilities provided by the COA and the Dutch Council for Refugees.

Broadening the existing facilities to create general reception centres would help all groups of migrants fnd their feet, bringing together public and private services in the process. The advantage of having one centre for everyone is that it brings greater coherence to reception functions and activities, as well as providing a better overview of the full spectrum of services available, addressing immigration and reception issues in a systematic manner and creating a broad network for sharing knowledge and so on (see Boxes 5.3 and 5.4). It might also help improve the intercultural competences of local 'front-line' workers. The great diversity of the migrant population requires professionals able to deal with a wide range of 'clients'.

One disadvantage of such a method is its complexity, not least because it calls for a high degree of public-private co-operation and funding. A now somewhat dated evaluation study of this type of 'one-stop shop' in six European countries concluded that there are benefts to be gained, but that they can prove troublesome at the operational level.22 Especially in western European countries that have been dealing with immigration for a long time, existing structures and rules may hamper innovation. A full-service reception centre stands the best chance when allowed to develop gradually. Where 'expat centres' or 'migrant worker desks' already exist, they can be expanded in phases to serve other groups and provide new services.

### **Box 5.3: International House Copenhagen**

International House Copenhagen (IHC) is a physical building housing all the reception functions for newcomers to the Danish capital. In fact, its role is regional as it operates on behalf of more than 30 municipalities making up the 'Greater Copenhagen Area'. The IHC regards improving reception policy as a must to attract human capital, but also to make Copenhagen the most inclusive city in Europe.

One of its occupants is the International Citizen Service (ICS). This is a reception centre at which migrants can arrange all matters relating to national and local administration, tax, work, income, and civic affairs.23 The integrated

(continued)

<sup>22</sup>Penninx (2009).

<sup>23</sup>Vertelyté (2016).

### **Box 5.3** (continued)

nature of the service – all the relevant organizations are under the same roof, co-operate and attune their provision to each other's needs and those of the migrant – prevents conficts of interest, duplication, confusion and delay. Besides the ICS, other organizations represented at the IHC include the city job centre, the local university, and private entities such as removal companies and childcare providers. This 'one-stop shop' approach to services for migrants is a key pillar of Copenhagen's reception policy.

### *5.2.1 Role of the Reception Centre*

A reception centre of this kind can fulfl several functions. To start with, it helps local authorities to better *monitor* the nature, size, and dynamics of their migrant populations. Which new groups are settling locally, and where exactly? Where and in what groups is there a high rate of turnover? What does this mean for the provision of schooling and other public services? As things currently stand, the national population register is of only limited assistance with these questions. Different types of migrants register in different places, or they fail to do so at all.24 By offering them somewhere to do it easily face-to-face, this basic administrative procedure can be completed quickly and smoothly. As an example of what is possible, in the horticultural Westland district employers and the municipality organize registration evenings for labour migrants at their accommodation or place of work, where they also receive information about and referrals to any other services they may need.25

With a view to encouraging labour-market participation and social cohesion, it is important to *inform* newcomers about public services at the earliest possible opportunity and to *refer* them to appropriate services where necessary. This can be organized in different ways. The Hague, for example, holds dedicated 'consultation hours' for EU migrant workers.26 The advantage of this approach is that it delivers

<sup>24</sup>As soon as they arrive in the Netherlands, highly skilled migrants and postgraduate students can visit the Immigration and Naturalization Service (Immigratie- en Naturalisatiedienst, IND) desk at an expat centre to submit their application for a temporary residence permit, have their biometrics taken and collect a provisional residence document. They can also register their address in one of the affliated municipalities. EU labour migrants do not usually visit expat centres and are only required to register with a local authority after 3 months. A proportion of this group remains unknown to the authorities, because as EU citizens they are not subject to immigration controls and they never register their address. Some avoid doing this because they live in irregular or overcrowded accommodation. People who reside in the Netherlands for less than 4 months can be signed onto the Register of Non-Residents (Registratie Niet-ingezetenen, RNI); currently, 19 local authorities have a dedicated registration desk for this group. Asylum migrants report frst to an asylum application centre (at Amsterdam Schiphol Airport, Ter Apel, Den Bosch or Zevenaar) and are then sent an asylum reception centre. Family migrants must arrange their residence permit in advance, in the country of origin, through the IND; upon arrival, they then register with the local authority.

<sup>25</sup>See www.gemeentewestland.nl/verhuizen/inschrijving-arbeidsmigranten.html

<sup>26</sup>Norder (2014).

a customized service, but one drawback is that there is no continuity of provision and other migrant groups do not beneft.

At a reception centre, newcomers would receive *advice* about civic integration programmes, language schools and so on. Civic integration is usually mandatory for asylum migrants, but for most highly skilled ones it is voluntary. Nevertheless, many of them would like to learn Dutch. A reception centre can tailor its advice to these specifc requirements and wishes. Together with training colleges and employers, it can also provide guidance regarding the local jobs market. And it can connect newcomers with organizations, businesses and individuals relevant to their own situations, be they refugee counsellors, language 'buddies' or local networks of migrant entrepreneurs.

A fnal possible function is *socialization*. Students and highly skilled migrants, for example, often say that they miss social contact with Dutch people.27 Whilst 70% of highly skilled migrants have a generally positive view of the Netherlands, to the extent that they would like to extend their stay here, they are also critical of the country as a place to settle and build a career. They are less than satisfed with their social reception, too, including opportunities to learn Dutch. Some even leave prematurely as a result.28

Meanwhile, more and more local authorities are choosing to immerse asylum migrants in a process of socialization from the moment they arrive, with introductory days and accelerated civic integration trajectories. Increasingly, they are also asking the COA to allow asylum seekers accommodated locally to stay in the area once they have been granted refugee status and are being assigned a municipality to reside in permanently.29 This allows their socialization to continue uninterrupted, using the contacts and networks they have already built up.

### **Box 5.4: Canadian Welcome Centres: An Integrated Approach**

In Canada, so-called Welcome Centres have been set up all over the country. It is not uncommon for new immigrants to visit one as soon as they arrive, sometimes straight from the airport. They are comprehensive facilities at which newcomers all kinds are welcome, from asylum seekers to migrant workers, students to family migrants. The centres give out free information and offer referrals to specifc service providers. Subjects covered include government paperwork, naturalization, the labour market, education, assessment of foreign qualifcations, language courses (English and French), healthcare, housing, legal matters, volunteering, recreation, income, taxes, food and clothing. The Welcome Centres thus familiarize immigrants with every aspect of life in Canada, from socio-economic participation to the nation's social structures.30

### (continued)

<sup>27</sup>Buers et al. (2018). For student migration, see the box in the next section. See also van Bochove and Engbersen (2015).

<sup>28</sup>Buers et al. (2018).

<sup>29</sup>Vriesema (2016).

<sup>30</sup>www.welcomecentre.ca/settlement/index.html

### **Box 5.4** (continued)

For some newcomers, the process begins before they even reach the country. The Canadian Orientation Abroad programme is designed to prepare selected immigrants, through interactive sessions, for a successful life in Canada. Evaluations show that participants signifcantly increase their knowledge of Canadian society and have a more realistic view of emigration and their destination country than migrants who do not follow the programme.31

### *5.2.2 Summary*


### **5.3 Residence: More Consideration of Temporary Stays**

Migrants differ widely in their reasons for coming to the Netherlands and their length of stay. We have seen in Chap. 2 that, on average, nowadays they remain for shorter periods than ever before. This applies to labour and student migrants in particular, but also to some asylum migrants.32 In other words, large numbers leave again sooner or later – sometimes to try their luck elsewhere, sometimes to return to their country of origin and sometimes because they are forced to move on due to the fact that they have no means of support or have not been granted legal residence. Others shuttle back and forth, living part-time in the Netherlands. This fuidity is part and parcel of a global economy, but it does not leave the social fabric unaffected.

### *5.3.1 Housing*

Many local authorities struggle to house temporary EU labour migrants, in particular (see Box 5.5).33 The migrants themselves also indicate that they have diffculty fnding a home that matches their expectations.34 Due to the shortage of suitable

<sup>31</sup> IRCC (2018a, b).

<sup>32</sup>Leerkes and de Hoon (2019).

<sup>33</sup>Van Ostaijen et al. (2018).

<sup>34</sup>Holtslag et al. (2012).

accommodation, some end up living in unsafe and overcrowded conditions.35 More and more local authorities and housing corporations are therefore experimenting with temporary accommodation for short-stay migrants. There is a huge need for such solutions.

It is important not only that there be a suffcient supply of housing, but also – given that it is hard to predict how long people will stay – that this supply can quickly be scaled up or down. Temporary migrants generally have little or no need to be at the heart of the community or to develop a local network, so for them such forms of 'new' housing can sometimes be located away from built-up areas. This also reduces the pressure on the regular housing market. Examples of such solutions include so-called 'Polish hotels', where hundreds of seasonal workers are accommodated. These are often in old offce buildings in otherwise non-residential areas. Another example is 'container homes' – literally converted shipping containers – for refugee status holders, in some cases mixed with people with a Dutch background. The accommodation of short-stay labour migrants is an important responsibility for employers and requires better regulation of the temporary employment sector.36

### **Box 5.5: Migrant-Worker Housing Problems in The Hague**

In The Hague, the main problems associated with migrant housing affect those staying no more than 1 year. Of the 30,000 central, eastern, and southern European workers living in the city in 2017, fewer than 1000 were in dedicated short-stay accommodation such as 'Polish hotels'. The rest had to rely upon the regular housing market. In response to this demand, unscrupulous landlords and employment agencies buy up low-quality housing to rent to migrant workers. These properties are often overcrowded and generate nuisance for local residents.37 Moreover, much of this transient population is unknown to the city authorities. Research reveals that approximately half of labour migrants in The Hague do not register as residents;38 either their stay is too short or their landlord and/or employer does not allow them to.

The Hague Housing Inspection Bureau (Haagse Pandbrigade) increasingly fnds migrant workers being accommodated in business units and commercial premises with inadequate fre precautions. They also come across bunkhouses in buildings owned by property investors, where middlemen rent out individual beds. In many cases there are too many people living in a dwelling, sometimes due to the offcial tenants allowing relatives, friends, or acquaintances to move in as well. With each occupant being charged rent on an individual basis, more people means more income. Amongst other tools, the Housing Inspection Bureau uses the app Meld een Vermoeden ('Report a suspicion') to gather information about such situations.39

<sup>35</sup>Rijksoverheid (2019a).

<sup>36</sup>Aanjaagteam bescherming arbeidsmigranten (2020).

<sup>37</sup>Gemeenteraad Den Haag (2019).

<sup>38</sup>Nienhuis et al. (2017).

<sup>39</sup>Gemeenteraad Den Haag (2019).

### *5.3.2 Education*

An infux of migrants, often on a temporary basis, can place local educational provision under a lot of pressure. Schools must deal with unexpected spikes in pupil numbers, high rates of turnover and intake and outfow at irregular times. These are issues associated with various types of migrant, ranging from highly skilled professionals and EU workers to refugees.40

For highly skilled migrants, the availability of (private) international schools is a magnet.41 But although the Netherlands has the highest rate of growth for these institutions in Europe, many still have waiting lists. Even for well-paid professionals, moreover, the high fees can be a barrier. Only a minority of employers contribute towards these costs. That aside, a lot of these migrants prefer to send their children to 'local' schools. Combined, these factors mean that approximately half of the highly skilled migrants in the Netherlands make use of mainstream education.42

Amongst this group, Dutch schools providing bilingual education are particularly popular. This redoubles the requirement that the quality of teaching in the second language (usually English), in particular, stay up to standard.43 Based upon earlier forecasts, by 2020 the Amsterdam metropolitan region expected to have more than 23,000 children of highly skilled international professionals at its schools, about half of them attending regular ones and the other half international institutions.44 Elsewhere, such as in the Laakkwartier district of The Hague and in horticultural districts, schools are dealing with an infux of large groups of children of EU labour migrants.

The arrival en masse of asylum migrants in 2015 revealed that it is diffcult to scale up educational provision for newcomers when there is a sudden peak in demand. This was due to a combination of accommodation and staff shortages, funding shortfalls, lack of knowledge about the new students and insuffcient public support.45 In the face of those challenges, regional co-operation between local authorities and educational institutions proved decisive in ensuring that such a large group of new arrivals was provided with high-quality education, as well transport to and from school where necessary. Because of the unpredictability of current and future migrant fows, it is important that these regional collaborations be maintained on a permanent basis to mitigate the need for unsatisfactory ad-hoc arrangements. Another lesson learnt from 2015 is that it is better for children's social, emotional and cognitive development for the COA to minimize relocations of families between asylum reception centres. This remains common practice, but is tough on the children and diffcult for schools. By accommodating asylum seekers with a good chance of being granted offcial refugee status at locations in the region where they

<sup>40</sup>Leeuwestein and Bokhorst (2018).

<sup>41</sup>Ministerie van Economische Zaken (2015).

<sup>42</sup> van der Wel et al. (2016).

<sup>43</sup> Inspectie van het Onderwijs (2019).

<sup>44</sup>Decisio (2018: 9).

<sup>45</sup>Onderwijsraad (2017), 'Vluchtelingen en onderwijs' [Refugees and education].

will eventually be resettled permanently, the disruption of school careers can be kept to minimum.

Greater professionalization is also needed to guarantee the quality of education for newcomers. The teaching of Dutch as a second language is a key pillar in this respect. Teachers must have the right skills and their classroom materials are also in need of improvement. Because the target audience for this subject is so disparate, many commercial publishers do not see it as a worthwhile market and so shy away from investing in the production of high-quality textbooks and other materials. A more active government role in this domain could contribute towards the necessary professionalization. We examine this topic in more detail in Chap. 6. 46

For schools educating children who stay in the Netherlands only temporarily, good basic support is important. For example, funding based upon the length of their pupils' stays. There is also a need for standardized non-linguistic tests to enable schools to properly assess pupils' subject knowledge at intake without the results being affected by their lack of profciency in Dutch.47

### *5.3.3 Summary*


### **5.4 Settlement: Civic Integration for All**

Approximately half of the immigrants to the Netherlands stay longer than 5 years, although for many that was not their original intention. A good proportion of asylum migrants, for example, especially those from politically and economically unstable regions such as the Middle East, are simply unable to return to their countries of origin. International students fnd a job or partner here. And labour migrants see their children growing up settled at school and so put off their return again and again. Some of these groups are required to undergo intensive civic integration programmes, complete with demanding mandatory language acquisition and socialization targets and intended as a pathway to Dutch citizenship. For many others little or nothing is arranged.

Current civic integration policy focuses exclusively upon newcomers from outside the EU who are taking up long-term residence in the Netherlands. For them, participation in such a programme is mandatory. This does not apply to other groups,

<sup>46</sup>Onderwijsraad (2017), 'Vluchtelingen en onderwijs' [Refugees and education].

<sup>47</sup> Inspectie van het Onderwijs (2016).

though. For example, European regulations exempt intra-EU migrants from civic integration requirements. And domestic rules extend that dispensation to non-EU students and highly skilled migrants who are staying only temporarily. As a result, large numbers of immigrants receive little or no guidance on settling into Dutch society. Yet that is something they all need to a greater or lesser extent, albeit in different ways. Below we outline the variety of situations facing newcomers not subject to mandatory civic integration.

### *5.4.1 EU Labour Migrants*

The number of intra-EU migrants, especially from central and eastern Europe, has increased considerably in recent years. They are not obliged to undergo civic integration and often have little contact with the authorities. Many do not even register as residents and make almost no use of local facilities. But their settlement is not always problem-free. According to research by Platform Integration & Society (Kennisplatform Integratie en Samenleving), no fewer than 70% of municipalities hosting EU migrant workers in any numbers experience problems as a result, mostly in the areas of housing, registration, language and public nuisance.48 But only 20% have drawn up policies specifc to this group, covering civic integration, the education of their children and the jobs market.

For EU migrants who remain in the Netherlands for only a short period of time, the most pressing issues are practical matters such as working conditions and housing. In practice, though, it is often unclear how long they stay because many do not register as residents. It is therefore important that local authorities arrange an early contact moment with all of them to record when they arrive in the country. Those who do settle can then be offered access to a range of provision to help them integrate. This approach is in line with a call from the Advisory Committee on Migration Affairs (Adviescommissie voor Vreemdelingenzaken, ACVZ) to create a (voluntary) civic integration scheme for these migrants and to encourage their participation in it.49

Research shows that EU labour migrants intending to stay in the Netherlands for longer periods have a great need to learn Dutch. Four out of fve Poles living here have diffculties with the language, even after years in the country.50 Many are keen to learn it, but say that that is diffcult to combine with long working days.51 For this group, then, it is important that evening classes be provided. These could be organized through their employers, but local authorities can also offer advice about

<sup>48</sup>De Gruijter and Razenberg (2015).

<sup>49</sup>See ACVZ, 2019, 'Wetsadvies "Voorstel van wet houdende regels over inburgering in de Nederlandse samenleving" (Wet inburgering 20…)' [Advisory submission concerning the draft bill regulating civic integration into Dutch society (Civic Integration Bill 20…)].

<sup>50</sup>Klaver and Odé (2017: 46).

<sup>51</sup>Dagevos (2011).

suitable voluntary civic integration programmes from private providers, including e-learning options.52

EU labour migrants also indicate that they have a wide range of practical questions concerning such matters as registration, housing, employment and benefts.53 The information available online is often too general or provided only in Dutch. This is particularly diffcult for the less well-educated. In addition, they need information about healthcare, education and debt.

### *5.4.2 Highly Skilled Migrants*

Highly skilled migrants often come to the Netherlands on a temporary basis and are therefore not subject to mandatory civic integration. Currently, they can turn to various expat desks or international centres for help with practical matters like registration, residence permits, healthcare and housing. Nonetheless, members of this group often indicate that they fnd it diffcult to participate fully in Dutch society. Once again, the main obstacle is the language.54 These migrants therefore need Dutch courses, too, but they also beneft from informal contact with native speakers. Local authorities can refer people to language schools and associations, of course, but they could also combine informal contact with language learning by, for example, sponsoring mentoring or buddy programmes. Although currently used mainly used by asylum migrants, initiatives of this kind would also be useful for highly skilled international professionals. Their integration is important because some end up staying longer than initially foreseen.55

### *5.4.3 Student Migrants*

More and more foreign students in the Netherlands are staying on after they have completed their course here.56 This group is not subject to mandatory civic integration but would still beneft from provision in that area. Many have diffculty fnding their way in Dutch society and so hardly ever encounter the local culture,57 a fnding reiterated by the Annual International Student Survey.58 Seventy-fve per cent have

<sup>52</sup>Engbersen (2012: 116). Labour migrants can learn Dutch through the national 'Count on Skills' ('Tel mee met taal') programme, as well as adult-education courses provided by local authorities. Employers can apply for subsidies to support language tuition for their staff, too, but not all make the effort. See: www.telmeemettaal.nl/subsidie-en

<sup>53</sup> de Gruijter et al. (2016).

<sup>54</sup> van Bochove et al. (2010).

<sup>55</sup> van Bochove et al. (2010).

<sup>56</sup>Nuffc (2017).

<sup>57</sup>LSVb (2019).

<sup>58</sup>Verbeek (2019).

little contact with 'locals', and more than a third say they are 'dissatisfed' or 'very dissatisfed' with the available opportunities to learn Dutch.59

These students often fnd themselves stuck in an 'international bubble', due to a combination of factors including the language barrier, separate accommodation and the fact that Dutch student societies and sports clubs are not always open to them. Of these, language is the main hurdle preventing them from remaining longer. They can manage with English during their studies, but those who stay on often fnd it hard to build a social network or to communicate without profciency in Dutch.60

### *5.4.4 Bespoke Civic Integration for All*

We are doubtful that a civic integration policy designed primarily as a path to Dutch citizenship for permanent migrants is appropriate to the realities of the modern migration society. The sharp distinction between those subject to mandatory integration and that exempt is out of line with the actual needs of the various migrant groups in the Netherlands, whether their stay is long or short. Those needs are more a spectrum than the dichotomy created by the current legislation.

In practice, to some extent central and local governments are already responding to the new reality. For example, by providing language-learning opportunities for those not subject to mandatory civic integration.61 Some local authorities are making certain services, such as job-application training and introductory tours, available to all residents regardless of their migrant background. Amsterdam, for example, offers thematic work-related language courses to anyone who needs them.62 This accessibility also encourages intergroup contacts and bolsters public support for such schemes.

Given the increasing diversity of the Netherlands' migrant population – by origin, length of stay and schooling – what is really needed is a differentiated policy including provision for groups not subject to mandatory civic integration. Local authorities do not have to do everything themselves; they can also refer newcomers to external providers of language courses and workshops, say, or they can buy in these services – perhaps in partnership with employers, volunteer projects and refugee organizations.63

<sup>59</sup>Nuffc (2019).

<sup>60</sup>Blaauwberg (2016) and Stravens and Cornelissen (2017).

<sup>61</sup>See the subsidy scheme outlined at www.telmeemettaal.nl/subsidie-en

<sup>62</sup>Noordhuizen et al. (2016).

<sup>63</sup>Another crucial issue here, of course, is the fnancing of voluntary civic integration. The ACVZ (2019, 'Voorstel van wet houdende regels over inburgering in de Nederlandse samenleving') has looked at the situation in Germany, where "large numbers of EU and Turkish migrants participate voluntarily in civic integration programmes. Moreover, the German government bears much of the cost of the course and the fnal test, and the percentage of migrants achieving B1-level language profciency (52%) is much higher than in the Netherlands (only 2%). Germany thus achieves better results with fewer obligations."

How exactly this broad range of integration services is organized will vary from place to place, according to local needs. In this respect, too, it is wise to allow differentiation. In communities with very high levels of diversity, such as The Hague, a central 'integration desk' can help refer different migrants to the right programme or course. Where the degree of diversity is far lower, as in those municipalities where most newcomers are Poles, say, or Germans, local policy can focus more upon these specifc groups. Under the auspices of the Association of Dutch Municipalities (Vereniging van Nederlandse Gemeenten, VNG), agreements can be reached as to what basic provision is essential to prevent legal inequalities arising between municipalities.

### *5.4.5 Summary*


### **5.5 Consider Departure Too**

Now that more and more migrants in the Netherlands are staying only temporarily, greater consideration needs to be given to facilitating their departure. Just as when ensuring the smooth arrival of newcomers, specifc policy is required for a smooth exit from Dutch society. Current efforts in this respect are pragmatic: migrants deemed unwanted are encouraged to leave and those who are wanted are encouraged to stay and settle. In practice, this means that departure-related measures selectively target failed asylum seekers and labour migrants with poor employment prospects.

Whether the government can infuence migrants' decisions to stay or to leave through a 'soft-touch' integration policy is a matter of debate. Past research amongst asylum migrants is inconclusive in this respect.64 Those who do leave are often wellintegrated in socio-economic terms. Migrants with a job, for example, are more likely to go than migrants without one.65 A Swedish study focusing specifcally upon refugee status holders shows that those in the highest income group and with

<sup>64</sup>Engbersen et al. (2015).

<sup>65</sup> Jennissen and Oudhof (2008).

the best schooling are most likely to leave the country eventually. One possible explanation for this is that work experience, income and education are important resources for the ability to emigrate.66 On the other hand, not being able to fnd suitable work is sometimes also a reason for leaving.67 Departure can therefore be the result of both 'success' and 'failure'. Sociocultural integration, such as contact with people with a Dutch background and identifying with the Netherlands, appears to reduce the motivation to leave.68 Since socio-economic and sociocultural integration often go hand in hand – but may thus have opposite effects – the question of their combined bottom-line impact remains open.69

Research has shown that good fnancial support and mediation by someone of the subject's own nationality seems to work in encouraging the departure of failed asylum seekers who have exhausted all legal avenues.70 The Ministry of Justice and Security subsidizes courses for people in this category to equip them better for the future, in the hope that a positive outlook will increase the chances that they depart voluntarily. These courses are not evaluated as to their effectiveness, however. A review of repatriation policy has shown that it is not possible to determine the effectiveness of specifc measures. And an inventory of the literature reveals a range of opinions about the extent to which repatriation support affects actual decisions to leave the country.71 For those already thinking about going, it may play a role in fnally deciding to pack their bags or bring forward their departure. But this is so not much the case when departure is not yet being considered or when a person has decided to stay in the Netherlands at all costs.

### *5.5.1 Local Departure Policy*

At present, there is strong central control over measures to repatriate failed asylum seekers and undocumented migrants with no further legal recourse. Five major cities have been designated to operate a so-called 'Bed, Bath and Bread' scheme, which provides only basic, subsistence-level support in the hope that this will persuade them to leave of their own accord. Meanwhile, the Repatriation and Departure Service of the Ministry of Justice and Security employs an intensive one-to-one approach to the same end.72 In Belgium, too, the government encourages such migrants to start thinking realistically about their future and not continue to regard securing refugee status as the sole solution to all their problems.73

<sup>66</sup>Klinthäll (2007).

<sup>67</sup>Cassarino (2004).

<sup>68</sup>De Haas and Fokkema (2011) and de Vroome and van Tubergen (2014).

<sup>69</sup>Cf. de Vroome et al. (2012).

<sup>70</sup>Leerkes et al. (2014).

<sup>71</sup>Ministerie van Justitie en Veiligheid (2019: 8).

<sup>72</sup>Winter et al. (2018).

<sup>73</sup>Petrovic et al. (2014: 29–32).

Local authorities should be given greater scope to organize their own repatriation support services for specifc groups, including failed asylum seekers and undocumented migrants. For some, it may also make sense to help labour migrants return to their home countries or move on elsewhere (see Box 5.6) in times of economic decline. Since some employers and agencies provide short-stay workers with housing linked to the job, they are in part responsible for an increase in homelessness when contracts end or they lay people off.74 They need to be held accountable for this, and should be required to arrange repatriation or provide help fnding alternative accommodation.

### **Box 5.6: Self-Help to Repatriate Homeless Polish Labour Migrants**

The Barka Foundation is a Polish charity which manages dozens of sheltered workplaces, residential communities, and reintegration projects in Poland itself, but also has an offce in the Netherlands. For homeless eastern European migrants who want to stay in the country and still have some chance of success here, Barka NL offers help to re-enter the housing and labour markets. In 2013 it opened a Social Economy Centre in Utrecht to provide training, advice and support for this purpose. It also organizes voluntary repatriations to Poland, currently operating projects in Utrecht, The Hague, Amsterdam, Rotterdam, Arnhem, Eindhoven, Nijmegen and Breda with funding from central and local government.

Barka NL works with mobile teams deployable throughout the Netherlands and consisting of a leader and an assistant, both from Eastern Europe. The leader is always an 'expert by experience' in addiction and homelessness, whilst their assistant is formally trained as a psychologist or social worker and is familiar with the Dutch language and institutions. The charity maintains relationships with hospitals, day and night shelters, the police, social services and other agencies to make and maintain contact with disadvantaged Poles and other eastern Europeans. If necessary, its counsellors attempt to restore broken ties with family in the home country and to facilitate voluntary returns to help people overcome an addiction and/or reintegrate into society.

Each year Barka NL teams speak to about 2000 central and eastern Europeans, around 500 of whom return home. According to its own fgures, a total of 2678 people were repatriated voluntarily between its launch in 2012 and the end of 2017.75 Most were Poles, but they also included other eastern Europeans. The majority returned to family, with the rest going to Barka communities in Poland or to shelters or addiction clinics in their home countries.

<sup>74</sup>Rijksoverheid (2019b).

<sup>75</sup>*Barka Reconnection Projects for homeless Middle and Eastern European migrants in the Netherlands – Results (01.01.2012–30.09.2015)*; Jongejan and Partners (2018).

It is also worthwhile, though, to think about how best to facilitate the departure of migrants who have found success in the Netherlands. As we have seen, large numbers across all categories leave the country after 5–10 years here. Local authorities could smooth this process by, for example, introducing simple municipal and school deregistration procedures and by aiding the fnal settlement of social security or pension entitlements. A local or regional reception centre, as proposed above, could also play a role in this. In addition, it is well worth trying to ensure that children's development suffers as little as possible in their move to another country. Consider designing curricula dovetailed as far as possible with education elsewhere, for instance, as international schools already do.

### *5.5.2 Summary*


### **5.6 Conclusions**

Government needs to make systematic provision for the reception of new immigrants. This is primarily a task for local authorities, working in co-operation with civil society organizations and employers. They are going to have to think about how they can create *structural* facilities to help *all migrant groups* fnd their way in Dutch society. And to prevent frictions affecting social cohesion and labour-market participation, they will need to meet the challenge of effectively organizing a variety of reception services.

A more proactive reception policy at the local level requires appropriate fnancial and material support from central government. Along with the VNG, central government can also help here by developing know-how and by identifying and sharing best practices. Moreover, not all costs need be borne out of the public purse. Particularly in the case of EU migrant workers and highly skilled professionals, part of the bill for reception and civic integration can be passed on to the employers who bring them here. This also prevents employers from shifting the entire social burden of labour migration to the community and overloading society's ability to absorb newcomers. In the case of family migrants, partners can reasonably be expected to bear part of the cost of reception and civic integration. As for students, educational institutions can offer language courses and mentoring schemes. Finally, language

buddies, residents' organizations and community centres can all help asylum migrants fnd their way in Dutch society.

The main conclusions of this chapter are as follows.


Local authorities can do more to facilitate the smooth departure of migrants – by reaching agreements with employers and agencies, for instance, but also by offering support through the reception centres.

**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 license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license 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 Strengthening Social Cohesion**

The increasing diversity of the Dutch population is making conviviality more complicated. Not only in the big cities and their suburbs, but also in medium-sized cities, expat and horticultural municipalities (see Chap. 3). Moreover, modern migrants do not stay in the Netherlands as long as they used to; half have left again after 5 years. This places considerable demands upon schools, for example, which have to deal with pupils arriving and leaving throughout the course of the year. For voluntary associations, too, a high turnover of members is not conducive to cohesion. And the same applies to neighbourhoods where much of the population is just 'passing through'.

### **6.1 Diversity and Transience Put Social Cohesion Under Pressure**

Whether a lack of social cohesion is regarded a problem is, ultimately, a political consideration. Whilst certainly important, it is not the only thing that counts in life. Many people regard personal freedom and self-fulflment as equally valuable, and not everyone needs close contacts with their neighbours. After all, an absence of social obligations can also be liberating. Ever since Max Weber and Georg Simmel, social scientists have pointed out that urban environments offer their residents greater scope to shape their own identity and individuality. In a certain sense, therefore, a relatively low level of social cohesion and security are the price the inhabitants of a city pay for the freedom and privacy it also offers. These observations call for a certain degree of caution in the policy arena. Moreover, the ways in which people coexist are not set in stone. Social cohesion evolves and takes on different forms over time.

Nevertheless, a great many people view lack of cohesion as one of the most serious issues in Dutch society today. Four times a year the Netherlands Institute for Social Research (*Sociaal en Cultureel Planbureau,* SCP) polls public perceptions of the main problems facing the country. Figure 3.1 (in Chap. 3) shows that the subjects 'immigration and integration' and 'conviviality' have long been high on this national list, and in recent years have consistently topped it – beating even concerns about incomes and the economy. According to the SCP, approximately 40% of adults agree with the statement that the Netherlands would be more pleasant as a country if it had fewer migrants1 and more than half believe that the nation is in danger of losing too much of its distinctive identity because of immigration and open borders.2 Furthermore, almost half (45%) of respondents without a migrant background have concerns about increased diversity.3

If these widely held apprehensions are not taken seriously, they have the potential to fuel social discontent and reinforce political polarization. When not everyone feels suffciently heard and represented, their dissatisfaction can lead them to drop out of the mainstream political process and reject democracy.4 In this chapter, therefore, we discuss what the government could do to strengthen the cohesion of a society in which both diversity and transience are on the increase.

### *6.1.1 Issues Around Social Cohesion*

In the Netherlands, the topic of social cohesion encompasses a whole gamut of more specifc issues. Traditionally, particular attention has been paid to those arising out of the defcient social integration of certain groups with a migrant background. This is a *compositional* effect. Some groups have been a source of concern because they display relatively high rates of antisocial behaviour. One example is young men of Moroccan origin, who are overrepresented in the crime statistics.5 The causes of this lie in both their upbringing and their living environment, and tackling the problem requires constant, targeted action on the part of local authorities. This is the classic 'integration agenda' aspect of social cohesion, about which much has already been written, not least by the WRR, and we have little to add about it in this study.6

Instead, we home in on another issue: the social unease caused by the increasing diversity of the population as a whole. A result of different groups living side by side, this can be observed across all groups. This is a *diversity* effect and involves what the American political scientist Robert Putnam has called 'hunkering down'.

<sup>1</sup> den Ridder et al. (2019: 20).

<sup>2</sup>Dekker and van Houwelingen (2017: 31).

<sup>3</sup> den Ridder et al. (2019: 57).

<sup>4</sup>Staatscommissie parlementair stelsel (2018).

<sup>5</sup>See also: Jennissen (2014, 2017).

<sup>6</sup>See also: WRR (2001, 2006, 2007), Gijsberts and Dagevos (2007), and Dekker and Bokhorst (2020).

As diversity and transience increase in a neighbourhood, so too do social diffdence and mutual incomprehension, leading residents to retreat more and more behind their own front doors. This is a 'diversity agenda' aspect of social cohesion.

That agenda is linked directly with contemporary migration patterns. Moreover, it is relatively new and unexplored. Local authorities these days are asking themselves how they can facilitate conviviality between different groups and in so doing create a new sense of community. It should be noted that these two issues, defcient integration and social unease, are increasingly coinciding so that many neighbourhoods now face general diversity effects on top of long-standing problems associated with specifc groups. Some of the measures we propose are therefore derived from research into the traditional integration agenda or may be relevant to it.

In compiling our proposals, we have looked at fndings from Dutch and international literature and considered whether there are lessons to be learnt from other countries. We do this at three different levels.


In 6.2 we discuss the neighbourhood as a place where social ties are forged, in particular considering its physical confguration and social infrastructure, in which public spaces serve as meeting places. In 6.3 we examine how diversity has impacted two important types of social institution: schools and voluntary associations. And in 6.4 we discuss whether cohesion also needs to be strengthened at the national level, and how this could be done. Finally, 6.5 outlines our main conclusions.

### **6.2 Strengthening Neighbourhood Social Cohesion**

What can local authorities do to maintain, and perhaps even strengthen, social cohesion in neighbourhoods in the face of increasing diversity and transience? Below we discuss some possible policy directions, based upon experiences in our own country and abroad. In so doing, we distinguish four broad categories of mechanism at work here.7 The frst is *social-interaction mechanisms* aimed at strengthening contacts between members of different population groups – a policy approach tried and tested in the United Kingdom and Canada in recent decades. Secondly, *environmental mechanisms* focus upon the confguration of public space. We discuss how projects in this domain have enhanced so-called 'public familiarity' in the Netherlands and Germany. Our third category is *institutional mechanisms*: how local institutions such as libraries, housing corporations and community-building organizations can contribute towards greater cohesion. Finally, we look at *cultural mechanisms*, paying attention to the importance of local identities for social cohesion.

<sup>7</sup>Based in part upon Galster (2012).

### *6.2.1 Strengthening Social Interaction: Experiences from the United Kingdom and Canada*

The Netherlands is not the only Western country struggling to manage increased diversity. For its 2018 exploratory study *The World in a City* (*De wereld in één stad*), the WRR collected lessons from research into the experiences of various European cities in dealing with this phenomenon. Their social problems are sometimes of a different order, and the administrative scale also varies, but it is clear nonetheless that cities across the continent face much the same problems and that there are no simple solutions. Everywhere there is a pattern of trial and error, and everywhere we see attempts to promote social contact between residents through small-scale local initiatives. The Netherlands is no exception in this respect. Such initiatives seem to be most successful when they are rooted in common goals or interests. It is worth noting, though, that there has been little in the way of reliable evaluation research into them.8 Almost everywhere, at home and abroad, good impact studies and policy evaluations are lacking. This makes it very diffcult to separate the wheat from the chaff.

For this report we have looked mainly at cohesion policy in the United Kingdom. After riots between various migrant groups in a number of cities in 2001, the British government set up a Community Cohesion Review Team (CCRT) to talk with local residents, policymakers and other leaders and to collect good practices*.* The resulting report centred on the key concept of 'community cohesion' as an alternative to the multiculturalism model, which was revealed as having led to groups with different cultural backgrounds leading separate, parallel lives.9

The CCRT report prompted the development of community cohesion programmes, which sought to encourage interaction and mutual understanding between groups with different origins. The term 'intercultural interaction' soon came to be used by some advocates of the new cohesion-based policy.10 Following the report's publication, community cohesion became an important aim of British national and, especially, local policy.11 This approach is based largely upon Gordon Allport's contact theory,12 which states that, subject to certain conditions, direct contact and interaction between people with different backgrounds and lifestyles can increase tolerance and reduce prejudice.13 Through a whole raft of initiatives, policy in the

<sup>8</sup>See Gijsberts and Dagevos (2007).

<sup>9</sup>Cantle (2001, 2013: 10).

<sup>10</sup>Cantle (2013: 10), Dobbernack (2014: 130), and Lewis and Craig (2014: 23).

<sup>11</sup>For example, as recently as 2019 the Local Government Association (LGA) published guidelines for cohesion policy entitled *Building Cohesive Communities.*

<sup>12</sup>Allport (1954).

<sup>13</sup>This theory has since been supported by several empirical studies. For instance, psychologists Pettigrew and Tropp conducted a meta-analysis of more than 500 studies. This confrms that contact between different ethnic groups, for example, reduces prejudice. See: Pettigrew and Tropp (2006: 766) and Cantle (2013: 10).

UK14 over the past two decades has therefore sought to improve relations between groups.15 But what has it delivered? From the few, often unsystematic evaluations available, three things emerge.

First, facilitating intercultural contacts at local level appears to have had some favourable effect in increasing mutual understanding and weakening stereotypes. At the same time, it has proven diffcult to fnd common solutions to shared problems and many groups have shown little interest in co-operating further. Whilst a local intercultural approach can have some positive impact, then, expectations should not be set too high.

Secondly, it seems that a national discourse stigmatizing and excluding certain minorities – such as Muslims or central and eastern Europeans in the case of the UK – undermines an effective local approach. That discourse has exacerbated tensions between groups at the local level and so impeded intercultural interaction. For a pragmatic local approach to be effective, it helps if central government encourages a discourse on 'national identity' that is open and inclusive.

Thirdly, the research suggests that the promotion of contacts at the local level should go hand in hand with action to tackle deep-rooted inequalities, such as economic disparity and discrimination. A pragmatic approach aimed at forging contacts 'on the ground' is not a panacea but must be part of a broader policy. Isolated initiatives with no overarching plan are of little avail. This fnding is echoed in research into the social integration of minorities in Canada,16 which also revealed that they experience wide-ranging discrimination. Moreover, it was found that the social integration17 of immigrants with a non-European background is slower than for those with a European background.18

According to the researchers, one major shortcoming of the Canadian policy is that although many broad goals and ideals have been formulated, such as 'equal opportunities', few frm and explicit targets have been set. Also, Canada's policy of multiculturalism has been insuffciently evaluated. In this respect it resembles British cohesion policy: there are plenty of noble objectives, but these have not been built upon in a concrete and consistent manner. Nor has the policy's impact been assessed.19

These fndings underline the importance of an integrated cohesion policy which seeks not only to improve intercultural relations but also to address discrimination and economic disparities and to provide a policy idiom that works for all migrant

<sup>14</sup>Although this has not been and is not applied uniformly across the country. See Lewis and Craig (2014: 24).

<sup>15</sup>Local Government Association (2019: 10, 19–20, 30) and Ipsos MORI (2007: 5–9).

<sup>16</sup>Reitz and Banerjee (2007).

<sup>17</sup>The term 'social integration' as used in this study means "the extent to which individual members of a group form relationships with people outside the group – relationships that help them to achieve individual economic, social or cultural goals". This is measured using of a number of indicators, such as self-identifcation as Canadian and feelings of belonging; see Reitz and Banerjee (2007: 18).

<sup>18</sup>Reitz and Banerjee (2007: 38).

<sup>19</sup>Reitz and Banerjee (2007: 35–36, 39).

groups, not just some of them.20 It also shows that far more attention needs to be paid to systematic evaluation based upon clearly defned objectives.

### *6.2.2 Promoting Public Familiarity in the Neighbourhood*

Social interaction between neighbours is in part a product of the way their shared living environment is confgured. Intuitively, we often think that it is necessary to forge close ties between them in order to achieve social cohesion. The academic literature, however, indicates that interactions at this level are in fact so-called 'weak relationships', and that this also has advantages. Not everyone has a need to maintain intensive contacts with the people living around them.21 Moreover, encouraging strong feelings of belonging can lead to some residents appropriating public space. For others, though, that can undermine their sense of familiarity and safety, intensify conficts and even cause some to withdraw from that space.22 Communities that are too tight-knit can exclude 'outsiders', even if they are neighbours.23 Bonding – forging close contacts between people from the same social group – may thus form an obstacle to bridging, building contacts between different groups.

Sociologists like Talja Blokland and Jan Willem Duyvendak have therefore argued that governments should focus upon promoting what they call 'public familiarity'. By this they mean that residents of a neighbourhood recognize each other in public space, even if they have no personal contact.24 This concept assumes that people who cross paths on a regular basis develop a form of 'passing acquaintance', even though they are otherwise complete strangers who have nothing to do with each other.25

Public familiarity enhances the experience of social safety. It provides people with a framework to 'place' themselves and others in a social context,26 which in turn makes it easier for them to call each other to account – in the event of disruptive behaviour, for instance.27 People also feel more at home in a neighbourhood when they have a reasonable idea of the social codes and unwritten rules applicable in its public domain.28 Finally, public familiarity is a precondition for the emergence of mixed social networks: stable, diverse neighbourhoods with a certain degree of public familiarity provide a good basis for people to connect in a positive manner.29

<sup>20</sup>See also Kremer et al. (2014).

<sup>21</sup> van Eijk (2010).

<sup>22</sup>Duyvendak (2017).

<sup>23</sup>Kleinhans and Bolt (2010: 78) and WRR (2005: 93).

<sup>24</sup>Blokland (2009).

<sup>25</sup>Blokland (2009: 182).

<sup>26</sup> Jacobs (1961).

<sup>27</sup>Hoekstra (2019), Blokland and Nast (2014), and Kleinhans and Bolt (2010).

<sup>28</sup>Blokland and Nast (2014).

<sup>29</sup>RMO (2005).

Local authorities can promote public familiarity in various ways. In the frst place, through the confguration of the physical environment.30 Safe, clean public space is essential for successful public familiarity and for social cohesion in general. In many deprived neighbourhoods, litter and derelict buildings are a major source of irritation.31 People are more likely to engage in antisocial public behaviour if they see others already doing so by, for example, littering.32 As certain physical signs of neglect, such as broken windows or graffti, become more common, so do behavioural transgressions like fy-tipping and petty theft.33 Visible signs of decay thus lead to a weakening of social norms, which in turn spreads delinquency. This makes it important that the authorities intervene early when deterioration occurs.

### **Box 6.1: Rotterdam: People Make the City**

People Make the City ('Mensen Maken de Stad', MMS) is a municipal project to improve the sense of community on residential streets in Rotterdam. It backs grassroots initiatives in streets where residents have little contact with each other. MMS has adopted an assertive approach, in the belief that people want a safe, clean and pleasant living environment but often have no idea how to go about creating it. Social professionals go door-to-door to ask about the street's particular problems and fnd out who might be willing to help solve them. Then, together with a group of active residents, they draw up a so-called 'street agenda' and work with council services and housing corporations to agree a set of rules for its implementation. All concerned commit themselves to goals such as organizing an event or renovating a playground. The agenda is offcially adopted when at least a third of households in the street sign up to it. Once implementation begins, the professionals increasingly step back and let the residents take charge.

Justus Uitermark and Jan Willem Duyvendak have studied the impact of MMS in 40 Rotterdam streets34 and found that it succeeds in overcoming residents' reluctance to take responsibility for the living environment beyond their own front doors, increases their mutual trust and reduces feelings of insecurity. Neighbours become more likely to speak to each other and to engage with the community. Trust in professionals and organizations also increases. And in streets where the active resident 'teams' are very mixed, there is less thinking in terms of groups and more recognition of people as individuals.

<sup>30</sup> van Gemerden and Staats (2006).

<sup>31</sup>WRR (2005: 57).

<sup>32</sup>There is some debate about this 'broken windows' theory, but the main conclusions still stand. See, for example: Keizer et al. (2008) and Welsh et al. (2015). Also: Blokland (2009), Kleinhans and Bolt (2010: 114, 2014), and Uitermark and Duyvendak (2006a), Ross et al. (2001).

<sup>33</sup>Keizer et al. (2008: 1681–1685). See also Welsh et al. (2015).

<sup>34</sup>Uitermark and Duyvendak (2006a, b).

Once the basic condition of a safe and clean environment is met, the local authority can promote public familiarity by encouraging people to meet regularly in the public space. Recognition from earlier encounters, however superfcial, can contribute towards feeling at home, safe and connected. If its confguration invites people to linger in public space, there is a greater chance that residents will meet each other. Well laid-out and maintained parks, public gardens and playgrounds play an important role in this respect.35

An ethnographic study of parks close to highly diverse urban neighbourhoods in the UK shows that they provide the setting for contacts between a wide variety of groups, and that people thus become accustomed to each other there.36 The encounters might only be perfunctory, in playgrounds, at picnic sites and on walks, but it can also be more intensive. For example, at boot camps, on playing felds and at festivals. Pleasingly designed and recognizable parks also contribute directly to feelings of being at home because residents develop a bond with them.

Dutch research reiterates the importance of the informal appropriation of open spaces, playgrounds and public gardens.37 When residents gain a sense of coownership of their immediate living environment, informal social control increases, as do feelings of belonging and connection with the neighbourhood. This sense of ownership can be stimulated by, for example, allowing residents to participate in the maintenance of green space. Previous interventions have demonstrated that community gardens attract a broad range of residents, both with and without a migrant background.38 In neighbourhoods with a high degree of diversity and little social capital, however, it is often necessary for the local authority or some other institution to take the frst step in facilitating contact between residents.39 It is also important that community projects (see Box 6.1) be supported by migrant and non-migrant groups alike.

### *6.2.3 The Importance of Good Semi-public Facilities*

There is a considerable body of research indicating that public and semi-public facilities reinforce the social resilience of a neighbourhood.40 Where there is a good range of amenities like schools, playing felds, playgrounds, shops and community

<sup>35</sup>WRR (2005).

<sup>36</sup>Neal et al. (2015).

<sup>37</sup>Kleinhans and Bolt (2010: 117–118).

<sup>38</sup>Gijsberts and Dagevos (2007: 27).

<sup>39</sup>Gijsberts and Dagevos (2007).

<sup>40</sup>WRR (2005), Sanders and Dautzenberg (2010), Völker et al. (2007), Blokland (2009), and Klinenberg (2018).

centres, there is a greater sense of community and there are more social contacts between residents.41 A good sociophysical infrastructure also contributes towards the emergence of productive grassroots initiatives.42 In this context, Joke van der Zwaard and Maurice Specht43 refer to places where "public familiarity can arise so that mutual prejudices are broken down and people know what they can expect from each other".44 Ethnographic research in south Rotterdam, for example, suggests that public libraries are a forum for everyday encounters that help people feel more at home in their very diverse neighbourhood (see Box 6.2).45 Likewise, international studies highlight the great importance of local libraries for cursory forms of social bonding.46

### **Box 6.2: Libraries: Silent Places of Meeting and Bonding**

Between six and seven million people in the Netherlands visit a library on a regular basis. There are 770 branches in total and the average person lives 1.9 km from the nearest one. In 2017 they lent out a total of 67.3 million physical books. But libraries today perform many more functions. For example, the director of Eindhoven's public libraries, Albert Kivits, states that they play an important role in connecting the city's international community with the local population.

The modern library is a meeting place for a highly diverse group of users. In Eindhoven, for instance, so-called 'language cafés' are held for anyone wanting to practise their Dutch in an informal setting. Expats can also fnd help to understand offcial documents such as letters from the tax authorities. And many also bring their children to Dutch reading sessions or take part in the library's live 'travel book' project, at which international residents talk about their country of origin, language and culture. As well as catering for Eindhoven's large expat community, the library is a 'living room' for other groups. Schoolchildren from cramped homes do their homework there. And people in need are welcome, too: the libraries operate a so-called 'suspended coffee' scheme, where visitors can buy a drink or even a meal in advance for someone unable to afford it themselves.47

<sup>41</sup>Völker et al. (2007). See also Kleinhans and Bolt (2010).

<sup>42</sup> van der Zwaard and Specht (2013).

<sup>43</sup> van der Zwaard and Specht (2013: 5).

<sup>44</sup> van der Zwaard and Specht later add (2013: 47), "Simply closing facilities because that is the cheapest solution comes at the expense of public familiarity and the self-esteem of users/residents, and thus their collective self-reliance."

<sup>45</sup>Peterson (2017).

<sup>46</sup>See Klinenberg (2018).

<sup>47</sup> https://innovationorigins.com/en/library-helps-expats-integrate-eindhoven

Local community centres can also contribute towards public familiarity, with the potential to forge deeper and more lasting contacts. But to achieve this it is important that they organize activities open to all and not limited to one group, otherwise there is a risk that they actually intensify segregation, competition for hegemony and feelings of exclusion.48 These centres do not need to be expensive purpose-built buildings: the literature on self-organization and grassroots initiatives contains many successful examples of residents converting vacant property – disused schools, historic buildings, churches, shops, business premises or offces – for community use.49

Housing corporations, too, have a role to play in enhancing social cohesion. In the Netherlands, social housing traditionally has been provided by large, publicly funded housing corporations. These semi-public bodies have traditionally taken on civic responsibilities above and beyond their core task of building and managing social rented housing, such as enhancing the general liveability of neighbourhoods in which they are active.50 Following amendments to the Housing Act (Woningwet) in 2015, however, their activities in this domain were restricted to housing-related social work, the development of small-scale infrastructure around their own properties and the promotion of a clean, safe and nuisance-free living environment. Since these changes, their spending on liveability has fallen well below the statutory maximum. Almost four out of ten local authorities fnd that the corporations are now doing less in this feld than they used to, and indeed less than is necessary.51 In 2019 the Minister of the Interior announced her intention to relax the rules to allow housing corporations to conduct activities on a small scale to promote encounters between residents.52

Housing corporations can reach agreements with residents to promote social safety in their living environment; for example, in the form of codes of conduct or 'city etiquette' rules. These give everyone, whatever their background, a better idea of what is expected of them and what they can expect from others – something often much needed in socially diverse neighbourhoods.53 They also encourage public familiarity.54 When undertaking initiatives of this kind, though, it is important that the residents involved receive institutional backing or professional support.

Social initiatives by housing corporations, community centres, libraries and so on have no chance of success without proper funding. The Council for the Environment and Infrastructure (Raad voor de Leefomgeving en infrastructuur, Rli) therefore advises that all towns and cities draw up an investment strategy for provision of this kind.55

<sup>48</sup>See, for example, Hoekstra and Pinkster (2017).

<sup>49</sup>See, for example, van der Zwaard et al. (2018).

<sup>50</sup>Engbersen and Engbersen (2014).

<sup>51</sup>Beuzenberg et al. (2017).

<sup>52</sup>Ministerie van Binnenlandse Zaken en Koninkrijksrelaties (2019: 14).

<sup>53</sup>Kleinhans and Bolt (2010: 15).

<sup>54</sup>Kleinhans and Van Marissing (2008).

<sup>55</sup>Rli (2020: 35).

### *6.2.4 The Importance of New-Style Community Work*

A public infrastructure is important to achieve public familiarity and social cohesion, but this alone is not enough. Dedicated professionals are needed to ensure that people use the infrastructure or contribute towards its creation. In the past, this task was assigned to community or neighbourhood workers. After the abolition of the government support scheme for community work and the decentralization of welfare responsibilities to local authorities, however, the number of professionals active in this domain fell sharply. In part, their place has been taken by specialists in specifc policy felds such as child protection, youth work, home care, disability care, nursing care and mental healthcare. In recent years, though, the number of dedicated community workers has begun rising again – often under new titles such as 'quartermaster', 'neighbourhood co-ordinator', 'social co-ordinator', 'district coach' or 'area broker'.56 There has also been an increase in the number of social enterprises developing cohesion-related initiatives of various kinds.57 But many of these come from outside the area, are contracted only temporarily or depend upon project subsidies related to new policy priorities, such as combating loneliness.58 Such arrangements are generally unsatisfactory, because the problems in vulnerable neighbourhoods are so deep-rooted that they demand a systematic and long-term approach, such as that provided by the National Programme for South Rotterdam (Nationaal Programma Rotterdam Zuid).59

For this category of professionals, it is particularly important to possess a thorough knowledge of a neighbourhood's social fabric, demographic make-up and population turnover. Their role encompasses providing residents with guidance and support, encouraging the various origin groups to make use of the established sociophysical infrastructure and galvanizing them to develop community initiatives of their own.60 According to social research institute Movisie, community-building requires active input from professionals to make full use of the resources available to forge contacts between residents. Diffdence and embarrassment often stand in the way of spontaneous encounters, especially in a 'superdiverse' community.61

<sup>56</sup>Engberts and Bijl (2020).

<sup>57</sup>According to the defnition used by the Social and Economic Council of the Netherlands (SER, 2015), social enterprises are "independent enterprises which supply a product or service and which primarily and explicitly pursue a social purpose; that is, seek to resolve a social problem."

<sup>58</sup>See also Bokhorst (2020).

<sup>59</sup>See also Jennissen et al. (2018: 161).

<sup>60</sup>WRR (2005), Kleinhans and Bolt (2010), and van der Zwaard and Specht (2013).

<sup>61</sup>Engberts and Bijl (2020).

### *6.2.5 Strengthening Local Identities*

Finally, to strengthen solidarity within communities or regions, it may make sense to focus upon reinforcing local identities. A study by I&O Research shows that many people in the Netherlands feel a strong connection with their place of residence, province or region.62 Residents of Friesland, Groningen and Limburg identify more strongly with their province than with the Netherlands as a whole, whilst those living in Zuid-Holland, Noord-Holland and Utrecht tend to have a close attachment to their own town or city. According to the researchers, this is due primarily to the allure of the major cities in these provinces.

Several studies indicate that young people with a migrant background living in Rotterdam63 and Amsterdam64 consider themselves 'Rotterdammers' or 'Amsterdammers' frst and Dutch only second. Respondents state that the intercultural character of these cities gives them a sense of belonging there, far more than 'in the Netherlands'. *Being* Dutch, they say, is about being born and raised here, whilst *feeling* Dutch comes from acceptance by the majority – and in that, outward appearance often plays a role.65

Attitudes of this kind are forging new forms of 'citizenship' and self-identifcation in the big cities, especially.66 'Urban citizenship' is undemanding and easily accessible; in principle, it is open to anyone who settles in a given city. And crucially, such an identity related to a particular place is not incompatible with others – one rooted in a country of origin, for instance. Many highly skilled professionals and EU labour migrants have no intention of settling permanently or becoming Dutch nationals, and sooner or later they either return home or move on. So pursuing legal citizenship is of no interest to them. Urban citizenship, on the other hand, can easily be combined with a foreign nationality or with continuing ties to the country of origin.

There are also downsides to the promotion of local identity, of course, especially if that is overly homogeneous in its exposition. After all, the natural human desire to develop a social identity almost inevitably leads to categorization, determination and group comparisons. The ingroup-outgroup distinction and the resulting bias in favour of the ingroup is a classic theme of experimental sociopsychological studies.67 An excessive sense of 'us-and-them' can amplify dichotomies between regional or local identities: Limburgers versus Hollanders, for instance, or Amsterdammers versus Rotterdammers. Although this may strengthen social

<sup>62</sup>See I&O Research (2019).

<sup>63</sup> van Bochove et al. (2009).

<sup>64</sup> van der Welle and Mamadouh (2009) and van der Welle (2011).

<sup>65</sup> van der Welle (2011).

<sup>66</sup>Hirsch Ballin (2014).

<sup>67</sup>Tajfel (1981).

cohesion at a local level, it undermines it on the broader national stage. It is therefore important not to overemphasize place-based identity in public policy.

From the perspective of the new diversity, then, any policy involving this form of identity needs to satisfy a couple of criteria. First, the identity concerned should be strong and distinctive enough to give residents a sense of attachment to it. This can be done by utilizing widely recognized physical identifers of the place in question, such as a unique skyline, famous buildings, bridges or other landmarks.68 Or by highlighting its cultural symbols, in the broadest sense of the term: specifc festivals and events (historical commemorations, carnival, an annual parade), the local dialect, 'local heroes', sporting events (the city marathon) or clubs and so on.

At the same time, a place-based identity must be general enough to encompass everyone living there. Even short-term residents should be able to identify with it reasonably quickly. This means that the symbols and identifers used should be as inclusive and accessible as possible, even for people who have not lived in the city or region all their lives. It also helps to give residents co-ownership and control; for example, by letting them organize local festivals.

### *6.2.6 Summary*


<sup>113</sup>

<sup>68</sup>Engbersen and Engbersen (2008).

### **6.3 Social Cohesion at Schools and Voluntary Associations**

### *6.3.1 Segregation Undermines Cohesion Through Education*

In so-called 'high-contact' settings in the semi-public service sector, the increasing diversity by origin and length of stay of the migrant population is very visible. This demands large numbers of professionals in education, care and the social domain. Education, in particular, is a sector in which people from different backgrounds come into intensive and lasting contact with each other. It is also a setting in which approaches to cope with diversity are developed actively.

Schools are both *sources* of cohesion and *places* of cohesion. To start with the former, they are a perfect example of institutions where people come into direct contact with members of other groups and so must learn to deal with the new diversity. The Dutch government considers it vital that education build social bonds: "Schools, as important meeting places for young people, play a prominent role in the development of active citizenship and of the knowledge and skills pupils require for that."69

There are indications, however, that the role of education in bringing together different groups is declining. The Netherlands already has highly segregated social networks, and such segregation is actually increasing in the areas of housing and education.70 The neighbourhood in which you live very much determines what school your children attend.71 Hence the growing social, economic and ethnic segregation in schooling observed by the Inspectorate of Education (Inspectie van het Onderwijs).72

The relatively early age at which children in Dutch secondary education are streamed by ability, about 12 years old, does little to help them learn to deal with diversity. The gap between ability groups in the feld of civic engagement is wider in the Netherlands than in countries with a common curriculum in secondary education.73 Our schools, according to Herman van der Werfhorst, are homogeneous islands when it comes to views regarding citizenship. "It is more diffcult in the Dutch system than elsewhere to come into contact with people who hold different opinions," he writes.74 Only to a limited extent, then, does education in the Netherlands ensure that children of different origins meet and mix, thus strengthening social cohesion.

<sup>69</sup>*Voorstel van Wet tot wijzing van een aantal onderwijswetten in verband met verduidelijking van de burgerschapsopdracht aan scholen in het funderend onderwijs* ('Bill amending a number of education acts to clarify the citizenship remit of schools in primary and secondary education'), ID8437.K-1: 4.

<sup>70</sup>Bovens et al. (2014: 24).

<sup>71</sup> Inspectie van het Onderwijs (2019).

<sup>72</sup> Inspectie van het Onderwijs (2018: 5).

<sup>73</sup> van de Werfhorst (2017).

<sup>74</sup> van de Werfhorst (2019). Based upon Munniksma et al. (2017).

### *6.3.2 Diversity and Belonging*

The segregation we fnd within the Dutch education system as a whole does not alter the fact that there are many schools with a high degree of diversity in their classrooms. So how can they best deal with this phenomenon? This question is really all about whether schools are *places* of cohesion, a topic on which there exists a substantial body of research.75

We look frst at how highly diverse schools can promote a sense of 'belonging' amongst their pupils. An international group of researchers led by Laura Celeste has studied the effects in this respect of different approaches to diversity at secondary schools in Flanders. They distinguish four such approaches, which closely resemble those we fnd in Dutch local policy.76 The frst is *colour-blindness*, in which religious and cultural neutrality is paramount and schools pay no attention to the origins of their pupils. This is the stance most commonly adopted at Flemish schools. Secondly, *assimilationism* requires everyone to adapt to the majority culture and there is no room for minority languages or cultural expressions. A third approach is *equality*, in which equal treatment and countering discrimination take precedence. The fourth and fnal option is *multiculturalism*, in which pupils learn about each other's cultural heritage and there is room for mutual cultural understanding.77 This, incidentally, is not the same as the much more far-reaching form of multiculturalism that was in vogue in the Netherlands in the late twentieth century, which emphasized education in one's own language and culture and self-organization along ethnic lines. To avoid confusion in this regard, we refer to the 'multiculturalism' described by Celeste et al. as *interculturalism* (see Box 6.3).

### **Box 6.3: Interculturalism or Multiculturalism?**

In the academic literature on educational theory and psychology, the terms 'multiculturalism' and 'interculturalism' are both used to describe a pedagogical approach that allows room for the recognition of cultural diversity. The label 'multiculturalism' is most common in the United States, whereas in Europe 'interculturalism' has been used more widely since about 2000.78 In the US, moreover, the multicultural approach has been translated into set of specifc and concrete measures, such as coverage of different cultures in the curriculum, countering prejudice and discrimination, actively promoting equality and the empowerment of students from minority groups. The

(continued)

<sup>75</sup>We are deeply indebted to Batja Gomes de Mesquita, Professor of Social and Cultural Psychology at Louvain (Belgium), for her assistance in accessing this feld of research.

<sup>76</sup>Celeste et al. (2019).

<sup>77</sup>See also: Verkuyten and Thijs (2013), Apfelbaum et al. (2016), Civitillo et al. (2017), and Schachner et al. (2016).

<sup>78</sup>See also Verkuyten and Thijs (2013: 180).

### **Box 6.3** (continued)

Netherlands adopted its own interpretation of multiculturalism in the latter half of the twentieth century, one strongly coloured by the national tradition of politicoreligious 'pillarization' (see 4.1). That resulted in a strong emphasis upon educating migrants in their own language and culture, along with selforganization along ethnic lines. This approach has since become discredited, however, and has been replaced by a more generic focus upon citizenship skills.

To avoid confusion with this earlier, specifcally Dutch, form of 'multiculturalism' in education, we refer to its more recent incarnation as *interculturalism*. By this we mean an educational approach in which pupils learn about each other's cultural heritage, there is scope for mutual cultural understanding and particular attention is paid to countering discrimination.

For children with a migrant background attending the Flemish schools, such an intercultural approach proved the most effective by far. They felt as great a sense of 'belonging' at school as their classmates with a non-migrant background. Their academic achievement – measured using their Dutch-language grades – was also better than under the other approaches. It was striking, too, that this approach did not compromise the performance of the children of Flemish origin. And interculturalism was found to work especially well at 'superdiverse' schools, where more than 60% of pupils have a migrant background.79 In schools adopting a colour-blind approach, by contrast, children with a migrant background performed signifcantly worse. And the stronger the colour-blindness, the lower their grades. Assimilationism also produced mainly negative results. It did not lead to better grades in Dutch, and feelings of belonging at school were far weaker. Indeed, the stronger this approach, the weaker they were. Finally, the researchers found that the equality approach had no signifcant effect.

Other international studies have generated similar results.80 School curricula that pay some attention to other cultures make pupils from them feel acknowledged and involved, because it means that their particular knowledge in certain areas counts for something.81 Conversely, children who are not allowed to speak their own language in the school playground – as is often the case in Flanders – take that as a signal that their mother tongue and their identity in general are considered worthless. In the labour market, too, research shows that minority employees feel more at home in companies where management considers space for other cultures important. When leaders – in business or in education – acknowledge and value cultural

<sup>79</sup>Celeste et al. (2019: 12).

<sup>80</sup>See, for example: Baysu et al. (2016), Stephens et al. (2019), and Walton and Cohen (2007). 81This also has positive effects upon their results at school and reduces drop-out rates. See Cabrera

et al. (2014).

differences, minorities have a greater sense of belonging than when they adopt a colour-blind approach. Moreover, colour-blindness is more frequently associated with confict and distance.82 One fnal but also crucial fnding in this regard is that interculturalism must be fully inclusive, in the sense that the culture of the majority group is recognized and appreciated as well,83 otherwise there is a risk that it in turn will feel unvalued or even threatened.84

### *6.3.3 Diversity and Interethnic Relations in Schools*

One upshot of greater diversity in the classroom is that teachers increasingly face the challenge of channelling social tensions within the school.85 Whilst a lot of research has been conducted into the effects of diversity upon classroom relationships,86 the fndings are far from unambiguous. Some studies indicate that the risk of bullying is lower when a school or a class is made up of many different ethnic groups of roughly equal size,87 whereas others fnd no correlation between the ethnic composition of a school or classroom and the quality of its interethnic relations.88 Dutch research shows that pupils in secondary education have a strong preference for interaction with peers from the same ethnic group. The greater the ethnic diversity of a school, the greater the chance that bullying occurs89 – both interethnic and intra-ethnic.90 About 45% of bullying in a class was found to be related to its composition. The more diverse the class, the more bullying there was. Because teachers in mixed classes may have more diffculty managing social relationships between pupils, the researchers believe that teacher-training programmes should pay more attention to the handling of group dynamics.

<sup>86</sup> In particular, we rely upon the literature review by Thijs and Verkuyten (2014).

<sup>82</sup>Meeussen et al. (2014).

<sup>83</sup>Meeussen et al. (2014: 638).

<sup>84</sup>See also Morrisson et al. (2010).

<sup>85</sup>At schools with large numbers of Muslim pupils, for example, there is sometimes consideration tension around the acceptance of homosexuality. See Leon Meijs' position paper (Meijs, 2020) for the Parliamentary Commission of Inquiry into the Undesirable Infuence of Unfree Countries (Parlementaire ondervragingscommissie ongewenste beïnvloeding uit onvrije landen, POCOB).

<sup>87</sup>Pupils in classes where the ethnic groups are more equally divided indicate that they feel safer, less lonely and less bullied. Munniksma (2013: 14) writes, "When more groups were more equal in size, as measured by the Simpson (1949) diversity index, students indicated to feel safer, to feel less lonely and to be less peer victimized. Juvonen et al. (2006) argued that when groups were more equal in size, there would be a balance in power that would be related to a stronger sense of safety and social satisfaction at school among ethnic minority groups. Stark (2011) argued that in order to examine the effect of classroom ethnic composition on outgroup attitudes, interpersonal relations within this context need to be taken into account. Also, when there are more groups, and when those groups are more equal in size, the chances of cross-ethnic friendships are more likely." 88Bekhuis et al. (2013), Stark (2011), and Vermeij et al. (2009).

<sup>89</sup>See: Fortuin et al. (2014), Smith et al. (2014), Stark and Flache (2012), and Hooisma (2020: 133). 90Tolsma et al. (2013).

A review paper by Jochem Thijs and Maykel Verkuyten91 summarizes research perspectives of relationships between the ethnic composition of schools or classes and interethnic relations within them as follows: ethnically diverse schools produce more interethnic friendships in absolute terms, but this average effect is not very strong and may also be negative under certain circumstances. Thijs and Verkuyten thus conclude that ethnic diversity *in itself* does not guarantee interethnic tolerance. They point out the importance of other aspects of the school environment, such as emphasizing a common, inclusive institutional identity. It also helps if the teaching imparts knowledge about cultural differences and if standards are set which promote interethnic tolerance, such as a strong antidiscrimination policy. Drawing upon several empirical studies conducted at Dutch primary schools, the authors conclude that an intercultural educational approach can lead to more positive interethnic relations.92 Moreover, this effect is discernible amongst the majority with a Dutch background as well as the minority groups in question; the underlying mechanisms explaining it are greater knowledge of cultural diversity and the dissemination of antidiscrimination norms.

### *6.3.4 Policy Signifcance: Scope for Cultural Familiarity and Greater Professionalism*

Based upon the scientifc literature, an intercultural approach seems to work best at schools with a diverse pupil population. When children feel seen and recognized at school, they have a greater sense of belonging there and perform better. This means that schools need to create scope for a certain degree of *cultural familiarity*. Within the regular curriculum, for example, subjects such as history and geography could address pupils' various backgrounds and origins as a matter of course. In this way, education can play a role in improving knowledge of different migrant groups, their reasons for migrating, their cultural backgrounds and the possible diversity within groups from the same country of origin. At the same time, an intercultural approach also addresses the cultural backgrounds of the more long-established groups in society.

Schools need to prepare themselves for a pupil population that is evolving constantly in terms of its cultural diversity, and they have to be able to absorb new groups smoothly. When the Netherlands had only a few migrant communities, specifc expertise and dedicated policies could be developed for each of them. With dozens, from all parts of the world, that is impossible. What worked in the past for pupils with a Moroccan or Turkish background will not necessarily be relevant for the new infux of Eritrean, Polish or Indian schoolchildren.

This means that schools and teachers alike need far more knowledge and skills about how to cope with pupils of different origins and with highly diverse classes.

<sup>91</sup>Thijs and Verkuyten (2014).

<sup>92</sup>Verkuyten and Thijs (2013).

Schools, for example, need to set clear standards to increase interethnic tolerance and to counter discrimination and racism. Through peer-to-peer learning, teachers can help each other fnd ways to raise sensitive issues and to act decisively. Intercultural knowledge and skills should be an integral part of their professional portfolio, so acquiring them should be a standard aspect of teacher training – as has long been in the case in traditional nations of immigration such as the US, New Zealand and Canada. At present, teachers in the Netherlands receive far less instruction in dealing with cultural diversity than their counterparts in many other countries.93

Nor is there yet any Dutch teacher-training programme or Master's degree with a specifc focus upon intercultural education. In this regard we can learn from the International Baccalaureate scheme overseen by global organization IBO. As part of this, 19 international secondary schools in the Netherlands have developed an intercultural learning profle with an emphasis upon curiosity, inquisitiveness, thinking skills, communication, integrity, openness, caring, entrepreneurship, balance and refective ability. To provide something similar across the Dutch education system, all existing teachers and future trainees would have to undergo a professional development trajectory (see Box 6.4).94

### **Box 6.4: Interculturalism in Education**

Many teachers in the Netherlands already encounter cultural diversity in their classrooms or will do in the future. It is therefore important that they be able to deliver inclusive and intercultural education. Sabine Severiens and her colleagues argue that teacher-training courses should devote more attention to the knowledge and skills important to be able to do this. In their view, that means a particular focus upon the following points.95


<sup>93</sup> Inspectie van het Onderwijs (2019: 31–34).

<sup>94</sup>Leeuwestein and Bokhorst (2018: 30, 36).

<sup>95</sup>Severiens et al. (2014).

### *6.3.5 Diversity and Transience Also Put Pressure on Voluntary Associations*

In many Dutch communities, voluntary associations are cornerstones of social cohesion. Sports clubs, choirs, carnival associations and other such groups are a means of making contacts and giving residents a sense of belonging. Many local authorities therefore put a lot of effort into supporting associations in pursuit of a variety of underlying policy goals. Sports clubs, for instance, are encouraged to broaden their scope by providing training sessions for people with disabilities, refugees and psychiatric patients, or by opening their clubhouses and changing rooms to after-school clubs and other community groups. Indeed, in many places they are expected to play a leading role in the integration of migrants and their children.96

There are indications, however, that the increased diversity and transience of the migrant population are straining many local associations. Not much research has been conducted into the effects of short-stay migration upon organizations of this kind, but it seems highly likely that a high turnover of members puts their continuity at risk. In the Netherlands, such groups are almost always run by volunteers working for free in their spare time. The shorter people stay, the harder it becomes to bond them with the group and recruit them to positions of responsibility.97

More studies have been undertaken into the ways diversity affects voluntary associations. These reveal clear differences between people with and without a migrant background when it comes to participation in such groups and volunteering in general. For example, people with a Dutch background are more likely to be members of a sports club. According to research by the Mulier Institute, 14% of Dutch residents with a non-European/Anglosphere background belong to a sports club, compared with 19% of those with a European/Anglosphere migrant background and 25% of those with a Dutch background.98 Likewise, the SCP reports that someone with a European/Anglosphere background is far more likely to volunteer at a cultural institution than a person with a different migrant background.99 Moreover, people not or only partially raised in the Netherlands feel far less attracted to organized group activities here.100

Public opinion tends towards a cultural explanation of these dichotomies: many people with a migrant background come from cultures with no tradition of community organization or volunteering. The research, however, shows that they do in fact join sports clubs and put themselves forward as volunteers just as often as anyone else; their lower overall rate of participation is actually due to the fact that, on

<sup>96</sup>Krouwel et al. (2006), Vermeulen and Verweel (2009), and Janssens and Verweel (2014).

<sup>97</sup>One exception to this is student associations. Special provision for them includes management grants to fund the appointment of full-time sabbatical offcers.

<sup>98</sup>Brandsema et al. (2018: 212). See also van Haaften (2019).

<sup>99</sup>The aggregate indices relative to the Dutch population (2012–2016) are 105 and 63 respectively. SCP (2019: 214).

<sup>100</sup>Kullberg et al. (2019: 21).

average, they cancel their membership more quickly.101 And this is especially so when a high proportion of the other members have a different background.102 Indeed, the same applies even more to members with a Dutch background: at the few clubs where they are in the minority, their departure rate is even higher than that of members with a migrant background who are also in a minority.

This suggests that the lower rate of participation by people with a migrant background is caused not by any cultural factor, but rather is a direct effect of diversity. Or, to put it in our terms: it is not a compositional effect attributable to the characteristics of a specifc group but a diversity effect common to all groups, including those with a Dutch background. The primary underlying mechanism, it would seem, is homophily: the principle that people have a strong need to associate with similar others. If a voluntary association like a local sports or hobby club is very diverse, a large of proportion of its members will lack a sense of belonging and so be inclined to leave.103

This applies equally to members with a Dutch background, although such feelings are still relatively rare for them because they form a signifcant majority in most associations. Certainly when it comes to sports clubs. The majority of amateur football clubs in the Netherlands, for instance, are made up predominantly of members with a Dutch background, who feel at home in the company of people 'like themselves' and renew their membership year after year. Members with a migrant background are almost always in a minority and therefore feel less connected with the group.104 All this means that, as clubs become more diverse, they have a higher rate of 'churn' and run a greater risk of falling apart. Or, alternatively, that increased diversity is only a temporary phenomenon and they again become more homogeneous over time. In the latter case, their stability increases as the diversity dissipates, because everyone is once again 'the same'.105

In other words, diversity makes life more complicated for voluntary associations. Highly diverse clubs and societies must make more of an effort to keep themselves together than homogeneous ones. No studies have yet been conducted into effective strategies to achieve this, but it seems likely that the fndings from the educational research outlined above are also relevant here. For members to feel that they belong, the association must work to build cultural familiarity and members from other backgrounds should feel acknowledged. This could well be achieved through relatively simple measures, such as providing a varied menu in the club canteen and respecting each other's holidays. It might also be useful, as in schools, to focus upon a common identity that transcends members' different backgrounds. And highly

<sup>101</sup>Wiertz (2016: 820) and van Haaften (2022).

<sup>102</sup> van Haaften (2022).

<sup>103</sup>See, for example, McPherson et al. (2001).

<sup>104</sup>For example, research by Wiertz (2016) amongst residents with Turkish, Moroccan and Dutch backgrounds shows that all three groups strongly prefer to join clubs, societies or associations in which their own group is already well-represented.

<sup>105</sup>For a fascinating description of that process and the tensions it creates, see van Slobbe, 2019.

diverse associations will have to learn to deal with an increased rate of member and volunteer turnover. Funds permitting, taking on paid staff such as a (part-time) manager or caretaker would probably ensure greater continuity.

### *6.3.6 Policy Signifcance: Do Not Expect Too Much of Voluntary Associations*

Three policy conclusions can be drawn from the above analysis. Firstly, it is questionable whether the local authorities in municipalities with a high HHI (see 2.4) can use voluntary associations to leverage a variety of goals in the social domain. Diverse ones are likely to lack the structures and social capital needed to undertake wide-ranging community work. Indeed, they themselves might well need local government backing just to survive.

Secondly, this means that local authorities should not expect too much of this sector as a source of social cohesion *between* population groups.106 Voluntary associations are suited primarily to building cohesion *within* groups. A certain degree of segregation is probably inevitable in organized group activities.

Thirdly, it may be worth considering ways to support alternatives to the structured clubs, societies and associations traditionally responsible for organizing sporting, cultural and suchlike activities in the municipalities in the Netherlands. For example, local authorities in areas with a high turnover of migrants such as expat and horticultural municipalities could invest in more informal groups. Or even promote the use of commercial facilities like gyms, boot camps and yoga studios, which have a low accessibility threshold but still provide a certain degree of public familiarity. And sports clubs could think about arranging 'casual' leagues for expats and migrant workers, which are easy to join or leave as a team or a player. Another option is to focus more upon competitions between schools, as is common in the Anglosphere world – they are more solid institutions than clubs, and already have professional staff and facilities at their disposal. Moreover, participation in sports competitions, cultural activities and so on at highly diverse schools could serve as a source of common identity.

### *6.3.7 Summary*


<sup>106</sup>Cf. Wiertz (2016: 823) and van Haaften (2022).

includes an understanding of the culture of the majority group. It is also important to set standards which promote interethnic tolerance, such as clear antidiscrimination policies.


### **6.4 Strengthening National Cohesion**

### *6.4.1 Diversity Reduces Overlapping Cleavages*

Large sections of the Dutch population are concerned that increased immigration and diversity are straining national cohesion.107 The main integration-related problems identifed by people with a Dutch background are failure to adapt suffciently and pressure on 'national' standards and values. More than half (54%) believe that people with a migrant background should participate in Dutch cultural traditions. People with a migrant background, for their part, say that it is important to respect each other and each other's cultures, but that this does not necessarily mean that they must adopt the same standards and values or participate in cultural traditions.108

In academic research on the consequences of immigration for social cohesion, these concerns are usually nuanced. Sociologists and political scientists argue that Putnam's fnding that people tend to hunker down in the face of increased diversity applies only at the local level.109 There is little evidence, for instance, that

<sup>107</sup>See den Ridder et al. (2019).

<sup>108</sup> den Ridder et al. (2019).

<sup>109</sup> van der Meer and Tolsma (2014) and Scheepers et al. (2013).

overall social trust amongst the Dutch population is in decline. Nor do the numbers of people engaging in voluntary activities and informal care seem to be decreasing.110

Furthermore, it could be argued that the increased diversity of the migrant population makes it less likely that the social fault lines between established groups and newcomers will widen at the national level. And the same also applies, mutatis mutandis, to increased transience. For the stability of a society, it is important that it not have too many 'overlapping cleavages' and that ethnic, religious, geographical, social and political divisions do not overlap excessively. If this happens, a society may be in danger of falling apart. Ominous examples from the past include Northern Ireland, where religious, socio-economic and political dividing lines converged and tore the province apart. Or Belgium, where linguistic differences overlapping with geographical and sometimes socio-economic divisions led to a long and sometimes ferce language confict. Or Lebanon, where religious and political fault lines coincide and the traces of the civil war are still close to the surface. Incidentally, until the so-called 'Pacifcation' of 1917 brought an end to its long-standing 'schools struggle' (over state funding for church schools), the Netherlands also had some very strong centrifugal forces of its own.

By contrast, the new diversity of the migrant population by origin, educational attainment and social background means that the differences between established groups and newcomers at the national level are less clear-cut now than they were in the past. No longer are a few relatively homogenous groups with a migrant background pitted against equally an homogeneous non-migrant population. This has reduced the likelihood of 'us and them' perspectives clashing and of differences *by origin* largely coinciding with differences of religion, language and social status.

Amongst new migrants and their children, for example, we encounter a multitude of faiths: not only are there Muslims, Hindus and a few Buddhists, but also many Roman Catholics, Protestants, Copts and other Orthodox Christians. There are now about one million Christian migrants in the Netherlands, as many as there are Muslims.111 According to many commentators, Islam is a major source of social antagonism in the Netherlands.112 In certain specifc neighbourhoods or schools, it is certainly true that radicalization has given rise to tensions between groups, but nationally it is defnitely not the case that either new immigrants or settled residents

<sup>110</sup>Two relevant studies with somewhat similar results are those by Tolsma et al. (2009) and Gijsberts et al. (2012). Both fnd that diversity at the neighbourhood level does not negatively impact overall trust or willingness to undertake voluntary work. The same also applies to informal care. For overall trust, see also Schmeets (2018). For voluntary work, see also Dekker (2017); page 8 contains a table showing the fgures for volunteering between 1993 and 2017.

<sup>111</sup> de Hart and van Houwelingen (2018).

<sup>112</sup>See, for example: Huijnk et al. (2015: 14), Elchardus (2018: 24–25), and van Houwelingen (2019).

with a migrant background are predominantly Muslim. At present, for example, only 5% of the population identify as Muslim.113 Moreover, their ethnic and religious diversity is also quite considerable. They hail not only from Turkey, Morocco and Suriname, but also from former Yugoslavia, the former Soviet Union, Bulgaria, Afghanistan, Iran, Iraq, Eritrea, Somalia and Syria. And as well as being divided along traditional Sunni-Shia lines, there is also a good deal of variety within both of these branches of Islam.114

Nor has migration to the Netherlands led to the emergence of a de-facto second language here, as is the case with Spanish in the United States and Russian in the Baltic states. Migrants and their children speak dozens of different languages at home, but their main medium of communication with others is Dutch or, for some groups, English. Even in the most diverse neighbourhoods, Dutch is the 'lingua franca' used by children from different backgrounds – although their street slang does include numerous loanwords drawn from a variety of migrant languages. Furthermore, there is no regional dominance by any ethnic group; migrants are spread fairly evenly over large parts of the country. There are no overwhelmingly Surinamese, Turkish or Polish neighbourhoods, and in hardly any does one particular group – with the exception of those with a Dutch background – constitute the numerical majority.115

Moreover, the Netherlands has not seen the formation of a homogeneous 'migrant bloc' on the political or cultural stage. Engagement in the current public debates around the history of slavery and the blackface character Zwarte Piet, for example, is confned mainly to participants of 'indigenous' Dutch and of sub-Saharan African, Afro-Surinamese or Dutch Caribbean heritage. People of Polish, Turkish or Moroccan origin, say, are far less visible in these discourses. And whilst 'multiculturalist' party Denk attracted strong support amongst voters with a Turkish or Moroccan background in the 2017 general election,116 those with an Indo-Surinamese background have been relatively quite likely to back the right-wing populist Party for Freedom (Partij voor de Vrijheid, PVV) in recent years.117

Finally, we observe little socio-economic homogeneity amongst the new generation of migrants to the Netherlands. Unlike the guest workers of the 1960s and 1970s, a fair proportion of those now coming into the country are university educated and work in professions with a high social status. This applies not only to highly skilled migrants from India and China, for example, but also to many of

<sup>113</sup>Schmeets (2018).

<sup>114</sup> van de Donk et al. (2006) and Martinovic and Verkuyten (2016).

<sup>115</sup>Hartog and Zorlu (2009). The only exceptions are the Moluccan neighbourhoods in a number of towns and cities, amongst them Culemborg, Alphen aan den Rijn and Maastricht. These are often a small number of streets with predominantly Moluccan residents.

<sup>116</sup>See Vermeulen and Kranendonk (2019).

<sup>117</sup>See also Roopram and van Steenbergen (2013).

those with an Iranian, Afghan or former Yugoslavian background. Today's immigrants from Turkey, Poland or Bulgaria, too, fnd their way not just to glasshouses and distribution centres, but to lecture theatres and research institutes as well.

### *6.4.2 But Diversity Also Increases the Risk of Fragmentation*

Potentially, then, the new diversity can mitigate 'old' concerns about segregation and a cleavage between the established population and newcomers. That potential is not realized automatically, however. And greater diversity also has a possible downside: an increased risk of fragmentation, with society breaking up into a constellation of smaller groups, each seeking to cling to its own identity and in so doing turning its back on the rest. As more and more different migrant groups settle in the Netherlands, so the risk of collisions between them grows. It is therefore wise to consider how to promote cohesion at the national level, too. It is with this in mind that countries with a long history of immigration, such as Canada, the United States and Australia, have traditionally placed a heavy emphasis upon rituals, symbols and institutions which promote national unity.

What can be done at a national level to guide the conviviality of all these different groups in the right direction? How can we give everyone living in the Netherlands, be they immigrants or have their roots here, the feeling not only that this is their home but also that they are part of a *unifed* society? To do this, we frst need to acknowledge that we are here dealing with different groups starting from different positions. First there are the *transients*, those such as temporary migrant workers, highly skilled migrants and international students who are just 'passing through' and have no intention of settling permanently. Then there are the *newcomers*, who do stay. And fnally the *rooted*, who were born and raised in the Netherlands.

### *6.4.3 Rules for Conviviality*

To regulate the conviviality of all these groups, *rules* are needed that create a frm framework for healthy social intercourse. These are not specifc to the Netherlands and can in fact be found in some form in almost all Western countries, as part of their own debates around immigration and national unity. Based upon the literature, we have identifed three sets of rules which are essential to uphold constitutional democracy, to safeguard national prosperity and the welfare state and to maintain a pluralistic society (see Box 6.5).118

<sup>118</sup>Based upon: Boutellier (2015), Elchardus (2018), and Engbersen and Gabriels (1995).

### **Box 6.5: Ground Rules for Conviviality**

To enable conviviality as a diverse society, everyone living in it must abide by three sets of basic ground rules. Respectively, these uphold constitutional democracy, safeguard national prosperity and the welfare state, and maintain a pluralistic society.


First, there are the *rules of constitutional democracy and the rule of law.* Everyone, be they transients, newcomers or rooted here, is expected to abide by the law and to respect the principles and institutions of the democratic constitutional state, including the principle of equality, the separation of church and state and fundamental rights. They must also treat each other with respect and refrain from racism and discrimination.

Secondly, everyone is expected to *contribute towards society*. After all, a healthy society can only exist by virtue of a certain degree of co-operation and reciprocity. This can be done by performing paid work and paying taxes, but also through other activities indispensable to a fourishing and healthy society, such as voluntary work. In this context, transients and newcomers are expected to make some effort to familiarize themselves with, and adapt to, the manners, traditions, customs and values of the Netherlands as a whole and of the region they settle in. This is why, in Chap. 5 on the reception and integration of migrants, we propose the development of forms of civic integration tailored to all migrants, irrespective of their legal status or intended length of stay: basic versions for transients and more advanced ones, including requirements for the acquisition of Dutch-language profciency and some knowledge of national history, for those aspiring to citizenship. This is crucial for solidarity at the national level. Incoming migrants who work hard to earn their own

living and engage with the community can generally count on more sympathy and goodwill than those who contribute little or nothing. We look at this in more detail in the next chapter.

Thirdly, there needs to be proper scope for the expression of *individual and collective identities*. In line with the intercultural approach described above for education and organized group activities, each group – the majority included – should have the freedom to celebrate its own origins, symbols and customs. Everyone living in the Netherlands has the right to feel at home here, including those with Dutch roots. So in regions of high diversity, for example, room should be created for each group to observe its own holidays: Sinterklaas (St Nicholas' Day), Holi, Christmas, Eid al-Fitr, New Year, Keti Koti (Surinamese Emancipation Day) and so on. This requires a certain fexibility when drawing up work rosters and school timetables. It could also mean giving each group room on the national stage to showcase to its own history, perhaps by supporting the establishment of 'migration museums' focusing upon the stories and backgrounds of the country's various ethnic groups as well as regional museums devoted to local history and customs.

### *6.4.4 Debates About National Identity Divide Rather Than Unite*

Besides upholding these basic requirements, is there a broader task for government in this domain? In the Netherlands, as in most other Western countries, whether the state should also endeavour to create or maintain a distinctive *national identity* has become a recurring topic of sometimes ferce debate – not least because there is rarely any consensus as to what that identity entails.119 If it is anchored solely in the traditional values and customs of the 'indigenous' majority, that leaves no room for those prized by newcomers and their offspring. Leading them to feel excluded and demeaned. Hardly anywhere, moreover, has it proven possible even to agree on what exactly the elements of such a 'Leitkultur' should be. In the Netherlands, the debate about national identity fares up with some regularity but has yet led to produce any widely shared conclusions about the nature and substance of that identity.

Will Tiemeijer shows that our neighbouring countries, too, have been unable to pin down what exactly typifes their national character.120 In France the debate has centred on identifcation with the nation's history and culture, and with its heroes, symbols and traditions: the Revolution, the Enlightenment, Napoleon, Voltaire, Victor Hugo, De Gaulle, the tricolour, *La Marseillaise*, wine, cheese, gallantry and romance. Adherence to basic 'French' values such as 'liberté, égalité, fraternité' and 'laïcité' (strict constitutional secularism) is often mentioned as well, but some also

<sup>119</sup>Cf. RMO (1999) and WRR (2007).

<sup>120</sup>The following is based upon Tiemeijer (2021).

equate 'Frenchness' with membership of a particular primordial group characterized by Caucasian descent, white skin and Christianity, especially Roman Catholicism. In Germany as well, there has been an emphasis upon its status as a nation of history and culture, the birthplace of Goethe and Bach. Religion is also said to be an important binding factor in society, although the state remains neutral in matters of faith. Germans are portrayed as enlightened patriots who love their country but do not hate others. Meanwhile, the people of the United Kingdom experience 'Britishness' as a combination of the general liberal values also mentioned elsewhere and unique characteristics such as the English language, the British landscape, the Royal Navy, pubs and fsh and chips – not forgetting cricket, of course.

What is striking in all these countries, though, is that their public discourses around national identity are divisive rather than unifying. When former president of France Nicolas Sarkozy opened a website and organized regional debates on the theme, they attracted not only edifying contributions but also a torrent of racist and anti-Islamic sentiment. The whole exercise has come to be seen as one of Sarkozy's most unsuccessful projects. Likewise, in Germany the discourse has largely been about what Germans are not – namely Muslim – and many have seen it as a thinly veiled message to migrants that they must assimilate. And in the United Kingdom the outcome has been inconclusive, leading the whole debate to fade quietly away. When *The Times* ran a competition in 2007 to devise a new national motto for Britain in no more than fve words, entries included 'Once mighty empire, slightly used', 'We apologize for the inconvenience' and 'At least we're not French'. The winner perhaps typifed the true national character: 'No motto please, we're British'.

### *6.4.5 Creating Unifying Stories and Institutions*

Despite all this, a society made up of a patchwork of ethnic groups has a lot to gain from fnding a common denominator which transcends and connects them all. Many traditional immigration nations therefore set great store by customs and symbols acceptable to all groups, such as respect for the national fag, singing the national anthem or the celebration of their independence day.

In fact, the same dilemmas as we have seen at schools and voluntary associations play out at the national level as well. An overly uniform majority cultural dominance leaves newcomers feeling excluded, but conversely a multicultural laissezfaire runs the risk of each group retreating into cultural isolation so that society breaks up into separate little islands. As with education and voluntary associations, we therefore propose focusing upon a small number of unifying, group-transcendent stories and institutions. (Here we mean 'institutions' in the sociological sense: certain fxed ways of acting that bring order and structure to social life.). All modern immigrant societies put some considerable effort into maintaining unifying national narratives and into socializing newcomers in line with them. The idea behind this strategy is that an emphasis upon common or inclusive goals positively infuences intergroup relations.

These inclusive general narratives and institutions need to meet at least four conditions: (1) they should transcend the different groups and not be linked explicitly to any one of them; (2) they should be accessible for newcomers; (3) they should by unifying at the national level; and (4) they should provide some continuity from the past to the present and future.

So what might be appropriate unifying stories and institutions for the Netherlands? To some extent this is a pragmatic, empirical question; ultimately, we will have to see what works in practice and what does not. At the same time it is a political issue, too, because such inclusive narratives also say something about what we as a society want to have in common. Below, therefore, we suggest a few avenues for further exploration in the full knowledge that other options could also be considered.121

### 1. *The Dutch language*

The Dutch language is one of the Netherlands' most distinctive national institutions; ours is the only in the nation in world where it is spoken by the vast majority of the population. And that singularity is widely acknowledged. In a survey to fnd out what Dutch residents consider the nation's most typical characteristics, the SCP presented 5000 people with no fewer than 185 possible answers. For each, the researchers asked frst how typical the respondents thought it was of the Netherlands and then how much it contributed towards their own sense of belonging here. In both cases the Dutch language topped the list.122 Speaking it is crucial to connect with others. As we mentioned earlier, Dutch is the principal 'lingua franca' of communication between the many migrant groups in the Netherlands. It also provides continuity over time, as the medium of the nation's spoken and written heritage and its literature past and present. The government should therefore continue to promote Dutch as the common language of the Netherlands. In the private sphere, of course, everyone can use whatever language they wish, and in certain situations and organizations it may be more practical to use another one – English, for example – as the working medium, but in the public sphere and public institutions Dutch should always remain our offcial language. For immigrants this does initially create a barrier because the language is not immediately accessible to them, but to overcome that the government should ensure that high-quality and affordable tuition in it is widely available for everyone settling here, including transients.

2. *The collective struggle for a safe living environment*

Psychology teaches us that a joint struggle against an external threat forges solidarity. In the Netherlands, the eternal fght against encroaching water is a concern for everyone: transients, newcomers and the rooted alike. After all, they all beneft equally from a safe physical environment – that is a classic 'collective good', but in the Dutch psyche it is also more than that: creating that environment has become an important symbol of national unity, transcending personal interest and individual origin. The struggle to hold back water and develop pro-

<sup>121</sup>The following is based in part upon Tiemeijer (2021).

<sup>122</sup> de Hart (2019: 20).

ductive land is not only a consistent theme throughout Dutch history but also a vital task for the future. In the coming decades, the whole nation faces a renewed challenge due to rising sea levels. This never-ending battle offers a narrative redolent of solidarity, a unifying story which not only does justice to the history of the Netherlands and all the groups settled here but is also forward-looking and includes newcomers. They too can, and indeed must, join the fght.

3. *Co-operation, compromise and consensus*

A third national institution consistent with the criteria formulated above is the long Dutch tradition of co-operation, compromise and consensus. Because it is about transcending the interests of any particular group for the common good, this tradition is eminently inclusive and can thus contribute towards the harmonious conviviality of all the different groups in the Netherlands. After all, ours is a nation in which a lot of people, each one of whom cherishes their own freedoms, must live together in a small geographical area – something they can only do successfully if they are prepared to give and take. Within the bounds of constitutional democracy, the rule of law and common sense, of course, it is therefore not unreasonable to expect everyone – transients, newcomers and the rooted alike – to give each other some space and to be prepared to co-operate in delineating it. This is a two-way process: it is not only up to newcomers to adapt to the established population, but sometimes also the other way round. Mutual forbearance is a trait many Dutch people consider important. When the SCP asked respondents in its study of Dutch identity what they hoped the country would look like in 50 years' time, they said they wished for tolerance above all else.123 This aspiration, moreover, is in line with the Netherlands' lengthy history of dialogue and conciliation, whereby groups with quite different outlooks on life have always managed to coexist peaceably. Research shows that emphasizing this tradition of tolerance has the potential to help maintain national cohesion.124

This is not the place to elaborate any further on these points. And as mentioned above, other options are also conceivable. Indeed, a number have already been tried out with some success, in such areas the culture, sports and the royal family. More details of these can be found in Tiemeijer's study.125

### *6.4.6 Summary: National Coherence Through Rules Backed by Unifying Stories*

• The new diversity of the migrant population by origin, schooling and social background means that the differences between established groups and newcomers at the national level are less clear-cut now than they were in the past. No

<sup>123</sup> van Houwelingen (2019: 7).

<sup>124</sup>See Smeekes et al. (2012).

<sup>125</sup>Tiemeijer (2021).

longer are a few relatively homogenous groups with a migrant background pitted against an equally homogeneous 'indigenous' population. This entails a risk of fragmentation, with society breaking up into a constellation of smaller groups, each seeking to cling to its own identity and in so doing turning its back on the rest.

	- Everyone living in the Netherlands transients, newcomers and the rooted alike – must abide by the law and respect the institutions of constitutional democracy.
	- Everyone is expected to contribute towards society.
	- Within reason, everyone's symbols and customs are recognized and acknowledged. This applies to all origin groups, the majority included.

### **6.5 Conclusion: Solidarity, Familiarity and Reciprocity**

The various policy directions discussed in this chapter all have one thing in common: they aim to strengthen the sense of solidarity of everyone living in the Netherlands. A high degree of diversity and transience can strain that. The more diverse a neighbourhood, the less its residents know and recognize each other and the less they feel at home there. Short-stay migration makes it harder to build up long-term friendships at schools and voluntary associations. At the same time, as shown repeatedly by SCP surveys of the national mood, people have a deep-seated need for cohesion and solidarity.

The British economist Paul Collier describes the great importance of a 'sense of belonging'.126 This concept is twofold; it encompasses both *feeling at home* somewhere and *being accepted* there. This duality is at the heart of our message in this chapter. Local authorities, schools and associations should endeavour not only to ensure that their residents, pupils or members feel at home, but also that they are accepted. This applies equally to newcomers (and their children) and to those long settled here (and their children).

One crucial requirement here is a high degree of reciprocity. Not only do established residents need to make room for newcomers and give them a sense of belonging, but the newcomers themselves have to appreciate local customs and habits and make a commitment to their neighbourhood, school or association. This also means that a more intercultural approach, taking into the backgrounds of all concerned, is best suited to coping with the new diversity.

<sup>126</sup>Collier (2018: 63–65).

Such reciprocity is achieved more easily at the local than the national level. Public debates about national identity are more likely to be divisive than a source of solidarity. Urban and regional identities, on the other hand, are relatively 'casual' and so within easier reach. They do not require a family tree going back several generations. A sense of belonging to a neighbourhood, a city or a region can be acquired by going to school there, supporting the local football club or picking up its accent. We thus see that many residents with a migrant background develop local allegiances earlier than national ones, and that in the second generation these are even stronger than ethnic identities.

Promoting familiarity in neighbourhoods, at schools and in voluntary associations is primarily a task for local authorities and institutions. Central government's principal role is to uphold the basic ground rules of conviviality: everyone living in the Netherlands should respect its constitutional democracy and the rule of law, and contribute towards society, whilst at the same time being allowed space to maintain their own identity. In addition, government has a responsibility to communicate unifying 'national stories' and to ensure that newcomers engage with them. The idea behind this strategy is that an emphasis upon common or inclusive goals positively infuences intergroup relations.

### *6.5.1 Summary*


**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 license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license 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 Towards Migration Policy with an Eye for Social Cohesion**

In Chap. 3 we established that increasing ethnic diversity puts pressure on social cohesion.1 In Chap. 4 we noted the lack of coherence between policies addressing migration and aspects of conviviality. These observations raise the question as to what extent the government could take more account of the issue of social cohesion in the design of Dutch migration policy.

### **7.1 Migration Policy as a Balancing Act**

Achieving better coherence between migration policy on the one hand and policy to increase social cohesion on the other is challenging because social cohesion is not the primary issue determining Dutch migration policy. That is actually the product of a complex balancing act involving different interests in all kinds of areas, in particular the economy, respect for private and family life and humanitarian considerations.2 We can link these three particular interests to three distinct types of migration: respectively, labour, family and asylum migration. Together, these account for the lion's share of all migration by non-Dutch nationals to the Netherlands.3

In outlining the contours of a migration policy that also serves Dutch society, we assume that there is a certain leeway for this to be decided at the national level. At the same time we acknowledge that Dutch policy is anchored in international

<sup>1</sup>See also Jennissen et al. (2018).

<sup>2</sup>See also Hampshire (2013).

<sup>3</sup>Most of the remainder is student migration. This is often regarded as a gateway to entice highly skilled workers to settle in the Netherlands, but in fact a high proportion of student migrants leave again within a few years (Bijwaard, 2010; Bijwaard & Wang, 2016). Since family and asylum migrants stay far longer on average (Schmeets, 2019), in practice they have a greater impact in increasing of ethnic diversity. Given the temporary nature of student migration, we do not address it in this chapter.

R. Jennissen et al., *Migration Diversity and Social Cohesion*, Research for Policy, https://doi.org/10.1007/978-3-031-14224-6\_7

regulations and agreements, including free movement within the European Union, the Geneva Refugee Convention, the European Convention for the Protection of Human Rights and Fundamental Freedoms (ECHR) and the Charter of Fundamental Rights of the European Union. International legal frameworks thus shape much of the legislation governing migration to the Netherlands. In many cases these have been established through complex negotiations between states, involving multiple issues. Naturally, the Netherlands must remain a reliable partner for other states bound by international law and so this factor should always be taken into account when considering any changes to the rules in the feld of immigration.

Despite the fact that international treaties outrank the Dutch Constitution in the legislative hierarchy and that the Dutch legal order is bound by EU law, so-called 'European policy variations' are possible.4 These can be applied particularly to asylum migration and to labour migration from outside the EU/EFTA zone. Individual member states are also allowed some fexibility with regard to family migration. European directives may be incorporated into national legislation in different ways.5

In elaborating a migration policy that serves social cohesion, our starting point is the three main types of migration to the Netherlands by non-Dutch citizens: labour, family and asylum migration.6 The policy governing each of these is rooted in different interests or principles.

In order to gain a better insight into the complex pattern of interests and principles associated with Dutch migration policy, it is necessary to draw an analytical distinction between these three types. We acknowledge that people's actual motive for migrating does not always correspond with their offcial one: the grounds on which they seek to gain access to another country. In many cases, moreover, they in fact have multiple motives. But whilst this plurality is an important factor when studying the phenomenon of international migration, it is advisable when it comes to the admissions policy of the Dutch government to draw a clear distinction between the 'offcial' migration types.7

<sup>4</sup>Hirsch Ballin et al. (2020). In their report *European variations as a key to cooperation*, Hirsch Ballin et al. distinguish three forms of variation: (1) type of membership/participation; (2) type of decision-making; and (3) policy content and interpretation of the legislation. Here we are referring here to the last of these.

<sup>5</sup>For example, the European Directive on the Right to Family Reunifcation does not apply to the United Kingdom, Ireland and Denmark (Groenendijk & Strik, 2018).

<sup>6</sup>These are ideal types, which do not exhaustively cover or describe reality.

<sup>7</sup>Policymakers in Germany, too, believe that the various types of migration should remain strictly separated. For example, a proposal by the social-democratic SPD to provide highly educated failed asylum seekers with a work permit anyway – the so-called 'Lane change' ('Spurwechsel') – was unsuccessful (*Süddeutsche Zeitung* 2 October 2018).

### **7.2 Labour Migration: Demand-Driven and Circular**

In discussing labour migration to the Netherlands, we have to distinguish between migration within the EU/EFTA zone and by so-called 'third-country nationals' from other parts of the world. Dutch policymakers are able to exert virtually no infuence over arrivals from other EU/EFTA member states because the Netherlands is bound by the 1957 Treaty Establishing the European Economic Community,8 which provides for 'freedom of movement for workers' (currently Article 45 TFEU).9 The majority of incoming labour migration, approximately 55–60% in 2016, originates within the EU/EFTA.10

Because the Netherlands is bound by EU treaties, in this section we focus entirely upon labour migration from outside the EU/EFTA zone. In so doing we outline means to improve policy in this domain so that it continues to bolster the real incomes of those already living in the country whilst at the same time harming social cohesion as little as possible.

### *7.2.1 Complementary Workforce Required*

Government policy for labour migration from outside the EU/EFTA zone should in the frst place aim to ensure that this improves the real incomes of those already living in the Netherlands. There is a generally held assumption in this respect that when migrants play a complementary role in the labour market, not affecting the established workforce, that benefts the economy.11

At frst sight it seems relatively easy to formulate a widely accepted labour migration policy on the basis of this assumption. In fact, however, that gives rise to three problems. The frst is a distribution issue. It is quite possible that such a policy

<sup>8</sup>Now the Treaty on the Functioning of the European Union (TFEU).

<sup>9</sup>Hirsch Ballin et al. (2020).

<sup>10</sup>Our own estimate based upon: (1) the total number of immigrants in 2016 – about 34,000 – whose motive for coming to the Netherlands was to work; and (2) the number of those immigrants coming from non-EU countries and from 'other European' countries, as estimated by van Duin et al. (2018) – about 15,000 and about 2000, respectively. Note that we count only labour migrants on the national population register (Basisregistratie Personen, BPR). If we were also to include those who have not registered, intra-EU migrants would account for a considerably higher proportion of all labour migrants.

<sup>11</sup>Borjas (1999). When complementary roles are mentioned, people often think of highly skilled migrant workers. In fact, though, it is certainly not the case that they always make a positive contribution towards a country's economic development. They may actually cause displacement within the labour market (see, for example, Kemnitz, 2009). Likewise, the contribution made by low-skilled migrant workers is not necessarily negative. They are often willing to perform work most local people avoid (due to unsocial hours, heavy labour and so on), which allow them to play a complementary role in the jobs market.

ends up benefting employers in particular, whilst employees lose out.12 The Netherlands Bureau for Economic Policy Analysis (Centraal Planbureau, CPB) argues that settled workers with knowledge and skills similar to those possessed by immigrants are disadvantaged in this situation, but the opposite is true for those with no such overlap.13

The second problem is one of short-term gains versus long-term effects.14 Labour migration that flls current gaps in the jobs market, and therefore helps the economy in the short term, can still have a negative impact over a longer period. An example of this is the recruitment of low-skilled migrant workers from Morocco and Turkey in the 1960s and early 1970s.15

A third issue is that proponents of labour migration tend to focus upon its economic benefts, without considering its potential to disrupt social cohesion. The rewards are reaped primarily by employers, but the burden of any accompanying problems is shifted onto society at large. We all therefore share the cost of tackling them, which in turn leads to a decrease in real per-capita income. In fact, this too is a distribution issue. Moreover, it can take time for the social problems to set in.

### *7.2.2 The Importance of Circularity*

To a large extent, problems of social cohesion seem to be limited if labour migration remains temporary in nature. To clarify the relevance of a circularity-based system, it is useful to consider the period of reconstruction in the wake of the Second World War. Those years were characterized by unprecedented economic growth, which resulted in a serious shortage of workers in the most labour-intensive sections of the economy.16 Dutch industries, in co-operation with the government, tried to fnd a solution to this by recruiting in a number of countries around the Mediterranean Sea.17 The idea behind this was that it would result in a 'win-win' situation. On the one hand the 'guest workers' would strengthen Dutch industry by doing jobs for which no local employees could be found, whilst on the other they would be able to convert the (often modest) wages they earned in the Netherlands into a high real income in their country of origin through remittances or by saving for their return home. This assumption proved partially correct: the vast majority of Italian and Spanish guest workers (see Box 7.1), and also some of those from Turkey and Morocco,18 did indeed migrate back to their country of origin with a considerable amount of money in local currency.

<sup>12</sup>Borjas (1999) and WRR (2001).

<sup>13</sup>Roodenburg et al. (2003).

<sup>14</sup>Borjas (2006).

<sup>15</sup>See, for example: WRR (1989) and Lucassen and Lucassen (2018).

<sup>16</sup>See, for example: Kuipers and van Zon (1982) and Hartog and Vriend (1989).

<sup>17</sup>See, for example: van de Beek (2010) and Jennissen (2013).

<sup>18</sup>Ten years after entering the country, 40% of Turkish and 30% of Moroccan guest workers had returned to their country of origin (Schoorl, 2011).

### **Box 7.1: The Arrival and Return of Italian and Spanish Guest Workers**

Labour migration by Italians to the Netherlands (and other parts of northern and western Europe) has a long history. Most of those who arrived between the wars came from northern and central Italy to work as terrazzo workers (specialist foorlayers), plaster fgurine makers, artists or ice-cream makers.19 After the Second World War, the majority hailed from southern Italy20 and worked mainly in industry and mining.21 But as their own country's industrial north began to fourish, more and more migrants from the south chose that as a destination over a move to north-western Europe.22 As a result, Italian labour migration to the Netherlands declined from the 1960s onwards and many of those already here returned to their homeland at the end of their contract.

Up until 1957, the Franco regime in Spain pursued a policy of autarchy. Consequently, there was hardly any Spanish labour migration to north-western Europe. That changed with the so-called Stabilization Plan (Plan de Estabilización) of 1959, which liberalized the international movement of people and goods. Labour migration was now actually encouraged, in part through the establishment of the Spanish Institute of Emigration (Instituto Español de Emigratión).23 This prompted an exodus of workers to other European countries, the United States and Latin America. The period 1960–1973 saw several years in which the number of registered labour migrants with another European country as their destination exceeded 100,000.24 In the 1960s, Spaniards made up the largest group of guest workers in the Netherlands,25 but their role in the migrant workforce declined in the early 1970s due to more intensive recruitment in other countries. Also, as in Italy, albeit on a more modest scale, domestic migration began to supplant international movement. This was due in part to the success of the Stabilization Plan itself: between 1959 and 1973 Spain had the world's second highest economic growth rate, behind only Japan, and became its ninth largest economy.26 Catalonia, the Basque Country and the region around Madrid, in particular, fourished27 so much that they become appealing alternatives for potential migrant workers from the rest of the country.

<sup>19</sup>Chotkowski (2006).

<sup>20</sup>See, for example, Haug (2000).

<sup>21</sup> van der Hoeven (2012).

<sup>22</sup> del Boca and Venturini (2003).

<sup>23</sup>Mansvelt Beck (1993).

<sup>24</sup>Dirección General de Migraciones (1993).

<sup>25</sup>See, for example, Lakeman (1999).

<sup>26</sup>Baklanoff (1976).

<sup>27</sup>Harrison (1980).

The abiding legacy of the 'guest worker' period, however, comes from those who stayed in the Netherlands and eventually brought over their families from their countries of origin. As a result, the Turkish and Moroccan communities are now the two largest groups here with a migrant background.28 Unfortunately, even into the second generation their socio-economic integration has sometimes fallen short.

National governments would therefore be well-advised to learn from the mistakes of the past and consider more circular forms of labour migration. Just as the decades following the Second World War saw what seemed at the time to be a neverending economic boom, so we are now experiencing a fundamental development very likely to result in a persistent shortage of labour. That is the ageing of the Dutch population, which will continue until around 2040 and is steadily increasing the proportion of economically inactive people in society.29 Due to the ageing population and the economic development of the EU Member States in central and eastern Europe, future labour migrants will increasingly come from outside the EU/EFTA zone. This offers opportunities for setting circular conditions for labour migration.

### *7.2.3 Demand for Different Types of Labour Migrant*

It is often assumed that the Netherlands' primary need is highly skilled migrants, but in fact this is not the case. As early as 2003, for example, the Advisory Committee on Migration Affairs (Adviescommissie voor Vreemdelingenzaken, ACVZ) foresaw impending labour shortages in the care sector, infrastructure, construction, hospitality, agriculture and horticulture.30 This makes it important, as well as focusing upon highly skilled professionals, to consider how to fll thousands of potential vacancies for low-skilled and medium-skilled personnel if their principal current source, central and eastern Europe, threatens to dry up. If policymakers wait until that tide starts to ebb, there is a risk that other countries will gain a head start in attracting workers at these levels. Indeed, something similar happened in the 1960s. Partly because Germany started recruiting Turkish guest workers earlier, it succeeded better than the Netherlands in attracting relatively well-educated people from more developed regions. The same could now happen again.31

On 1 March 2020 a new law came into force in Germany: the Skilled Immigration Act (Fachkräfteeinwanderungsgesetz). In many respects this is similar to the Dutch Highly Skilled Migrants Scheme (Kennismigrantenregeling). There is a crucial difference, though: the German version allows 'skilled workers with approved vocational qualifcations' ('Fachkräfte mit qualifzierten Berufsausbildungen') to work in the country under certain conditions, not just the 'highly skilled' professionals

<sup>28</sup>See, for example, Jennissen et al. (2018).

<sup>29</sup>See, for example, Ekamper and Van Nimwegen (2018).

<sup>30</sup> de Lange et al. (2003).

<sup>31</sup>Dagevos et al. (2006).

('kennismigranten', effectively meaning university graduates) covered by the Dutch scheme. Through this measure the German government is attempting to fnd a solution for shortages in the labour market and demographic developments (the ageing population).32 This approach offers a good starting point for a reassessment of Dutch labour migration policy.

### *7.2.4 Do Highly Skilled Migrants Make Us Any Wiser?*

By focusing upon the development of a so-called 'knowledge economy', the Dutch government hopes to improve labour productivity and hence national prosperity.33

It is obvious that increasing the amount of human capital at work has a role to play in achieving this goal, but it is questionable whether all forms of knowledge contribute towards it. In certain disciplines, that seems doubtful. Nevertheless, many researchers and policymakers assume that a labour force with better skills overall – regardless of the specifc areas in which they have improved – will automatically generate higher labour productivity and so increase GDP per capita.34

Views differ on the relationship between educational expansion and economic growth, as well as the underlying mechanisms involved. Some believe that technological progress increases the need for a highly skilled workforce.35 This leads to relative growth in wages for that group, making it more attractive to stay in education and obtain the necessary qualifcations. In this view, a person with a higher relative level of education is more productive.36 And if the average educational attainment of the population as a whole increases, that leads to greater collective prosperity. Others, however, take the view that ongoing educational expansion is caused by individuals trying to defend their (future) position in the labour market relative to others.37 A person with a higher education simply displaces someone with a secondary education, and they in turn displace someone with only a basic education, without this resulting in additional productivity.38

This latter point of view plays only a minor role in Dutch policymaking concerning labour migration, as exemplifed by the Highly Skilled Migrants Scheme. In certain cases, someone seeking to enter the country under this arrangement has only to meet a set salary threshold (and in some cases not even that),39 regardless of

<sup>32</sup>For more information, see the German government bill published on 13 March 2019 (BT-Drucksache *19/8285*). This can be found on the website of the Bundestag, at https://dip21. bundestag.de/dip21/btd/19/082/1908285.pdf

<sup>33</sup>See, for example, Jacobs and Bovenberg (2005).

<sup>34</sup>See, for example: Canton et al. (2005) and Mous (2008).

<sup>35</sup>Lipset and Bendix (1959) and Blau and Duncan (1967).

<sup>36</sup>Becker (1964).

<sup>37</sup>See, for example, Thurow (1975).

<sup>38</sup>Collins (1979).

<sup>39</sup>For academic staff and junior doctors, for example (Buers et al., 2018).

whether their specifc skills are actually needed here. If the truth in the choice between the two visions described lies somewhere in the proverbial middle, then ideally we would opt for a sector-specifc admissions policy for labour migrants.

### *7.2.5 Labour Migration and Social Cohesion*

Since the end of the 1990s, cities like The Hague and Rotterdam have faced a rapid and large infux of labour migrants for which they were not prepared.40 This has led to social problems. The most obvious is the lack of adequate and fexible living accommodation, which has resulted in issues of overcrowding and put great pressure on the housing market. Others include a sudden lateral infux of young children into education, public nuisance and homelessness.41 This is why we have argued for the creation of structural facilities aimed at temporary residents, especially in housing (see Chap. 5).

Employers also have a responsibility for these facilities. If the housing, education or healthcare burden arising from labour migration exceeds the capacity of neighbourhoods, communities and regions, that affects the living situation of the existing population. This could be an argument for temporizing or reducing such migration, despite the economic interests at stake.

We therefore suggest taking into account local social problems resulting from temporary labour migration when issuing work permits. Currently a permit can be refused if an individual employer is unable to provide the applicant with suitable accommodation, but the collective social costs for a community or region associated with the arrival of large groups of labour migrants are not taken into account. We propose that these broader costs also be borne in mind when assessing individual applications for work permits for potential labour migrants from outside the EU/ EFTA zone. For example, by considering whether there are adequate available facilities in such domains as housing, education and civic integration42 to integrate the migrants – and possibly their families as well – into local society.

### *7.2.6 Summary*

• We advise the Dutch government to end the use of educational qualifcations and/or salary level as the principal criterion for the admission of migrant workers from outside the EU/EFTA zone, and instead consider actual labour-market

<sup>40</sup>Holtslag et al. (2012).

<sup>41</sup> van Ostaijen et al. (2018).

<sup>42</sup> In the case of civic integration, we proposed in Chap. 5 that the government should make civic integration provisions for all migrants, including those who are not subject to an integration obligation. Employers may be asked to contribute to the costs.

needs. This means no longer seeking only or primarily to attract highly skilled migrants, but focusing upon all sectors of the economy in which there are labour shortages.


### **7.3 Family Migration: Respect for Private and Family Life with an Eye for Social Cohesion**

This section looks at whether migration policy in respect of family formation can be adjusted in such a way that it places less pressure on social cohesion. The relevance of this question stems in part from the scale of family migration. Every year since the recession of 1973, this has accounted for the largest number of non-Dutch nationals settling in the Netherlands. Figures from Statistics Netherlands (Centraal Bureau voor de Statistiek, CBS) show that more than 800,000 family migrants have come to the country since the beginning of this century. Their demographic impact is therefore considerable. As a result of their arrival, the relatively modest group of fewer than 50,000 labour migrants from Turkey and Morocco in the early 1970s had grown to 800,000 people by 2020.43

### *7.3.1 Family Reunifcation and Formation*

We differentiate two forms of family migration: family reunifcation and family formation.44 The former refers to the partner and/or the children of someone who has previously settled in the Netherlands – as a labour or asylum migrant, say – joining them here. They thus already had a family in their country of origin.45 Family

<sup>43</sup>According to Statistics Netherlands, there were approximately 48,500 immigrants with a Turkish or Moroccan background in the Netherlands in 1972.

<sup>44</sup>See, for example: de Beer and Noordam (1992) and Nicolaas et al. (2011).

<sup>45</sup>Following asylum migration, the rules for family reunifcation differ. For a period of 3 months after obtaining an asylum residence permit, anyone – even a minor – can act as a sponsor of family members travelling to join them (Cleton et al., 2017).

formation, by contrast, is when someone brings in a partner they did not have before they became lawfully resident in the Netherlands. In this case the person already living here – the so-called 'sponsor' – need not be a migrant; their 'lawful residence' could have begun at birth. According to Article 8 of the ECHR and Articles 7 and 9 of the Charter of Fundamental Rights of the EU, persons seeking to reunify their family or to form one can invoke the right to respect for family life.

In the Netherlands, an income requirement of 100% of the minimum wage currently applies to both forms of family migration.46 In the case of family formation, both the sponsor and the immigrating partner must also be at least 21 years old.47 The lower age limit for family reunifcation is 18 years.

Reunifcation always follows a previous form of migration. This does not necessarily apply to family formation. In many cases the sponsor was born in the Netherlands, but sometimes they are migrants themselves. A distinction can also be drawn between sponsors born in the Netherlands with a migrant background and those with a Dutch background.48 By way of illustration, during the period 2007–2011 nearly 30% of the sponsors of migrants who came to the Netherlands for family formation had a Dutch background49 and in more than 40% of cases the migrant and the sponsor had the same country of origin.50

### *7.3.2 Family Migration and Social Cohesion*

Family formation can result in long-term chain migration. And when migrants in a vulnerable social position form families with partners who barely participate in Dutch society, if at all, this may undermine social cohesion. As can the fact that some migrant communities are strongly homogamous in terms of their partner choices (see Fig. 7.1), meaning that two people from the same origin group are involved in family formation. Homogamy in partner choice can thus result in substantial chain migration by people who are vulnerable socially and have diffculty fnding their way in Dutch society.51

<sup>46</sup>See, for example, Lodder (2019). There are legal differences between family reunifcation and family formation. Apart from the fact that reunifcation can also involve minor children, the law also distinguishes between them with regard to the family members of asylum migrants coming later to join them (Cleton et al., 2017).

<sup>47</sup>See, for example, Lodder (2019).

<sup>48</sup>Strictly speaking, a sponsor may also be foreign-born without a migrant background.

<sup>49</sup>Sterckx et al. (2014). These percentages vary fairly widely by origin group. The sponsors of Turkish, Moroccan, Afghan, Somali, Iranian, Iraqi family migrants to the Netherlands rarely have a Dutch background, whereas almost 80% of those pursuing family formation with someone from Thailand or the Philippines do.

<sup>50</sup>Sterckx et al. (2014). In 17% of cases, the two partners had different migrant backgrounds. In 12% of the cases, the sponsor's country of origin was unknown.

<sup>51</sup>Bonjour (2009).

Homogamy in partner choice is common amongst asylum migrants from Afghanistan, Somalia, Iraq, Eritrea and Syria as well,52 also groups often in a vulnerable social position. There are differences between these groups, though. For instance, refugee status holders from Afghanistan who came to the Netherlands in the period 1995–1999 are somewhat better positioned in the labour market than their counterparts from Iraq and Somalia. In addition, research has shown that the educational performance of children from refugee groups is relatively good.53 And we know that as the level of educational attainment rises, so the degree of homogamy in the choice of partner decreases.54 As a result, family formation migration declines.

<sup>52</sup>Sterckx et al. (2014: 62) write, "Marriage migrants of Afghan, Somali, Iranian and Iraqi origin also rarely have sponsors with a Dutch background. In these groups, it is relatively common for the sponsor and their prospective spouse to have different countries of origin. These marriage migrants are probably members of the diaspora. This applies particularly to Somalis."

<sup>53</sup>Hartgers et al. (2017).

<sup>54</sup> van Tubergen and Maas (2007).

### *7.3.3 Stricter Requirements Have Boosted Labour-Market Participation*

In order to enhance socio-economic participation by family migrants, over the years the Dutch government has steadily tightened the necessary requirements. In particular those for family formation. This is considered benefcial for both society and the migrants themselves.55 In applying these measures, the government has sought to strike a balance between respecting the right to a private and family life and issues of integration. The three principal requirements imposed56 are as follows.


<sup>55</sup>Bonjour (2009).

<sup>56</sup>Other, somewhat less signifcant measures introduced in the early 2000s were the equalization of the income requirement for married and unmarried people (2001) and a reform of the fees system (2005). See Kulu-Glasgow and Leerkes (2009).

<sup>57</sup>See Nicolaas et al. (2011).

<sup>58</sup>See CJEU Chakroun, 4 March 2010, C-578/08, par. 64. See also Sterckx et al. (2014).

<sup>59</sup>Bonjour (2010).

<sup>60</sup>The accompanying explanatory memorandum states, "The integration of newcomers is a twoway process. Dutch society is expected to offer them opportunities, and they are expected to make an effort and to contribute towards Dutch society. It is also a long process. Although integration may proceed smoothly in the individual case, the process has to start all over again for each new migrant. This repeatedly sets it back at the macro level, which does nothing to beneft social cohesion or broad support for the acceptance of new migrants … Signifcantly, too, continuing immigration with consistently defcient integration can give rise to processes that eventually result in the marginalization of certain groups, in the sense that they are increasingly unable to participate in society, lack opportunities in the labour market and become systemically dependent upon benefts in lieu of earnings" (*Kamerstuk* 29700, no. 3, 2003–2004: 2–3).

response to a draft version of the Act, the ACVZ stated that "this argumentation is suffcient in principle to justify an obligation to integrate in advance … The structural problems with integration justify an unorthodox approach, one which appeals to the foreign national's own responsibility and capabilities, and to their commitment towards Dutch society, and thereby lays a foundation for an effective integration process in the Netherlands."61

Research shows that tightening the age and income requirements has had a positive effect with regard to the labour-market participation of family migrants.62 And the Civic Integration Abroad Act has improved their participation in Dutch society.63 That law was amended in April 2011 to increase the required standard of language profciency from level A1-minus to level A1 in the Common European Framework of Reference for Languages (CEFR). This provides better preparation for the civic integration programme to be followed in the Netherlands, which has a pass requirement of level A2. In addition, the examination was expanded to include a test of literacy and reading comprehension. Evaluations show that these changes have resulted in family migrants having a better command of Dutch upon arrival in the Netherlands, and that they integrate somewhat more successfully once here. Their average educational attainment is also slightly higher now.64

### *7.3.4 Are More Requirements Needed?*

Despite the stricter criteria they have to meet, substantial numbers of family migrants still do not participate in the Dutch labour market. And they often live in isolation within their own origin communities.65 This would seem to argue in favour of imposing additional requirements upon sponsors of *family formation* migration.66 An educational criterion, for example, such as holding at least a basic

<sup>61</sup>ACVZ (2003: 5).

<sup>62</sup>Muermans and Liu (2009). These policy measures also had a positive effect upon sponsors' labour-market participation. See also the memorandum of 2 October 2009 from the Minister for Housing, Communities and Integration, the Minister of Justice and the Minister of State for Justice on marriage and family migration (*Kamerstuk* 32175, no. 1, 2009–2010).

<sup>63</sup>Brink et al. (2009) and Odé et al. (2014).

<sup>64</sup>Odé et al. (2014). See also Klaver et al. (2016).

<sup>65</sup>See, for example: Dagevos et al. (2016) and van Gaalen et al. (2018).

<sup>66</sup>This suggestion is in line with the analysis and proposals by the responsible ministers in the fourth Balkenende government (2007–2010) concerning the imposition of stricter requirements upon both the sponsor and their partner in cases of migration for family formation. See the memorandum of 2 October 2009 from the Minister for Housing, Communities and Integration, the Minister of Justice and the Minister of State for Justice on marriage and family migration (*Kamerstuk* 32175, no. 1, 2009–2010). This stated that they would "argue within the EU to allow an educational requirement to be imposed upon the sponsor in the Netherlands, and for the sponsor to be required to have passed their civic integration examination in the Netherlands".

qualifcation,67 could counter early school-leaving. This is also a better predictor of active social participation than an income equivalent to at least 100% of the national minimum wage.68 In addition, such a requirement would strengthen the position of young women from origin groups with a high degree of homogamy, making them less susceptible to external pressure from members of their family.69

However, any measure of this kind would be in breach of the current EU Family Reunifcation Directive.70 An earlier plea by the Dutch government to be allowed to impose an educational requirement for sponsors in the Netherlands failed to receive any support.71 Criteria can be set regarding income (100% minimum wage) and age (21 years), but not education.

### *7.3.5 Summary*


### **7.4 Asylum Migration: A More Active Government Approach**

Since the mid-1980s, asylum has been one of the main factors shaping patterns of migration to the Netherlands. Many of the groups involved are relatively large, including refugees from Iran, Somalia, the former Yugoslavia, the former Soviet

<sup>67</sup> In the Netherlands, the ACVZ suggested in 2010 that a system be developed with an "income or educational requirement" rather than an income requirement alone. On this point it stated (ACVZ, 2010: 7) that "the underlying assumption is that a sponsor with a high level of educational attainment will probably be able to meet the income requirement in the future without any problem." 68See, for example, Lilly et al. (2007).

<sup>69</sup> In addition, this provides a response to undesirable situations arising out of the income requirement. For example, there are known cases of individuals abandoning their studies in order to meet the requirement temporarily or incurring debt in order to 'buy' themselves into a job for a year. See Kulu-Glasgow and Leerkes (2009: 3).

<sup>70</sup>Article 7(2) of the Family Reunifcation Directive does not provide a basis to introduce national measures affecting the sponsor. Were integration measures to be used as de-facto means of restricting reunifcation, they would constitute an additional requirement at odds with the purpose of the directive, which is to promote family reunifcation.

<sup>71</sup>See the appearance of 15 February 2012 by Gerd Leers, Minister for Immigration, Integration and Asylum, before the parliamentary Select Committee on Immigration and Asylum (Algemene commissie voor Immigratie en Asiel) and Standing Committee on European Affairs (Vaste commissie voor Europese Zaken): https://zoek.offcielebekendmakingen.nl/kst-30573-98.html. See also the EC guidance on family reunifcation.

Union, Afghanistan, Iraq, Syria and Eritrea. According to a rough and conservative estimate by Statistics Netherlands, based upon data from the 1980s onwards, approximately 500,000 people now living in the country either arrived as asylum migrants themselves or have at least one parent who did so.72

Asylum is also a substantial contributor towards the increasing diversity of the nation's migrant population. This is because these groups come from a multitude of countries of origin, many with no signifcant previous history of migration to the Netherlands. To a large extent, asylum migration contributed to the 'new diversity' which places pressure on social cohesion. The question here, then, is how well asylum policy can stay true to its core objective of providing refugees with a safe haven whilst at the same time taking account of issues related to social cohesion. Below we outline two possible responses to this dilemma.

### *7.4.1 More Invited Refugees*

A frst option is for the Netherlands to gain greater control over the refugees it receives. This can be done by, for example, increasing the proportion invited to the country. To illustrate what this means: some 30,600 asylum seekers arrived of their own accord in 2017, compared with 2265 who were invited under international resettlement schemes – the highest number since the resettlement policy was introduced in 1977. So for each one of them, approximately 15 asylum seekers came to the country uninvited.

No-one can be denied the right to claim protection as a refugee on his or her own initiative, but the Netherlands could, as part of a more effective system of dispersal within the EU as a whole, take a larger share of UNCHR refugees and limit the number of other asylum seekers it accepts. This would give it a greater say in who exactly comes to the country. Moreover, such an arrangement could ensure that even the most vulnerable have access to the Dutch asylum system. This might well increase public support for the presence of refugees, as well as acceptance of the related diversity. After all, asylum migration in its present form is highly selective. There are strong indications that it is primarily the strongest and most prosperous individuals who are currently gaining access to the European asylum system.73 A broader invitation-based policy would also make it possible to respond more quickly to sudden crises, such as those involving the Yazidis in the Sinjar Mountains or at the Mória camp on Lesbos.

When inviting refugees, the question arises as to who is deemed eligible to come to the Netherlands. Can the government make a selection using criteria indicating that the person concerned is likely to be able participate well in Dutch society, or

<sup>72</sup>Scheffer (2018).

<sup>73</sup> van Liempt (2007), Betts and Collier (2017), Lange and Pfeiffer (2019), and Koikkalainen et al. (2020).

should it base itself exclusively upon humanitarian considerations such as vulnerability and risk of persecution? In our opinion, humanitarian determinants should always be paramount but other factors may play a subsidiary role. To an extent, this is in fact already standard practice. Since 2005, for example, 'integration potential' (see Box 7.2) has been one of the criteria used in the selection of invited refugees.74

### **Box 7.2: Integration Potential as an Additional Criterion for Invitation75** The basic starting point when inviting refugees to the Netherlands is their vulnerability as individuals. The Immigration and Naturalization Service (Immigratie- en Naturalisatiedienst, IND) obtains information about this from UNCHR reports and often also from the Regional Development and Protection Programme (RDPP) of the European Commission. For example, the service examines whether people have been victims of torture and to what extent their lives are under threat in the country where they are currently located. However, it does also take 'integration potential' into account when choosing which countries to invite refugees from.

The UNCHR offce in the selected country nominates candidates likely to be eligible for resettlement in the Netherlands. To help with this, the IND prepares a so-called 'pre-mission questionnaire' as one of its selection tools. Interviews are then held with the nominees, covering such topics as medical, work-related or religious issues. Their own expectations of resettlement and integration potential are also discussed.

In a 2006 submission to parliament, the then Minister of Justice listed a number of negative indicators of integration potential which could lead to a nominee being rejected: "Not being prepared to learn the Dutch language and integrate into Dutch society, displaying inappropriate conduct, intent to cause social unrest or holding militant/fundamentalist views that could give rise to undesirable behaviour."76 At the interview, the IND offcer also tries to ascertain whether resettlement in the Netherlands would actually be in the candidate's best interests given that it might involve them leaving behind an established network in their current country of residence. Conversely, already having contacts in the Netherlands is often seen as an advantage.

<sup>74</sup>Besters and Diepenhorst (2016).

<sup>75</sup>The information in this box is drawn largely from a conversation with Nicolien Rengers (Central Agency for the Reception of Asylum Seekers, COA), Tim Sparla (IND) and Afke Siezen (Ministry of Justice and Security, Directorate for Migration Policy, DMB).

<sup>76</sup>Tweede Kamer, vergaderjaar 2005–2006 (Netherlands House of Representatives, 2005–2006 session), 19637 and 29237, no. 1071, 3.

Refugees' family situation can also play a role. For example, large sections of the Dutch population are open to taking in families with children but would prefer not to accept large groups of young men.77 Due to increasing feelings of insecurity, the Canadian government decided in 2015 that Syrian families and vulnerable refugees would be given priority over single men for resettlement in that country.78

The workability of this system of resettlement by invitation is subject to two conditions. The frst is better reception facilities for refugees in their regions of origin,79 which in turn will require good European and international co-operation.80 With such facilities in place, it should be possible to substantially shift the ratio of invited refugees to ad-hoc asylum seekers.

Were other European countries to adopt the same approach, moreover, this system could mobilize the fnancial resources needed for a better, more generous reception in the region. The entire budget of the UNHCR in 2015 was €4.7 billion euros,81 from which it accommodated more than 12 million international refugees outside Europe. Even if the agency had spent every penny on those activities, that would still have amounted to only about €400 per person. By comparison, in 2018 a oneyear stay at an asylum reception centre in the Netherlands cost around €27,900 per person. With 20,500 people submitting an initial application for asylum in that year, total spending on these facilities reached €590 million.82

A second condition is that it must always remain possible to claim asylum in the Netherlands itself. In this respect, closer co-operation between European countries is desirable. It is also important that various weaknesses in the EU legislation be ironed out, such as the regulations concerning the so-called Dublin System, asylum seekers from safe countries and repatriation policy.83

<sup>77</sup>Kloosterman (2018). This is also the case in Germany (von Hermanni & Neumann, 2019).

<sup>78</sup>See, for example, Tyyskä et al. (2017).

<sup>79</sup>The ACVZ (2018: 80) has thought through various possible future scenarios. In all of these, better reception in the home region is key: "The creation and fnancing of high-quality reception facilities in the region (providing not only shelter but also adequate food, sanitary and medical provision, suffcient employment and, above all, good education for children and young people) has been identifed as a benefcial policy in all cases included in this study, regardless of which vision of migration is taken as the starting point." See also Hirsch Ballin (2019).

<sup>80</sup>See also Donner and den Heijer (2020).

<sup>81</sup>UNHCR, 2016.

<sup>82</sup> www.rijksoverheid.nl/onderwerpen/asielbeleid/vraag-en-antwoord/uitgaven-rijk-provinciesgemeenten-voor-opvang-asielzoekers

<sup>83</sup>The Dublin Convention aimed to prevent secondary asylum migration by making one member state responsible for considering an individual's application for asylum in the EU. Another issue it addresses is that of asylum seekers from so-called 'safe countries'. They have a right to remain until they have exhausted all legal remedies, which can take some considerable time and so has the potential to overburden the Dutch and other European asylum systems. Finally, there is repatriation policy. Responsibility for this still rests with the member states. They reach bilateral agreements with countries of origin, which may lead to inconsistencies. Donner and den Heijer (2020: 28) cite this example: "It is diffcult to explain why Spain and Germany, for example, can reach effective repatriation agreements with Morocco, but the Netherlands cannot." The government could also consider, again preferably in a broader European context, extending opportunities to claim asylum from abroad. See also Boeles (2017).

### *7.4.2 More Consideration of Absorption Capacity*

A second option for asylum policy is to give greater consideration to Dutch society's ability to incorporate these new migrants. Sudden peaks in asylum migration, in particular, often polarize communities. Although, incidentally, the opposite also happens: local authorities and residents are sometimes disappointed when reception centres that are well integrated within their communities close due to a decline in demand. The fuctuating nature of asylum migration entails a permanent process of upscaling and downscaling, whereby facilities and professional know-how repeatedly have to be restored and renewed.84

Just about everyone agrees that the government should consider absorption capacity as part of its approach to asylum policy. The question, though, is how.

In this respect we recommend examining the German experience with numerical targets (see Box 7.3). A number of considerations are relevant here. In the frst place, setting targets might open up explicit political debate about how many asylum seekers the Netherlands can accept each year. This would avoid giving the impression that the country has lost control of asylum migration and could thus bolster public support for accepting refugees.85 Secondly, it would assist with the processing of asylum claims and allow the creation of more permanent reception facilities. At present, every infux forces the Dutch government to improvise anew with regard to reception facilities, the processing of applications and housing allocations. Combined with a systematic policy of resettlement by invitation, taking in refugees would become less of a rollercoaster ride, with huge peaks and troughs, and the arrival and reception of refugees could be better anticipated.

### **Box 7.3: Germany: Establishing Numerical Targets**

In Germany, which had to deal with more than 1.2 million asylum claims in 2 years (2015 and 2016),86 a discussion arose about how much migration of this kind the country could handle given the economic and social effort it required. An answer to this question was formulated in the coalition agreement between the social-democratic SPD and the Christian-democratic CDU/ CSU alliance in 2018, which set a numerical target of 180,000–220,000 asylum migrants per year.87

(continued)

<sup>84</sup>Cf. ACVZ (2017).

<sup>85</sup>See, for example, Scheffer (2018).

<sup>86</sup>Source: Federal Offce for Migration and Refugees.

<sup>87</sup>*Koalitionsvertrag* CDU/CSU/SPD [CDU/CSU/SPD coalition agreement], (2018: 103).

### **Box 7.3** (continued)

This target is not a strict upper limit, but a policy objective with no legally binding effect. A rigid cap on refugee numbers would violate the German constitution, human rights treaties and EU law,88 as well as the Geneva Refugee Convention because it could lead to 'refoulement'.89

Despite this limitation, the actual number of asylum claims in Germany did remain below the target fgure in the frst 2 years after it was set, 2018 and 2019.90 The country took a variety of measures to achieve this, amongst them supporting the reception of refugees in their home region, expanding the list of so-called 'safe countries', committing to better protection of the EU's external borders and attempting to reduce the causes of fight.91

Once again, establishing and achieving a target of this kind would require improved European and international co-operation, particularly with regard to the adequate reception of refugees in their regions of origin. In addition, it might lead to increased efforts to tackle the root causes of asylum migration so that people do not embark on a dangerous journey to Europe in the frst place. But there is also a risk that could encourage a one-sided focus upon guarding Europe's external borders – a narrow interpretation we insistently warn against.92

### *7.4.3 Summary*

• The Netherlands can gain greater control over asylum migration by increasing the proportion of refugees invited to the country. Humanitarian determinants should always be paramount in their selection, but their 'integration potential' may also be considered.

<sup>88</sup>Wissenschaftliche Dienste des Deutscher Bundestages (2019).

<sup>89</sup>The principle of non-refoulement prohibits a country that receives asylum seekers or refugees from sending them back to a jurisdiction where they fear persecution on account of their race, religion, nationality, membership of a particular social group or political opinion. This basic principle is enshrined in Article 33 of the Refugee Convention.

<sup>90</sup>See: www.welt.de/politik/deutschland/article199137315/Migration-Obergrenze-fuer-Asylbewerberwird-nicht-ueberschritten.html and www.bundestag.de/presse/hib/676488-676488

<sup>91</sup>See, for example: Ilgit and Klotz (2018) and Bleiker (2017).

<sup>92</sup>See, for example, Ilgit and Klotz (2018).


### **7.5 Conclusion**

How can migration policy play a role in mitigating the negative effects of international migration on social cohesion in the Netherlands? That has been question at the heart of this chapter. In addressing it we have distinguished three main types of migration, each with its own underlying rationale: labour, family and asylum migration. We then went on to look at the policy scope available to the Netherlands within the European and international legal order. This revealed that the current supranational policy frameworks provide no room to impose additional requirements upon family migration. There are, however, more possibilities for European variations in labour migration and asylum policy.

Some of the options outlined in this chapter are worth exploring further. Briefy, they are as follows.


**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 license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license 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 8 Reassessing the Policy Agenda**

The Dutch government must pursue a more active policy to familiarize all new migrants with our society and to incorporate them into it as effectively as possible. That is the main message of this publication. In recent decades, policy in this area has been too variable and too reactive. An active policy is necessary because migration to the Netherlands is structural in nature. The Netherlands is now a dynamic migration society, attracting people from all parts of the world. As a result, its diversity by origin is increasing. In addition, we have to deal with more and more transient migration: many immigrants who come to the Netherlands are just 'passing through' and so eventually leave again.

### **8.1 New Migration Patterns Are Not Yet Suffciently Anchored in Policy**

Current policy does not take suffcient account of this new reality. It is still too much rooted in the world of yesteryear, when newcomers came mainly from the 'traditional' countries of origin: Turkey, Morocco, Suriname and the Dutch Caribbean. That is now history. Today's migrants come from a wide range of very different countries, such as the former Eastern Bloc nations, Syria, India or China. And they no longer settle only in the established multicultural districts in the major cities. Metropolitan suburbs, border towns, horticultural regions and expat communities also house many migrants, as do a variety of neighbourhoods in the major cities. In demographic forecasts, just about every future population scenario indicates that the number of people living in areas of high diversity by origin is set to increase. A development with considerable implications for all kinds of policy domains.1

<sup>1</sup>NIDI and CBS (2021).

<sup>©</sup> The Author(s) 2023

R. Jennissen et al., *Migration Diversity and Social Cohesion*, Research for Policy, https://doi.org/10.1007/978-3-031-14224-6\_8

Most new immigrants, and their offspring, fnd their way in Dutch society on their own. Their greater diversity by origin and shorter average length of stay also reduces the likelihood of social cleavage; there is less chance that a few relatively homogeneous groups of newcomers will fnd themselves pitted against equally homogeneous groups of settled residents.

Nevertheless, these new patterns require vigilance. They complicate conviviality in residential neighbourhoods, at schools and in voluntary associations. There is also a risk of fragmentation, with society breaking up into a constellation of smaller groups, each seeking to cling to its own identity and in so doing turning its back on the rest. Researchers presume that any problems of this kind will eventually resolve themselves, dissipating over a few decades or generations,2 but we believe that the government can *accelerate* this process through targeted policy*.* An acceleration that is essential given the systemic nature of international migration and the importance of maintaining social cohesion.

In our view, an active government policy of this kind should have three components.


In this chapter we frst formulate the core principles of such a systematic and proactive policy approach. We then fesh out the three policy components. In so doing, we have no illusions that that our proposals will dispel all concerns. Migration is one of the major social issues of our time, after all, and even in classic immigration countries it is a theme that divides politicians and citizens. The focus in this publication is the problems associated with the increasing diversity and transience of the migrants coming to the Netherlands, and the proposals we make are intended to help improve their reception into Dutch society and to strengthen the nation's social cohesion.

<sup>2</sup>See Putnam (2007). A similar point has been made about the positional improvement of people with a guest worker background: this can take two or three generations (Vermeulen & Penninx, 1994). For many children of asylum migrants, the process is faster due to their good educational performance (Hartgers et al., 2017).

### **8.2 Basic Principles: More Permanence, Coherence and Local Variation**

The Netherlands needs to recognize more explicitly that it has become a migration society with a high degree of diversity and transience. This requires a proactive policy approach with three key pillars, as discussed below.

### *8.2.1 From Ad-Hoc to Systematic Policy*

The systemic character of migration to the Netherlands requires systematic policies for the effective reception and integration of all immigrants rather than ad-hoc responses for the 'groupe du jour'. In classic immigration countries like Canada and Australia, such an approach is standard practice. Dutch migration and integration policy, by contrast, is characterized by considerable volatility (see Chap. 4). The past 60 years have repeatedly seen more or less improvised responses to new developments, each with their own emphasis. And there has been little coherence between migration and integration policy. This volatility resulted in part from evolving economic conditions, alternately creating and suppressing demand for labour migrants, and in part from an increase in family and asylum migration. But there also was a third contributing factor: the fact that the Netherlands failed to properly realize that it had become an immigration country and that this required a coherent migration and integration policy.

### *8.2.2 Consider Conviviality as Well as Integration*

This report has shown that a high degree of diversity by origin can undermine mutual trust and exacerbate feelings of insecurity or of not belonging, particularly at the local level.3 These effects are strongly linked to the diversity of a neighbourhood and have nothing to do with its socio-economic composition; we also fnd the same correlation in wealthier localities and amongst residents with a migrant background. That diversity, along with increased short-stay migration, requires an infrastructure that facilitates the harmonious conviviality of residents of the Netherlands. And policy to that end should be based upon *reciprocity*: it makes demands not only of newcomers, but also of their host society. In practice, this requires that government create the right institutional conditions for newcomers to be able to participate in Dutch society, that individuals, businesses, schools, community facilities and libraries be open to them and treat them fairly and that the newcomers themselves understand the feelings of settled residents and are persuaded to play an active part in society.

<sup>3</sup>Smeekes and Mulder (2016), Jennissen et al. (2018), Glas (2018), and Mulder (2018).

This does not imply that current integration policy should be abandoned. The issues of diversity, transience and conviviality we have highlighted in this report come on top of longer-standing questions around the sometimes diffcult integration of specifc groups of immigrants and their offspring. Core aspects of current policy thus remain relevant, especially the stimulation of labour-market participation by vulnerable immigrant groups. In the coming decades the Netherlands will continue to have to deal with groups alienated from the Dutch labour market and requiring support to enter it. Work, after all, not only provides an income, social contacts and self-respect; it is especially important because it connects people,4 a vital function in a society with a high degree of diversity. As the traditional immigration countries teach us, participation in the labour market is often the primary route to social integration and also helps build social cohesion.

### *8.2.3 Room for Local Variation*

Local authorities play a crucial role in the reception and integration of immigrants. The huge variety of migrants now in the Netherlands, in terms of both background and length of stay, has resulted in major demographic variations across the country. In this report we have distinguished eight types of municipality based upon the size and characteristics of the migrant groups they host. Horticultural districts like Westland, which attract mainly temporary labour migrants from Poland, face very different policy challenges from expat communities such as Amstelveen, with large groups of highly skilled newcomers from countries like Japan and India. Central government should therefore give local authorities the policy space they need to be able to respond effectively to these variations.

In so doing, though, the government still needs to provide appropriate fnancial, legal and substantive support. Local authorities should have suffcient resources to accommodate new groups and to facilitate their conviviality with non-migrants. Together with the Association of Dutch Municipalities (Vereniging van Nederlandse Gemeenten, VNG), the central government can also help them with know-how and examples of best practices.

### **8.3 Policy Recommendations**

In describing the three key pillars of a proactive policy approach to migration in more detail, we have focused upon three particular issues. This in turn leads us to three generic recommendations on how the government can manage the arrival of the very different types of migrant now coming to the Netherlands.

<sup>4</sup>NIDI and CBS (2021).


We have already elaborated on each of these in turn in Chaps. 5, 6, and 7. Below we recapitulate our most important specifc recommendations.

	- Increase local authorities' responsibility for the settlement of migrants.
	- Create reception facilities for all migrants.
	- Provide differentiated civic integration services for all migrants.

Every new immigrant to the Netherlands settles in a particular community, and the local authority there is the institution best placed to introduce them to Dutch society and facilitate their participation in it. These bodies therefore need to think about creating more permanent facilities to help all arriving migrants settle in. Sometimes, however, they have only a limited insight into who is coming and how long they stay. Not until it becomes apparent after some time that a specifc group has settled in their area, and problems have perhaps already arisen, do they make appropriate provision. And with some communities now seeing new migrant groups arrive on constant basis, more is needed than ad-hoc facilities for the 'groupe du jour'.

### *8.3.1 Increase Local Responsibility for Settlement*

Local authorities need to think more carefully about which immigrants best suit their communities. They can exert some infuence over this by reaching agreements with employment agencies, employers, universities and the Central Agency for the Reception of Asylum Seekers (Centraal Orgaan opvang Asielzoekers, COA), as well as by establishing facilities tailored to certain groups. In the case of asylum migration, central government could allow them some freedom to select the background of the refugee status holders settled in their areas – although it is essential in this respect to maintain even dispersal across host communities. Local authorities should also be given greater scope to make their own arrangements for the repatriation of specifc groups, such as unemployed labour migrants. They can facilitate the departure of people with poor future prospects by co-operating with migrant selfhelp organizations and through agreements with employment agencies and employers.

### *8.3.2 Create Reception Facilities for All Migrants*

The government needs to create systemic provision for the reception of new immigrants. Again, this is primarily a task for local authorities; they should establish more *permanent* facilities that help *all incoming groups*, not just asylum migrants or highly skilled workers but also labour, student and family migrants, to familiarize themselves with Dutch society. This would do away with the repeated need to develop ad-hoc provision to deal with sudden infuxes of asylum or labour migrants. To this end it may be useful to set up a local or regional reception centre for all migrants settling legally in a particular area. One option here is to broaden the scope of existing expat centres, which tend to be rather exclusive. The new-style reception centres can help all newcomers, taking into account differences in their legal status, with housing, schools, healthcare, language tuition, sports facilities and organized activities. Together with regional training colleges (ROCS) and employers, local authorities can also provide guidance on work opportunities in their area. Finally, the reception centre can connect newcomers with civil society organizations, other local residents and entrepreneurs.

There should also be a systemic response to temporary migration, in particular with regard to housing and education. Especially in the densely populated Randstad conurbation, housing for all groups is in short supply and overcrowding is commonplace. Local authorities need to tackle illegal rental practices by unscrupulous landlords and employment agencies, and prevent temporary workers who lose their jobs from fnding themselves on the street. Meanwhile, schools sometimes have to deal with unexpected spikes in pupil numbers, high rates of turnover and intake and outfow at irregular times. Regional co-operation and professionalization are needed to guarantee the accessibility and quality of education for newcomers.

### *8.3.3 Provide Differentiated Civic Integration Services for All*

A civic integration policy designed primarily as a path to Dutch citizenship for permanent migrants no longer refects the reality of the contemporary migration society. Large groups are now staying only temporarily in the Netherlands, whilst many who do remain longer have no desire to become Dutch nationals or only decide to do so after years settled here. Nevertheless, it is vitally important for social cohesion that all these groups be incorporated into Dutch society. This can be done by offering language tuition, by introducing them to the history and culture of the Netherlands and by encouraging their participation in organized community activities. The government should therefore create civic integration programmes for all migrants, including those whose stay is only temporary and others for whom the process is not mandatory.

The nature of these programmes will vary according to legal status and individual needs. If they want to stay longer in the Netherlands, for instance, many migrant workers from other EU member states need to learn Dutch. Highly skilled migrants also want to know more about the country's culture and participate in local social life. Parents require assistance navigating the education system. And labour migrants need help with the many forms they have to fll in. All this calls for a varied portfolio of language and other civic integration services, tailored to the considerable diversity of today's migrant population. The same applies to fnancing. For asylum migrants, civic integration is compulsory and heavily subsidized. For highly skilled newcomers, on the other hand, the process is voluntary but requires substantial personal investment. For labour migrants it is also voluntary, but the employer may be asked to contribute towards the costs.

	- Provide good physical and social infrastructure at neighbourhood level.
	- Strengthen intercultural skills in education and other parts of the public sector.
	- Enforce the basic rules of conviviality.

Increased diversity by origin and greater transience are straining social cohesion, especially at the local level. The more diverse a neighbourhood, the less its residents know and recognize each other and the less they feel at home there. At schools and voluntary associations, a higher turnover of pupils or members makes it harder to build long-term relationships. Yet at the same time Netherlands Institute for Social Research (*Sociaal en Cultureel Planbureau,* SCP) surveys of the national mood repeatedly show a great need for social cohesion. The government should therefore pursue an active policy aimed at strengthening social cohesion. Again, this is also primarily a task for local authorities.

### *8.3.4 Provide Good Physical and Social Infrastructure*

Local authorities can promote social cohesion in different ways. In recent decades, all kinds of small-scale local initiatives have been launched to encourage social contact between residents from different ethnic groups. Due to the near total absence of good impact studies on these efforts, however, it is diffcult to separate the wheat from the chaff. This applies not only to the Netherlands, but also to many other Western countries.5

There is evidence supporting the importance of good confguration of the physical environment, though. Safe, clean public space is essential for social safety and for social interaction in general. Once this basic condition has been met, the local

<sup>5</sup>Engbersen and Scholten (2018).

authority can encourage people to meet each other regularly in that space. This promotes a certain degree of public familiarity. Recognition from earlier encounters, however superfcial, can contribute towards feeling at home, safe and connected. If its confguration invites people to linger in public space, there is a greater chance that residents will meet each other. Parks, public gardens and playgrounds play an important role in this respect.

There is a considerable body of research showing that a rich social infrastructure helps increase cohesion. Semi-public amenities like playing felds, local shops, libraries and community centres strengthen the social resilience of a neighbourhood. In the Netherlands, housing corporations are key players in this respect. They should be given greater scope to undertake small-scale activities in support of liveability and cohesion. In highly diverse environments, local authorities should invest more in professional community work.

Increasing diversity and transience put voluntary associations under pressure, too. Consequently, local authorities should not expect too much of this sector as a source of social cohesion. Very diverse clubs, societies and associations tend to experience high member turnover and may even need support to stay afoat. In areas where many migrants stay only temporarily, such as expat or horticultural communities, it might be worth investing in more informal groups. And sports clubs could think about arranging 'casual' leagues for expats and migrant workers, which are easy to join or leave as a team or a player.

### *8.3.5 Strengthen Intercultural Skills in Education and Other Parts of the Public Sector*

Likewise, increasing diversity and transience have major repercussions for highcontact professions in the semi-public sector. They require distinct skills and make considerable demands of those working in the 'front line' of education, care and public housing, and not just in the big cities. Education, in particular, can serve as a laboratory for the development of institutional approaches and professional abilities to deal with diversity.

From the scientifc literature, it seems that an intercultural approach works best at schools with pupils from a variety of backgrounds. When children feel seen and recognized at school, they have a greater sense of belonging there and perform better. This means that schools need to create scope for a certain degree of *cultural familiarity*. Within the regular curriculum, for example, subjects such as history and geography could address pupils' various backgrounds and origins as a matter of course. An intercultural approach also takes into account the cultural backgrounds of the more long-established groups in society, and thus helps counter prejudice and cultural myopia in all directions.

In addition, schools need to prepare themselves for a pupil population that is evolving constantly in terms of its cultural diversity and have to be able to incorporate new groups smoothly. When the Netherlands had only a few migrant communities,

specifc expertise and dedicated policies could be developed for each of them. With dozens of different groups from all parts of the world, that is impossible. This also means that schools and teachers alike need knowledge and skills to cope with pupils of different origins and with highly diverse classes; these should be an integral part of their professional portfolio, so acquiring them should be a standard aspect of teacher and school-leader training. It is up to central government to ensure that these professionals and their schools possess the right know-how and resources.

### *8.3.6 Enforce the Basic Rules of Conviviality*

Thirdly, increased diversity by origin, educational attainment and social background means that the differences between established groups and newcomers at the national level are less clear-cut now than they were in the past. No longer are a few relatively homogeneous groups with a migrant background pitted against an equally homogeneous 'indigenous' community.

Nevertheless, it is wise to consider how to promote cohesion at national, city and town levels. Without that, there is a risk of fragmentation, of society breaking up into a constellation of smaller groups that seclude themselves from each other. Moreover, public debate about national identity has not been helpful in promoting cohesion; in the Netherlands, and in neighbouring countries as well, these have proven more divisive than unifying. By contrast, urban and regional identities offer a more promising route. A sense of belonging to a neighbourhood, a city or a region can be acquired by going to school there, supporting the local football club or picking up its accent. We see that many residents with a migrant background develop local allegiances earlier than national ones, and that in the second generation these are even stronger than ethnic identities.

The national government's main role in this respect is to enforce the basic ground rules of conviviality: everyone living in the Netherlands should respect its constitutional democracy and contribute towards society, whilst at the same time being allowed space to maintain their own traditions and symbols as long as this is in keeping with the rule of law and does not hinder participation.

Finally, the government can promote social cohesion by propagating unifying 'national stories' and institutions and by ensuring that newcomers engage with them. Examples include speaking the Dutch language, the struggle for a safe physical environment and the tradition of co-operation, compromise and consensus. Upholding these is quite different from establishing a national identity, which tends to engender a polarizing discourse and so fails to bond people. Emphasizing unifying or overarching goals and focusing upon things everyone living in the Netherlands has in common can promote more positive intergroup relations. After all, these general institutions are not linked to any one group: they are accessible to all, can forge links and offer continuity from the past into the present and future.

Dutch migration policy is a balancing act between various disparate interests and considerations, such as economic development and humanitarian principles like

### 3. **Make migration policy more conducive to social cohesion**


respect for private and family life and protection from persecution. Governed by different legal frameworks, these interests and principles, which are governed by different legal frameworks, can be linked to three distinct types of migration: respectively, labour, family and asylum migration. When it comes to family migration, it is not possible to impose additional requirements within the existing legal parameters because of the restrictions imposed by the EU Family Reunifcation Directive. In the case of labour migration from outside the EU/EFTA zone and asylum migration, however, there is some scope for policy adjustments.

We therefore recommend exploring a number of new directions for migration policy.

### *8.3.7 Ensure that Labour Migration from Outside the EU/ EFTA Zone Is Complementary and Circular*

Government policy for labour migration from outside the EU/EFTA zone should aim frst and foremost to ensure that this benefts the real incomes of those already living in the Netherlands. This occurs when migrants play a complementary role in the labour market.

In this respect, distribution issues also need to be considered. For example, the economic benefts of labour migration often accrue to employers whilst leaving wider society facing potential social problems. These result in costs to society that reduce real per-capita income.

Problems of conviviality can be reduced if labour migration is temporary in nature. Central governments would therefore be wise to continue promoting circularity in this domain, learning from mistakes made in the past.

### *8.3.8 Take Account of Society's Capacity to Incorporate Asylum Migrants*

Asylum migration is one of the main drivers of the increasing diversity of the Dutch population. This is due to its multitude of countries of origin, many with no signifcant previous history of migration to the Netherlands. We outline two possible options whereby Dutch asylum policy can remain tailored to its core objective of offering refugees a safe haven whilst at the time considering issues of conviviality.

The frst is for the Netherlands to gain greater control over the refugees it receives by increasing the proportion invited to the country. This would give it a greater say in who exactly comes. Moreover, a broader invitation-based policy would allow a quicker response to sudden crises such as those involving the Yazidis in the Sinjar Mountains or at the Mória camp on Lesbos.

When inviting refugees, humanitarian determinants always remain paramount. But within that framework, factors related to 'integration potential' can also be taken into consideration. For example, compatibility with existing migrant networks in the Netherlands. This should accelerate incorporation into Dutch society and might also enable better matching of arriving refugees to the characteristics and needs of the communities they are resettled in.

A second option is to develop a systemic policy taking into account Dutch society's capacity to incorporate asylum migrants. The German experience with numerical targets may be useful here.

Under both of these options, the policy's success is highly dependent upon improved European and international co-operation to ensure the adequate reception of refugees in their region of origin and shared responsibility for their migration to safe third countries.

### **8.4 Concluding Remarks**

### *8.4.1 A Greater Role for Local Authorities*

In this study we have argued for government to take on a proactive role in order to guide international migration in the right direction and to help the multitude of groups living in the Netherlands living together harmoniously. To achieve this, different branches of government need to assume distinct responsibilities.

*Central government* takes the lead when it comes to shaping migration policy, and also has a key supervisory role. For example, it oversees enforcement of the basic rules of conviviality, including those to combat discrimination. With regard to civic integration, its remit needs to be strengthened; activities here are still not properly regulated, allowing abuses by unscrupulous language schools and reintegration agencies. National frameworks will also remain necessary for the dispersal of asylum seekers and the organization of compulsory civic integration programmes. Above all, though, national government has to create the right fnancial and legal conditions for local policy in areas such as education, integration, access to the jobs market and the social domain. Many of these have been subject to major cutbacks in recent decades, and a great deal of expertise has been lost. That needs to be restored and renewed, with the appropriate resources.

*Local authorities* face major challenges with regard to strengthening cohesion and helping large numbers of people fnd work. We advocate a reassessment of their existing responsibilities, particularly in the area of civic integration. Maintaining the physical environment and social domain has also traditionally been a task for local government. It is crucial to create an infrastructure, supported by civic society organizations, individuals and companies, that facilitates conviviality But local authorities should also assume additional responsibilities for shaping reception policy and – to an extent – repatriation policy.

### *8.4.2 Employers Also Have Responsibilities*

Government lacks the resources, the knowledge and the authority to manage migration on its own. As in traditional immigration countries, a whole range of other parties need to engage with this issue – the private sector in particular. This applies frst and foremost to *employers*; they beneft directly from labour migration to the Netherlands, after all, be that by production workers or highly skilled professionals, so they are also responsible for its wider impact. For example, they should contribute towards local reception facilities and the fnancing of differentiated civic integration programmes. Or organize basic forms of such provision themselves. They should also make sure that temporary labour migrants are housed in decent accommodation.

As for the *migrants* themselves, Dutch society is entitled to expect them to make an effort to speak the language, to integrate and participate actively. Everyone in the Netherlands, migrant or otherwise, has a civic responsibility. For example, they sustain the nation's physical and social infrastructure by taking part in community and neighbourhood projects, by volunteering and by joining clubs, societies and associations. The task for government in this domain is to support the multitude of civic initiatives currently promoting integration, liveability, sports, culture and neighbourhood management, as well as trying to instigate such activities where they do not already exist. This is the core challenge of contemporary community work in a migration society. Finally, there is also major role here for *public institutions* like schools, housing associations and care providers, not to mention the increasing number of social enterprises active in the felds of integration and conviviality.

### *8.4.3 Focus Upon Everyday Diversity*

Ultimately, this is all about how we in in the Netherlands want to live together. The motto of the US migrant society is 'E pluribus unum': out of many, one. Although the reality is often different, this phrase retains a symbolic unifying function. For the Netherlands as a migrant society, unifying national institutions such as language, the struggle against water and the tradition of co-operation, compromise and consensus are also vitally important. A robust society knows how to propagate what people have in common and what binds them together. In this way it contributes towards the emergence of 'everyday diversity'.6

By this we mean that people living in the Netherlands consider diversity not as something exceptional but as a natural reality embedded in their day-to-day behaviour at school, in shops, in public spaces and at work. Everyday diversity ensures that we are able to live together harmoniously rather than in confict. And that everyone – transients and settlers, newcomers and lifetime residents – can feel at home in the migrant society.

**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 license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license 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.

<sup>6</sup>Wessendorf (2010, 2014) and Ignatieff (2017).

### **Bibliography**


R. Jennissen et al., *Migration Diversity and Social Cohesion*, Research for Policy, https://doi.org/10.1007/978-3-031-14224-6

<sup>©</sup> The Editor(s) (if applicable) and The Author(s) 2023


*Sportverenigingen in Nederland. Veerkrachtige verbanden voor sport* (pp. 208–225). Mulier Instituut.


Fettelaar, D., & Smeets, E. (2013). *Mogelijke indicatoren van schoolgewichten*. ITS.


### **Index**

### **A**

Absorption capacity, 83, 152–153


### **B**

Border communities, 6, 39, 56, 77

#### **C**


Civic integration, 7, 35, 69–71, 73, 75, 79–100, 127, 142, 146, 147, 159–161, 165, 166


#### **D**

Departure, 3, 7, 34, 53, 79, 80, 96–100, 121, 159 Disadvantages, 12, 60, 62, 65–69, 86

#### **E**

Education, 2, 7, 11, 51, 60, 63, 64, 66–68, 70, 74, 80, 81, 84, 85, 88, 91–94, 97, 99, 100, 114–119, 122, 127–129, 133, 141, 142, 148, 160–163, 166 Emigration, 10, 17, 27–30, 32, 89, 139 Employers, 7, 54, 63, 74, 80, 81, 84, 87, 88, 90, 91, 93–95, 98–100, 138, 142, 159–161, 164, 166 Ethnic diversity, 2–4, 24–33, 43, 45, 46, 50, 83, 117, 118, 135 Everyday diversity, 167 Expat communities, 7, 39, 55, 77, 81, 84, 109, 155, 158

#### **F**

Family migration, 20, 26, 73–75, 136, 143–146, 148, 154, 164 Feeling at home, 49, 108, 132, 162 Feeling of loss, 43, 46, 48 Feeling unsafe, 51 Forecast, 23, 24, 32–34, 91, 155

### **G**

Geolinguistic classifcation, 14 Guest workers, 4, 20, 26, 28, 39, 54, 60, 62, 63, 69, 125, 138–140, 156

### **H**

Herfndahl-Hirschman Index (HHI), 13, 37, 45 Highly skilled migrants, 2, 22, 39, 68, 72, 73, 75, 81, 84, 87, 88, 91, 93, 94, 125, 126, 137, 140–143, 161

© The Editor(s) (if applicable) and The Author(s) 2023

R. Jennissen et al., *Migration Diversity and Social Cohesion*, Research for Policy, https://doi.org/10.1007/978-3-031-14224-6

Homogeneous communities, 39, 45, 56 Horticultural districts, 6, 39, 55–56, 77,

81, 91, 158 Housing, 7, 9, 52, 60–63, 67, 70, 74, 80, 81, 84–86, 88–94, 98, 100, 103, 107, 110, 113, 114, 133, 142, 152, 160, 162, 166

### **I**


### **L**


#### **M**

	- 144, 145, 157, 163

Migration policies, 3, 9, 14, 72–76, 78, 135–154, 156, 159, 163–165 Motive for migration, 11, 35, 36, 136 Multiculturalism, 8, 29, 60, 61, 64–72, 77, 104, 105, 115, 116

### **N**

National cohesion, 123–132 National identity, 105, 128–129, 132, 133, 163 Natural population growth, 31–32 Neighbourhood cohesion, 46–48 Net migration, 3, 4, 27, 29–30

Numerical targets, 152, 154, 165

### **O**

Overlapping cleavages, 123–126

### **P**


#### **R**

Reception, 2, 3, 7, 11, 14, 20, 57, 77–80, 82–89, 91, 99, 100, 127, 151–153, 156–160, 165, 166

Residence, 11, 19, 50, 52, 55, 68, 73, 74, 89–92, 94, 112, 144, 150

### **S**

Schools, 2, 7, 9, 33, 34, 48, 52–57, 64–68, 82, 88, 91, 92, 94, 99, 101, 103, 108, 110, 114–124, 128, 129, 132, 133, 156, 157, 160–163, 165–167


Index

Social infrastructure, 103, 113, 133, 161–162, 166 Social safety, 46, 49–51, 106, 110, 133, 161 Sociocultural, 60, 65, 67, 69–71, 97 Socio-economic, 5, 8, 9, 44, 47, 50, 60, 62, 65, 67, 69, 75, 88, 96, 97, 124, 125, 140, 146, 157 Student migration, 24, 88

### **T**

Temporary stays, 89–92

Transience, 3, 4, 8–10, 33–37, 40, 41, 52, 56, 57, 80, 92, 101–103, 120, 123, 124, 132, 156–158, 161, 162

### **U**

Unifying stories, 103, 129–132

#### **V**

Voluntary associations, 9, 33, 52, 57, 101, 103, 114–123, 129, 132, 133, 156, 161, 162