Understanding “what” people think about European Integration: A Qmethodological Study of Dutch’ Discourses on the Freedom of Movement of Persons

(1)

Understanding “what” people think about European Integration

A Qmethodological Study of Dutch’ Discourses on the Freedom of

Movement of Persons

By Brigit de Kleer January 2019

A thesis submitted for the fulfilment for the degree of Master in Public Administration At Leiden University, in the Faculty of Governance and Global Affairs.

(2)

Preface

During both my bachelor and master study, I have learnt more and more the need to approach academics from a wide perspective. I have done a major in sociology and political science during my bachelor and during my master focused on public administration. I therefore fully believe those are linked, and the free movement of persons is a great example of this. To understand how European integration has evolved, it is important to have an understanding of politics, of the European Union and its multi-level nature, but also of the people that comprise the EU and their attitudes towards it. To approach one topic from so many points of view can be challenging, but is interesting and satisfying work. I hope reading this thesis will equally challenge and interest you.

As a thesis is rarely written without support, neither is mine. I would, therefore, like to thank prof. Steunenberg, who has not only introduced me to the interesting work of Dryzek and his framework of discourse analysis, but also kindly guided me since, always providing clear and thought-through feedback that truly helped improve my work. I also want to think Elitsa Kortenska, who, especially in the beginning of writing my thesis helped greatly with

explaining the framework and deciding on important methodological issues. Similar thanks go to my colleagues in my capstone group who I greatly appreciate for brainstorming as well as making our sessions more lively. Lastly, I owe thanks to my friends and family, who have put up with me without (too much) complaints when I was once again stressing about this thesis.

(5)

Abstract

As questions on European integration become more salient, so does the importance of considering public opinion and its influence on this process. This is stressed by post

functional theory and recently more studies focus on understanding why people exhibit pro-integration or anti-pro-integration attitudes. However, these theories forget to consider the content (the “what”) of those attitudes and as such miss an important opportunity to clarify their explanations.

This thesis attempts to unravel the subjective nature of attitudes towards European Integration by analysing the discourses of people surrounding the freedom of movement of persons policy of the European Union. Using Dryzek’s adaptation of Qmethodology I answer the question:

What kind of discourses on the freedom of movement can be identified in the population of the Netherlands?

I find three discourses: the two anti- and pro-integration attitudes as the literature suggests exist and another one that weighs moral values on migrants with personal interests. Examining the content of attitudes towards European integration, has therefore led to new insights on what considerations are influential in determining attitudes. Understanding “what” people think on European integrations may therefore just as important as understanding “why” they think this.

(6)

Introduction

Already quite a lot of literature is written on attitudes on European Integration. These studies either theorize whether and why people are positive towards European integration or not (and often link this to background variables such as education) or they focus on empirically testing those theories using large databases such as the Eurobarometer (see chapter 2). However, by only determining which variables coincide, these studies forget to ask the question what those attitudes towards the EU are comprised of and thereby miss an important element of EU integration studies. It is therefore important to understand the what (or the dependent variable) in order to better understand the why (the independent variables or the causes).

This thesis attempts to unravel the subjective nature of attitudes towards European Integration by analysing the discourses of people surrounding the freedom of movement of persons policy of the European Union. The freedom of movement of persons is well suited to study attitudes towards European Integration as it is both highly contested, and combines ideas on the European Union with concerns of national identity and migration. As attitudes towards migration are usually similar to attitudes towards the EU as a whole (De Vreese &

Boomgaarden, 2005; Toshkov & Kortenska, 2015) the freedom of movement of persons is particularly interesting to study with regards to European integration.

As the main idea of this thesis is to gain a better understanding of the content of attitudes towards European Integration, focusing on the freedom of movement, the research question remains broad:

What kind of discourses on the freedom of movement can be identified in the population of the Netherlands?

To answer that question I will use Dryzek’s adaptation of Qmethodology as a methodological framework. This allows the limitation of the influence of the researcher on the data, thereby enabling a kind of “bottom up approach. ” Thereby I will be able to find new discourses even if they are not predicted to be there based on the academic literature. A detailed account of this method and how I used this in my research can be found in the methodology section of this thesis (chapter 4).

(7)

Before that, I will explain the freedom of movement policy in relation to European Integration (Chapter 1), and provide an overview of the main theories on European Integration and show that the content of attitudes is not sufficiently considered in those theories (Chapter two). I then go on to explain the Dutch cultural background of migration and migration policy and suggest what –based on the theories and the surveys on Dutch attitudes towards migration and European Integration, can be expected to be found in the study. After the method section, I will analyse the result and link them to my literature (chapter 5). In my discussion I do not only focus on further research suggestions, but I also compare my study to a similar study done recently by Sluis (2018). Although he did not find exactly the same results, both our studies suggest that attitudes towards European Integration are more complicated than simply pro- or anti- integration. As such, the contents rather than the causes of views towards

(8)

Chapter 1: Free movement of Persons

1.1. Introduction

As explained in the previous section, the aim of this study is to understand more precisely what the views of Dutch citizens towards the free movement of persons are comprised of. I have chosen this topic because the free movement of persons is vital to the functioning of the European Union. Furthermore, the free movement of persons is often contested, specifically in recent years due to the migration crisis and surrounding the debate around the Brexit (Hobolt, 2016). This makes it well suited for a discourse analysis as described in chapter 4. This chapter will give some basic information on the background of the free movement of persons. Furthermore, I will explain the Schengen area and its relation to the free movement of persons and to European Integration.

1.2. The establishment of the free movement of persons

In 1951, the Economic Coal and Steal Community was established (ECSC) and for the first time in history, an organisation was (partly) organised at the supranational level. Then, in the decades following the establishment of the ECSC, the European cooperation expanded rapidly (Lelieveldt & Princen, 2015). In 1957 the treaty of Rome was signed, forming two new

communities: the European Economic Community and the European Atomic Energy

Community (Euratom). The single European Act from 1986 detailed plans to create a single market and establish the four European Freedoms (freedom of movement of goods, services, people and capital) and six years later with the Treaty of Maastricht, the European Union was founded. In short, the free movement of persons is part of a fast integrative process, that characterised the European community in its early years (Craig & De Búrca, 2015; Lelieveldt & Princen, 2015).

The free movement of persons is often considered to be at the core of the functioning of the European Union. Already in 1957, this right was incorporated in the treaty of Rome with the establishment of the European Economic Community (European Council, 1957). In this treaty in Article 3 already, the objective of achieving free movement of people is stressed:

For the purposes set out in Article 2, the activities of the Community shall include, […] in accordance with the timetable set out therein: […]

(c) the abolition, as between Member States, of obstacles to freedom of movement for persons, services and capital;

(9)

The free movement of workers is described in more detail in Article 48 of the treaty of Rome and is currently included in the Treaty on the European Union. This freedom is not the same as the Schengen agreement (which I will discuss later in this chapter) as barriers to the free movement are eliminated, but internal European borders remain intact.

At the creation of the European Coal and Steel Community, the goal of its initiators was clear: to advance European cooperation to such an extent that war within Europe was not only “unthinkable, but also materially impossible” (Schuman 1957 as cited in Iglesias, 2016, p. 56). Furthering European Integration was not only perceived to inevitable but also desirable and necessary, to such an extent even that the European Coal and Steel Community was considered a stepping stone for achieving a European Federation (European Parliament, 2018). The four freedoms were important in achieving such a union or the very least an efficient economic community as it eliminated important barriers to trade (Craig & De Búrca, 2015).

In his book the passage of Europe, Luuk van Middelaar (2013) explains how the free

movement of persons is not only a step in the furthering of European Integration, but was also meant as a way to legitimize it. According to van Middelaar, the expansion of the freedom of movement (first it only applied to mineworkers and steal workers) was enabled by the idea of creating legitimacy by creating citizens’ right (the so-called “Roman” strategy). The argument is that as war had started to fade from the public’s mind, the EU needed a way to legitimatize its existence beyond the EU’s function of maintaining peace. Allowing, the free movement of workers that should be treated equally to the citizens of the country they resided in, created rights that were clearly attributable to the European Community and as such were thought to increase its legitimacy.

However favourable to some this new right may have been, at its start it has been contested, mostly by the political elites. During the founding of the ECSC, six years before the treaty of Rome, Italy wished to include the freedom of movement of workers into the Schuman plan, in order to make it easier for its unemployed citizens to find work elsewhere and to ensure the rights of Italians already working abroad (Van Middelaar, 2013). However, Belgium and France were afraid that this right would lead to a large influx of Italian workers and increased competition for their own citizens. As the success of the Schuman plan was a high political priority for both countries, a consensus was reached. The free movement of highly skilled steal workers and coalminers was introduced, to be overseen by the High Authority (the

(10)

supranational body of the community). For national leaders whether to be in favour of the free movement of persons, thus depended according to van Middelaar (2013), on the specific socio-economic condition of that nation-state. Italy, dealing with high unemployment rates at that time, favoured a European policy that not only allowed its citizens to find work

elsewhere, but that also protected those who would or already did. Countries that already hosted a rather large amount of external workers and could not use more, would be more likely to oppose the freedom of movement as it would increase competition for their own workers (Van Middelaar, 2013).

Similar considerations apply to the level of European citizens. Van Middelaar suggests that the nature of the Roman strategy of the creation of rights for European citizens is inherently contradictory in its nature. This is because whereas workers living in low capital countries can benefit from the right to freely reside and work in another country, the majority of European citizens stay in the country they are citizens of and as such do not benefit from this right (Van Middelaar, 2013). Moreover, as explained above the freedom of movement leads to an

increase in competition. The Roman strategy of giving citizens the right to work and reside in different countries, is most likely to be considered favourable by companies benefitting from this as well as workers wishing to obtain a job outside their own country. The majority of people, therefore would according to this logic either be indifferent or opposed to the freedom of movement.

Vasilopoulou & Talvin (2018) tested this idea using data from the Eurobarometer by

investigating whether citizens from high capital countries were more likely to oppose the free movement of persons then those from poorer countries. This, they argued was to be expected as, especially since the expansion of the Union in the 2000s, intra-EU migration has for its greater part existed from poorer to richer EU countries as such increasing competition in the latter. This expectation was confirmed in their study. While support for the freedom of movement of persons was considered high in all EU memberstates (82.4%), those people in high capital countries opposed the freedom significantly more than others. For example, in the UK 27.3% said to oppose the right, whereas in Spain only 5.8% did. The premise of van Middelaar thus seems to hold here. Therefore, in a sense, the free movement of persons can be considered double sided: on the one hand it represents and seeks to further European

integration, but on the other hand it seems to create a sentiment against that same integration ideal. The above explanations of these contestations around the freedom of movement have focused on macro-economic arguments. Another explanation relates to the concepts of

(11)

borders, nation hood and national identity and is best explained by considering a further integrative step then the free movement of persons: the establishment of the Schengen area.

1.3. Schengen under siege

As mentioned, the Schengen area is different from the principle of free movement of people, workers, services and goods. In fact, the Schengen area opens the borders within the area, diminishing the border controls, whereas the freedom of movement eliminates barriers to trade. For instance, the UK currently has (although it is unclear for how long) national borders and passport controls and thus takes no part in the Schengen agreement, but is part of the Single European Market and as such the free movement of goods, capital, services and persons is applicable to the UK. This means for instance that the UK is not allowed to discriminate workers outside of the public sector on basis of their nationality. As borders have, in the past, been considered vital to the definition of a nation-state (Walters, 2002), Schengen can be considered a far bigger integrative step than the freedom of movement. As such its introduction was even more constrained than that of the four freedoms.

When the Schengen agreement was introduced in 1985, only 5 out of the 10 countries that were part of the EEC – West Germany, France, The Netherlands, Belgium, and Luxembourg – signed the agreement. This first agreement was only a preliminary one and concerned the measures required to be taken to achieve a “borderless” area within the EU (Walters, 2002). Schengen was therefore established by a voluntary (some may say intergovernmental)

scheme, that focused on collaboration and allowed countries to opt-out(Lelieveldt & Princen, 2015). This challenged the notion of the EU integration as a single Union into a single union as Schuman had intended the European Project to be. Rather, patches of countries exhibit different levels and intensity of integration.

However, even though the establishment of Schengen was perhaps not as Schumann intended, it did lead to the furthering of European Integration as it has been formalised. With the treaty of Amsterdam, Schengen was indeed incorporated in the Treaty on the Functioning of the European Union and nearly all member states (except the UK and Ireland) are now part of the Schengen zone as well as non-EU member states Norway and Iceland. For this reason, the Schengen treaty is often described as a “laboratory” that allowed an experiment to lead to further European Integration (European Parliament, 2018; Lelieveldt & Princen, 2015). However, more recently Schengen has been criticised by others than just elites seeking to further regional integration. In 2015, the migration flow over the Aegean Sea drastically

(12)

increased, resulting in a crisis of the Schengen project (Schimmelfennig, 2018). During its peak in the last quarter of 2015, Frontex detected nearly a million illegal border crossings at the EU external borders (Frontex, 2016). Since the second quarter of 2016, after the

implementation of the EU Turkey deal, this number has dropped drastically and the numbers have continued to decline (Frontex, 2016, 2018). However, both during and after the main crisis, the union seemed unable to divide the migrants over the member states. Division schemes were not implemented properly in all member states, and some states (e.g. Hungary) within the Schengen Zone even reinstated their borders, thereby endangering the open

borders’ project (Börzel & Risse, 2018). The EU was unable to deal with the migration crisis (as opposed to e.g. the Eurocrisis that did

lead to more integration), because the issue was highly politicised (Börzel & Risse, 2018; Schimmelfennig, 2018). Walters (2002) argues that the establishment of the Schengen is a reconstruction of the idea of borders. As such the external threat to the EU is no longer (geopolitically) seen as nations, but rather “the security threat takes the form of a host of

transnational, social threats, and is often personified in the racialized figure of Islamic and nonwhite people (Walters, 2002, p. 570).” This reconceptualization of border threat lends

itself to and is reinforced by elite and public discourses around refugees as a threat. According to Schmidt (2010), national – populist- parties were able to use/frame this discourse of

transnational threat to mobilize citizens with exclusive nationalists identities (see chapter 2 for an explanation of this).

As I will explain in the following chapter, theories on integration suggest that the more visible an issue becomes to the public, the more likely it is that the public mobilizes and restricts the ability of political elites to further integration independently of the public. Some (e.g. Börzel & Risse, 2018) argue that this (partly) explains why the EU is unable to deal with the

Schengen crisis as one united union. The following chapter will explain more about these theories.

(13)

Chapter 2: Theories on European Integration

With the EU being such a unique organisation and growing so rapidly in its initial year, academic interest in the European Union grew and theories were established to explain this rapid growth. Considering my research question1, the stream of literature focusing on

European Integration is most relevant to consider, as these theories investigate how and why power from sovereign countries, is (or is not) transferred to the European Union. Early theories of integration, stem from the field of international relations (Lelieveldt & Princen, 2015). Therefore in those theories, the entities under examination are always the national states or the elites leading those states. The most important classic integration theories are (neo)functionalism and intergovernmentalism. A more recent theory, postfunctionalism, also includes the influence of the public on European Integration and it is this theory I will be focusing on.

With the recent growth (or visibility) of Euroscepticism and the unwillingness of member states to further integrate on particular policy areas (such as immigration), the exclusion of the public in explaining European (dis)integration is no longer feasible. Postfunctionalism focuses on how public scrutiny can constrain integration and is thus better adapt at explaining recent developments. There are also those theories that aim to explain how attitudes towards integration are determined (economic interest and identity threat). These theories can be embedded within the theoretical framework of postfunctionalism. The issue with those theories is that they only study the link between public opinion and European integration, or focus on what determines whether someone has a pro- or anti migration/integration attitude. They never stop to consider that the content of attitudes towards European integration may be more complicated than simple pro- or anti integration and that the considerations of identity, party positioning and communication, and economic concerns can matter differently to different people. This is a missed opportunity as understanding the content of attitudes on European integration can shed more light on when which concerns matter most and what can be done to change attitudes towards European integration. My research is intended as both showing the importance of considering the content of attitudes towards EU integration, and as a first step towards understanding these contents better.

1

(14)

2.1.2. The predecessors of post-functionalism

Post functionalism can be seen as derived from neofunctionalism, although within its explanation of the visibility of topics can considered borrowing the intergovernmentalists concepts of high and low politics. Before diving into postfunctionalism a few things should be known about its predecessors.

Firstly, both neofunctionalism and intergovernmentalism assume that the agents of

integration are political elites. For neofunctionalism, these are not the nation-states, but could be policy-makers, leaders of political parties but also lobbyists from companies or other interest groups (Haas, 1958). In line with the theory of pluralism, De Haas argued that political decision making is the result of those political actors with diverging interests “taking a utilitarian approach” to fulfilling these (Rosamond, 2005, p. 242) and subsequently shift their loyalties to a supranational entity. Contrary, stemming from classic international relations’ theories, intergovernmentalism considers the nation state the agent of (non)

integration. For intergovernmentalists, the state is the primary decision maker in the European Union and as such determines the level of European integration.

Secondly, both theories can be considered to be explaining the particular time in which they were formulated. 2 Neo functionalism was introduced by the Haas in his book The uniting of

Europe (1958) in a time where the community seemed to be integrating in a linear fashion

without constraint (Rosamond, 2000). As such de Haas introduced the concept of spillover that assumed that as that as integration is initiated in a limited area –mostly economic- further integration is stimulated (spillover), either by the initiators to reap further benefits, by

transnational interest groups that mobilized as a result of the first integration steps or by the established supranational institutions (Lelieveldt & Princen, 2015). This is in line with the developments of the EU at the time that neofunctionalism was grounded. However, in the 1960s, as a result of the empty chair crisis, the Luxembourg3 compromise was established, and European integration at least on an institutional level seemed to come to a halt

(Rosamond, 2000). To explain this halt of integration, Hoffman proposed that the nation state is a factor of “non-integration (Hoffmann, 1966, p. 862),” that is, the state is in general

2

although see Rosamond, 2005 for an explanation of the relevance of neofunctionalism today

3_{The Luxembourg compromise entailed that if a decision (made by qualified majority voting) would go against}

the “vital interests” of an EU member state, that state could “veto” the decision. In general, while making decisions, the council tries to incorporate all the concerns of the member states and tries to reach unanimous decisions (Craig & De Búrca, 2015).

(15)

unwilling to transfer power to a supranational institution. A leader of a nation-state will only pursue integration on issues of low politics: that are not vital to the identity of the nation-state and that are governed by a logic of integration. Contrary, high politics refers to issues that are important to a nation’s identity and that are governed by a logic of diversity(Rosamond, 2000, 2005). Because states have so many diverging interests on those issues, they will never be able to fully integrate. Like neofunctionalism in the early years of the community,

intergovernmentalism explained the European integration process, based on what was known about European integration at its time of writing.

However, as Börzel & Risse (2018) point out, intergovernmentalism fails to explain other examples of (non) integration, such as the inaction or stalemate of the European Union during the migration crisis in 2015. After all, the nation states had real common interest to find a solution to the issue, to ensure political and economic stability and definitely to keep the Schengen zone open. Both neofunctionalism and liberal intergovernmentalism cannot explain this. Rather, a theory is needed that encompasses the influence of non-elites on European Integration (i.e. the public). Postfunctionalism is such a theory.

2.1.3. Post-functionalism

Post-functionalism is similar to neo-functionalism in the sense that it also attributes a certain agency to political elites regarding the furthering of European integration, but contrary to neo-functionalism, post-functionalism is a multi-level governance theory (Börzel & Risse, 2018). This means that post-functionalism looks at the different levels of governance and influence, stating that different actors at different levels of governance all influence European decision-making and implementation. Agency to limit or perpetuate European integration is thus not only limited to political elites, but is also possessed by e.g. public administrators and the general European public (Hooghe & Marks, 2008).

Postfunctionalists argue that the initial years of the European Community could be

categorised as years of permissive consensus.(Börzel & Risse, 2018; Hooghe & Marks, 2005; Steenbergen et al., 2007) During this time, political elites furthered integration, but the general public was not very engaged with the European Community and remained dorment. From 1991 onwards the political parties mobilized public identity and a period of

constraining dissensus emerged (Hooghe & Marks, 2008). During this period the public was

more engaged with the European Union and political elites had to be more careful while making a decision on the EU as they could result in a backlash by the national public. In a

(16)

sense, this postfunctionalism thus draws on the intergovernmentalists’ concepts of high and

low politics. As European integration becomes more visible, touches upon topics considering

vital to the nation state (such as national borders, see chapter 1) and therefore becomes more salient, the actions of political elites to further integration become more scrutinised by their electoral base. This does not necessarily lead to a halt of European integration, but it can limit the growth in particular policy areas as well as slow it down.

Postfunctionalism is more adapt in explaining recent developments with regards to

integration, than its predecessors as it includes the influence of public opinion. It is therefore important to understand how those attitudes are formed and how they can be influenced. There are currently three main theories that deal with this question: utilitarianism, identity and political communication theory. These will be described below.

Utilitarianism

The oldest approach to explaining pro- and anti-integration attitudes (or Euroscepticism) is utilitarianism. This theory is derived from the rational school of thought and assumes that rational actors (citizens) are in favour or against regional integration depending on whether they are likely to benefit or be disadvantaged by such integration (Schmidt, 2010).4

According to Gabel and Palmer (1995, as cited in Gabel, 1998), citizens with a higher human capital5 are more likely to benefit from the integration or liberalization of the labour market as they are more capable of adapting to the European occupational competition resulting from this. As the attitudes towards European Integration depend, according to this theory, on the rational cost-benefit analysis of integration by individuals, highly educated people will be more in favour of European Integration than people with a lower level of education

(Hakhverdian, Van Elsas, Van der Brug, & Kuhn, 2013; see also the first part of Vasilopoulou & Talving, 2018). Furthermore, Gabel suggests that the liberalization of capital markets is more likely to benefit EU citizens with a high income or that are wealthy, as they “can exploit the greater investment opportunities provided by more open financial markets (Gabel, 1998).”

4_{As I also refer to in my method section, economic considerations are also theorized to determine the likelihood}

that someone is in favour or against a liberal immigration policy.

5_{Someone with a high human capital is a highly skilled worker and therefore often has a good economic}

position. Often human capital is certified by a degree. To be able to remain a high human capital in other countries as well, these degrees must be recognized across border. If this is the case, human capital transferability is high (Castles et al., 2014b).

(17)

Thus, wealthy citizens are also more likely to support European Integration than people with a lower income (note that highly educated people are also more likely to have a higher income). This utilitarian principle as described by Gabel seems to conflict with the notion that on a macro-level high capital countries are more likely to oppose the free movement of persons than low capital countries (as described in chapter 1, following Vasilopoulou & Talving, 2018). However, this can be explained by the different levels of analysis. On a national level, if a country is rich, the free movement of persons is likely to result in migration flows

stemming from poorer countries within the EU and as such increased competition. On a whole, high capital countries may therefore be less favourable towards such measures (Vasilopoulou & Talving, 2018). However, within the countries, European integration may have a different effect on individuals. Those with high human capital are less likely to be threatened by increased competition, will be more likely to benefit from the opportunities the freedom brings and will thus be more likely to be in favour of European integration (Gabel, 1998). As such the economic approach may seem contradictory when comparing national and individual approaches to integration, but when considering the different level of analysis, the same logic applies.

It should be noted that the utilitarian approach to explaining attitudes to integration has been criticised. Firstly, it is questioned whether citizens possess information to make a cost-benefit analysis. Inglehart’s theory of cognitive mobilization suggests that well-developed skills are needed to understand and appreciate European integration, because the EU is complex and abstract (as cited in De Vreese & Boomgaarden, 2005; Hobolt & Brouard, 2011; Lubbers, 2008). It may therefore not be possible for all individuals to obtain enough information to weigh the cost and benefits of integration, neither may they be willing to. Instead, it is argued that economic considerations are mediated by the effect of political cueing (Hobolt &

Brouard, 2011, see also “political communication” ) Furthermore, whereas Gabel found his theory among four others (values, partisanship, trust in government and cognitive

mobilization) to be most relevant, more recent studies that include the political identity approach suggest that the latter is a better predictor of support for European integration (Hobolt & Brouard, 2011; Hooghe & Marks, 2005, 2008; Lubbers, 2008). These theories do not completely discount the utilitarian approach, but stress that these multiple concepts interact, with identity being the most influential factor.

(18)

Political Identity approach

Hooghe and Marks (2008) explain the different speed of integration the EU has witnessed over the years, by focusing on an inherent contradiction in the idea of supranational

governance. On the on hand, they argue that only by coordinating human activity on multiple levels of governance, governance can be efficient. On the other hand, they mention that: “governance is also an expression of community” or common identity, and that “communities demand self-rule (2008, p. 2).” To explain European integration, it is therefore important to comprehend when, and in which manner identity is mobilized in the integration debate and how this mobilization can result in pressures to halt integration.

According to Hooghe and Marks (2008), a constraining dissensus emerged because political party leaders saw that they could profit from mobilizing people with exclusive identities against the EU. People with exclusive identities, identify themselves exclusively with one (geographical) entity, usually the nation state (but sometimes this can also be a region or province). People with inclusive identities, on the other hand, consider themselves to have multiple identities, e.g. being both a Dutch and an EU citizen. As such they do not consider regional (European) integration as a threat to their national identity and are more likely to support it, than those who see European identity and national identity as competing. In line with postfunctional theory, Hooghe and Marks argue that as European integration expands, the integration will become more visible, salient and, therefore, threatening. As such, the permissive consensus is replaced with a constraining dissensus and anti-european attitudes become more pronounced. Hooghe and Marks (2005) also link this identity approach to political cueing (see next section).

Political communication

Political communication theory posits that it does not only matter what socio-economic position one has or whether one has an inclusive or exclusive identity, it is also important to consider how people or elites communicate around issues surrounding European integration. Besides identity and economic position, discourse or communication matters as well (Gabel, 1998; Hooghe & Marks, 2008; Schmidt, 2010; Schmidt & Radaelli, 2004; Steenbergen et al., 2007).

Schmidt & Radaelli (2004), for instance, focus on the link between discourses and policy and institutional change within the European Union (Schmidt & Radaelli, 2004). In her research, Schmidt shows that the discourses within the political community influence the ability of

(19)

institutional change. However, whereas her approach is multi-level considering next to European decision makers also the influence of national politicians, public administrators and (trans) national interest groups, she does not focus on the influence of the public on EU decision making and implementation. Rather she focuses on how political elites (such as national politicians) communicate to the general (or informed) public in order to legitimate and put into perspective EU policies. How the public in turn influences discourse around European Policy and the subsequent decisions, is not considered in her work.

Steenbergen et al.(2007), who also focus on political communication, do a better job here. In his paper Who’s cueing whom? Steenbergen investigates the very question of the influence the public has on political elites determining policy change and vice versa (Steenbergen et al., 2007). He argues that mass-elite linkages can be either top-down or bottom-up. Bottom-up means that political elites (often leaders of national parties) can take over the position of their public. Through an electoral connection, the mass can exert influence over the European Integration process. Moreover, the mass public can take over the opinion of the political elites. This is referred to as cueing. Steenbergen (2007) shows that both processes occur and interact with each other. Thus, whereas individuals do receive cues from e.g. political parties and can subsequently change their opinion

2.2. Linking the explanations and concluding remarks

However, it is nowhere true that these three explanations are mutually exclusive. Steenbergen does work together with Hooghe and Marks - who focus more on the identity approach- (Marks, Hooghe, Steenbergen, & Bakker, 2006) and mentions the relevance of utilitarianism (Steenbergen et al., 2007). Similarly, Gabel(1998) shows that net to utilitarian arguments, cues by political elites have an important influence on public opinion. Furthermore, Schmidt explicitly mentions that political communication is one of the factors influencing European integration but by no means the only or more influential one. Rather she advocates more research on when and to what extent political communication matters. Hooghe and Marks (2005) link this concept of cueing with their identity approach towards explaining attitudes to integration and suggest that political cues “mediate the effect of economic calculation and community”. They argue that those who have an exclusive identity are more easily cued to believe that furthering European Integration is incompatible with remaining a strong national identity and are therefore more easily influenced to oppose European Integration. People with inclusive identities are theorised to be more favourable towards the EU and less likely to change their opinion on the EU based on cues. An individual with an exclusive national

(20)

identity, on the other hand, will be more sceptic towards the EU “if they are cued to believe

that love of their country and its institutions is incompatible with European integration

(Hooghe & Marks, 2008, p. 13).

As such the above theories do suggest that the different determinants of attitudes matter to a different extent, but how they exactly interact is not abundantly clear.I argue that this is due to the fact that they focus too much on the causes of attitudes towards integration, without understanding the exact content of the attitudes, I expect this thesis to help shed some light on the proposed interaction. The following section looks at the migration history and culture in the Netherlands, as well as Dutch approaches towards European Integration. I then continue to formulate expectations of discourses (narratives of attitudes) I expect to find in my research, based on this and the following chapter. These discourses will, according to the literature, focus on economic and cultural threat consideration.

(21)

Chapter 3: European Integration in the Netherlands

In this chapter, I will explain more on the background of EU integration and migration attitudes in the Netherlands. I will explain why people can be in favour or against EU

integration and what, considering the Dutch cultural and historical background we can expect to come forward in the study. I will link back to this in the analysis part of my thesis.

3.1. Migration and the Netherlands

Migration patterns in the Netherlands (1945 – now)

During the first period after the second world war (1947-1973), migration to Western

European countries drastically increased, due to the stark economic growth. Castles, De Haas & Miller (2014a) note that there are two main streams of migrants in that period:

- Labour migrants from the European periphery that entered the western countries to work either spontaneously or through “guest workers” schemes. In the Netherlands, most guest workers were recruited from Turkey and Morocco.

- Migrants from former colonies to the country of their former occupiers. For the Netherlands, this were people from Indonesia as well as the Antilles.

The first group was expected to migrate to western Europe only temporarily (until economic downturn for instance) and the second group, for the most part, intended their stay to be permanent (Castles, De Haas, & Miller, 2014b). Another surge in migration within Europe occurred as a result of the 1968 free movement of workers policy within the European Community.

In 1973, with the global oil crisis, an economic downturn occured. However, rather than returning en masse to the source countries, many migrants stayed and brought over their families to western European host countries. This also occurred in the Netherlands and was strengthened by the recruitment freeze and immigration restrictions that were implemented by western European countries in reaction to the 1973 oil crisis. Many temporary workers were afraid of not being able to enter their host country again if they left it and subsequently decided to stay there permanently (Castles et al., 2014b). As can be seen from Figure 1, since the 1970s the Netherlands has been a receiving country, meaning immigration has been consistently more than emigration (CBS, 2017).

(22)

Figure 1: Immigration and emigration in the Netherlands (1945-2017)

(CBS, 2017) After the 1973 oil crisis, there was a restructuring of the world economy towards a more neo-liberal functioning and globalisation became more prominent, resulting in more migration. Further European integration, such as the signing of the Schengen agreement in 1985 and the expansion of the EU also led to an increase in migration (Castles et al., 2014b). These trends are visible in Figure 2 where immigration to the Netherlands by country of origin is shown. As can be seen, immigration from the EU has become more prominent over time, whereas the “guest workers” from Turkey and Morocco immigrate less to the Netherlands (CBS, 2017). This does not mean that they become less visible though. Contrary, due to the second and third generation of the migrants and the clustering of those groups in particular areas (such as neighbourhoods in major cities like Amsterdam, Rotterdam and the Hague) the migrant population has become more visible, resulting in a more negative attitude towards them (Castles et al., 2014a).

(23)

(CBS, 2017)

Integration policies in the Netherlands (1945-now)

What is most important to understand in relation to the research question, is not necessarily why migration to the Netherlands occurred, but how the Dutch dealt with those migrants and what effect this has had on the population. According to Koopmans, Statham, Giugni and Passy (2005), there are four main approaches countries can take towards migrants: assimilationism, segregationalism, universalism and multi-culturalism. These approaches depend on whether citizenship is granted on descendance (jus sanguinis acquisition) or birth on territory (jus soli acquisition), or whether or not migrants are allowed to retain their own culture, language and traditions (cultural pluralism vs. cultural monism). Figure 3 shows the matrix of these conceptions on citizenship.

Notably, in the past, the Netherlands was known for its multiculturalist’s policy (Koopmans et al., 2005). This means that people born in the Netherlands are considered Dutch citizens and that minority (immigrant) groups are allowed to retain their own culture with equal rights to execute that culture as the majority group. For instance, in the Netherlands publicly-run

(24)

schools as well as Christian, but also Islamic schools receive government funding (Rijksoverheid, 2018). However, during the mid-1990s, the Dutch (then left-liberal) government started to retreat from the multi-culturists approach due to the difficulties in maintaining the policy for a more diverse migrant population and the fact that it leads to the segregation and socio-economic deprivation of the migrant population (Joppke, 2004). Subsequently, Joppke (2004) argues, the Netherlands has taken a civic integration approach, which is similar to Koopmans’s concept of universalism. This approach focuses less on group rights to exhibit cultural difference and focuses more on the obligation of (potential) migrants to integrate into Dutch society, to conform to the Dutch culture and values, and to learn to speak Dutch. As such, migrants are required to pass an integration test (Joppke, 2004;

Koopmans et al., 2005). More recently the burden on migrants to integrate has been increased, by making it a requirement for a permanent residence permit and making the migrants bearer of the costs for the integration course and test (Castles et al., 2014a).

However, not all academics solely explain the shift towards universalism by the problems generated by the Dutch multicultural approach. Freeman argues (1995), similar to the post-functionalist explanation of European Integration (Hooghe & Marks, 2008),that political elites with economic interests dominated the debate on migration, until after the oil crisis of 1973 it became apparent that temporary migration (e.g. the guest worker system) would inevitably

(25)

lead to permanent settlement and more competition on the labour market, even in times of crisis. The idea was formed that the government had no control over border flows and settlement and as such the conversation on migration became more salient to the public and anti-migration groups became more organised and joined the debate.

Figure 4: Development in the national issue concerns, the six largest categories from the second quarter of 2018 (in percentages) (Sociaal Cultureel planbureau, 2018, p. 21).

The Dutch Sociaal Cultureel en Plan Bureau continuously monitors Dutch citizens’ perspectives on the most important issues in the society. As can be seen from their second report of 2018 (when I did my interviews) immigration and integration are considered one of the greatest issues by the Dutch population (see the arrow, Figure 4). Since its peak in 2016, the salience of this topic has decreased, but it still is –and consistently has been- cause for concern by the Dutch population. This is thus indeed in line with the postfunctionalist

explanation that the visibility of an issue, especially one that is considered to threaten national sovereignty or identity can lead to a constraining disconsensus, and as such, results in less openness to that threatening process.

(26)

3.2. The Netherlands and the European Union

The approach of the Netherlands towards European Integration is to a certain extent parallel to the approach towards migration, wanting more in the early years of the EU and more recently trying to slow it down. This is no surprise as immigration attitudes are a predictor (along with trust in national government) to European support or attitudes towards EU integration (Boomgaarden, Schuck, Elenbaas, & de Vreese, 2011).

The Dutch citizens’ approach to EU integration

Interestingly, a majority of the Dutch seem to have either an inclusive or a European Identity. According to the Eurobarometer (European Commission, 2018a, p. 37), 63 percent of the Dutch respondents identify as Dutch first and European as well, a further 7 percent as firstly European and then Dutch, and 1 percent identifies as European only. Thus, 71% of the respondents identified as at least partly European, which compared to the average of 63 percent of the entire EU is rather high. Furthermore, a large percentage of the population seems to have an inclusive identity. According to identity theory, this means that a large part of the Dutch population is likely to have a pro-European attitude. However, it should be noted that a still rather large portion (28%) identifies as only Dutch. This group, that seems to have an exclusive identity, is likely to oppose European integration.

It seems that the economic reasons are mostly called forward when explaining positive sides to EU citizenship, whereas the main reasons against EU integration are formulated around identity threat. This can be seen both from the data from the EU barometer and the Sociaal

Cultural Planbureau. When asked what citizens thought was the most important benefit of the

EU, 63% of the Dutch participants mentioned the free movement of people, goods and services within the EU as their first or second choice (see figure 5 European Commission, 2018a). The main assets of the EU were considered the EU’s regard for democracy, human

rights, and rule of law (43%) as well as the economic, industrial and trading power of the EU

(27)

When asking about the most important issues the EU was facing, however, migration was mentioned most regularly by the Dutch (45%) (European Commission, 2018b). The Sociaal Cultureel Planbureau (Ridder, Boonstoppel, & Dekker, 2018) shows similar results. When asked about globalisation, both the majority of the lower educated and the higher educated believe this brings opportunities for economic growth. However, when asked about identity, most highly educated Dutch citizens believe globalisation will not lead to a loss of national identity, whereas less educated citizens overwhelmingly do fear this. This is portrayed in Figure 6 that portrays the amount percentage of people rejecting the statement that

globalisation threatens the identity of the Netherlands, minus the percentage of people who agreed with this. As can be seen, people with a lower education tend to agree with the statement, whereas highly educated respondents reject it. In this case, the identity approach towards explaining attitudes towards immigration seems better applicable than the economic interests.

(28)

Figure 6: Netto rejection “globalisation threatens the identity of the NL”

Furthermore, research into Dutch Euroscepticism that mostly focuses on explaining the Dutch “no” to the European constitution in 2007, suggests that in the Netherlands, discourses around cultural threat were decisive in explaining the votes against (Hobolt & Brouard, 2011;

Lubbers, 2008). This seems to be culturally dependent, as in France, socio-economic concerns were far more prominent (although the far-left in the Netherlands also had socio-economic concerns). According to Hobolt & Brouard (2011), anti-establishment, anti-migration parties were able to frame the debate on the constitution in terms of cultural threat and thereby were able to cue a large group of people to vote against the constitution. The visibility of the

migrants due to the Dutch previous multiculturalist integration policy (Koopmans et al., 2005) may have enhanced the opportunity for this framing. This is in line with the political

communication theory described in chapter 2.

Concluding remarks and expectations

All the theories using individual characteristics to explain attitudes to integration seem to occur when looking at the large-scale opinion data of the Netherlands. What is left out is how these arguments are weighed by people or in other words how they interact.

The available data thus suggests that there are two groups of people. The ones that have an inclusive identity, mostly see economic benefits of European Integration, are highly educated and in favour of furthering European Integration. Another group that identifies exclusively or

(29)

mostly Dutch, has likely a lower socio-economic position and sees European integration as a threat to the national identity. I do expect to find these two discourses (pro-integration and anti-integration) in my thesis. However, as this chapter explains, there seems to be a

difference between the way economic opportunity and identity threat is approached. Whereas a majority of people, from all levels of education, considers the free movement of persons to offer economic opportunities, the level of educations greatly matters for whether or not Dutch citizens believe the globalisation to threaten national identity. This suggests there is a group that sees the economic advantages of the free movement of persons, but that also opposes it based on its threat towards national identity. If such a discourse exists, it is logical to assume an opposite discourse exists as well. I therefore expect another group of people to feel economically threatened by an increase of European integration, but –having an inclusive identity- feel migrants do not threaten the national identity of the nation state. In short, I expect the following four discourses:

Discourse 1: pro- integration

This discourse strongly favours European integration and the free movement of persons. It is likely that people with high human capital (i.e. education) as well as people with an inclusive identity favour this discourse, as they benefit economically from European integration, and are less likely to feel like their own identity is threatened by the immigration of people with different cultures.

Discourse 2: economic benefits, but threatened identity.

I expect to find this discourse among those with an exclusive identity, but who do not encounter more direct competition by international workers. As they do not suffer direct disadvantages from this freedom of movement from an economic perspective, they may look favourably towards it, as from a macro-economic perspective this may benefit the country. However, the visibility of the different cultural backgrounds of the migrants in combination with political cueing (such as by parties as the PVV), leads those people to believe that European integration and the freedom of movement, threatens the Dutch national identity. As a result they will –overall- oppose further European integration.

Discourse 3: no cultural threat, but economically disadvantaged.

This discourse reflects the idea that people with an inclusive identity, do not consider immigrants as a threat to their identity or to Dutch cultural values, norms and traditions, but

(30)

do experience increased competition due to the free movement of persons policy. Whether they oppose or advocate further European integration therefore depends on how they weigh this contradiction: whether they are led more by values or interests. Recent research that uses postfunctionalism as their framework, does suggest that identity is more important in

determining whether interests or values are more important in deciding people attitudes towards integration. (Hobolt & Brouard, 2011; Hooghe & Marks, 2005, 2008; Lubbers, 2008). I therefore expect this discourse to slightly favour European integration, but not to the same extend and speed as discourse 1.

Discourse 4: Euroscepticism

This last discourse, I expect juxtapose discourse 1. People favouring this narrative have an exclusive identity and are either competing or perceive themselves to be competing with labour migrants. As such, they feel the free movement of persons both results in a threat to the national identity and their personal job security. Furthermore, I expect those people to view Brussels or the EU as an abstract entity imposing its views. This is in line with political communication theory, suggesting that people who know little about the European Union can easily be cued to find it infringing on national sovereignty. Thus, people favouring this last discourse, will not only oppose further European integration, they will also believe it would be best to turn it around. They will stress the importance of retaining national borders and sovereign power to make decisions on a national level , thereby decreasing the current level of European integration. This is a clear Eurosceptic discourse.

To test whether these hypothesized discourse are present within the Dutch society, I used Dryzek’s framework of discourse analysis. This is described in the following chapter.

(31)

Chapter 4: Methodology

4.1. Introduction

In 1935 professor of physics and psychology William Stephenson wrote a letter to nature, detailing a new approach towards factor analysis. In this letter, Stephenson proposed to invert the analysis. Rather than measuring the differences between variables, tests could be

measured by focusing on the differences between individuals (Stephenson, 1935). Easier said, “Q correlates persons instead of tests (Exel & Graaf, 2005)”. Although not the only

contemporary playing with this idea –Thompson proposed a similar method in the same year- Stephenson was the first who saw potential in it and as such the letter is considered to be the start of the long list of literature detailing Q methodology (Steven R Brown, 1980).

Q methodology is most commonly used to measure subjectivity. According to the proponents of Q methodology (e.g. Brown, 1993; Brown, 1980; Stephenson, 1994), views, opinions or behaviours expressing such, cannot be considered “right” or “wrong”. Rather, they are self-referent, that is –opinions are always true to the person formulating them and their relative importance depends on the importance individuals –not the researcher- attributes to them. This makes it difficult to measure and interpret views, without imposing meaning on them. In his famous book political subjectivity, Brown explains why:

[O]perationally defined measures of liberalism and conservatism, to take a common example, usually postulate a continuum with left at one end and right at the other. A test is constructed containing statements of both kinds and each respondent receives a mean score, thereby creating the spurious impression that a left-right continuum exists in nature when, in actuality, the results are merely artifacts reflecting the way in which the data were collected. There is no doubt that subjects' responses to scales are

operant6 in some sense, but the concreteness of the response is bypassed in favor of the "objective" meaning of the scale. The individual's independent point of view, in effect, is considered to be dependent on the prior meaning of the scale (Brown, 1980).

Thus, because traditional social science research tries to find correlations between variables, such variables are measured (operationalised) using the researcher's idea of what answers indicate high scores on a particular variable. While this works when investigating facts, it

(32)

becomes more difficult to operationalise subjective ideas, because the respondent may not agree with the researcher’s categorization. To truly investigate what people believe, a different method is required. Qmethodology is such a method. The basic idea of

Qmethodology is that by allowing participants in a study to study a large amount of statement, they are given room to give meaning to the statements by putting them in relation to each other. As such, Qmethodologists hope to find “how its subjects actually do apprehend the world, not in how the researcher expects them to do so (Dryzek & Berejikian, 1993).”

Throughout this chapter, Qmethodology is further described along with the choices I made to make this research a robust as possible for the scope of this thesis.

4.2. Sampling method

I. Concourse

Q methodology is used to measure subjectivity. As such the phenomena it tries to understand is ordinary conversation, discourses, commentary, art, poetry etc. This “flow of

communicability surrounding any topic is referred to as a concourse (Brown, 1993)”. In this paper, the concourse refers to all expressions of opinion in the Netherlands surrounding the topic of the freedom of movement of persons.

It is impossible to gather all these expressions, not only because there would be far too many of them, but because these expressions are made continuously in everyday conversations and it is not possible to bear witness to them all the time in every place in the Netherlands. However, one of the main assumptions of Q-methodology being that the number of opinions present within a society is limited and can be found while using a small but varied population, it is not necessary to use the entire concourse, as long as the used concourse is representative (Dryzek & Niemeyer, 2008).

Ideally, to find a concourse (expressed in language) focus groups are used (Brown, 1993; Dryzek & Holmes, 2002). For the purpose of this study, however, this is non-feasible as it would require multiple focus groups in different locations throughout the countries composed of a variety of participants. Therefore, I looked at (opinion)newspaper articles, the comment section of newspaper articles and discussion fora. I exhausted as many sources I could think of. This means I used the lexisnexus database, sites of major Dutch newspapers (including

(33)

Topics7) and googled for discussion fora. I used words like free movement of persons,

European borders and Schengen to find statements. The key words were established by the

entire capstone group and were found by brainstorming initially about the freedom of

movement and then adding words to the list as the list of found sources progressed. The list I used can be found in appendix I.

II. Qset

The initial concourse I established contained a list of 158 statements. Asking respondents to sort this many statements would be too much of a burden on them. Therefore, I reduced the concourse to a Qset, the set of statements that respondents actually sort during the interviews. To make sure the Qsort was as varied as possible and to limit the influence of my own

judgement on the Qset, I used Dryzek’s method for sampling a concourse (Dryzek, & Holmes, 2002; Dryzek & Berejikian, 1993).

Dryzek proposes that statements can be arranged into a matrix as shown in Table 1, according to the type of claim and the discourse element (derived from discourse theory). The type of claim can be:

1. Definitive: concerned with the meaning of terms 2. Designative: concerned with matters of fact

3. Evaluative: concerning the worth of something that exists, or might exists 4. Advocative: concerning something that should or should not exist.

(p. 26, Dryzek, & Holmes, 2002)

The discourse element describes what “stuff” the statement is making claims about. Claims can be made about the ontology of an entity (or set of entities), the agency ascribed to these entities, the motives agents can have to act, and the (un)natural relationships between political entities.

A. An ontological claim refers to the (recognised) existence of entities (e.g. elites, nations, democracies, politicians, classes etc.)

B. Agency describes how much a statement claims entities have the ability to act. Statements can ascribe or deny the agency of entities.

7

I have a prescription on the paper The Volkskrant and as such have access to topics, a site that collects all articles from 14 major newspapers in the Netherlands (with different ideological backgrounds and audience)

(34)

C. Statements can make claims about motives of acting entities. These motives can be emphasized, but also denied.

D. The last discourse element refers to (un)natural political relationships. Statements in this category make claims about the relationship between entities (such as there is conflict, harmony between classes or there should/should not be competition between nation states).

Table 1: Matrix for Sampling a Concourse. Discourse Element

Type of Claim Ontology Agency Motivations Natural

Definitive 1 2 3 4

Designative 5 6 7 8

Evaluative 9 10 11 12

Advocative 13 14 15 16

I categorised statements according to the matrix resulting from combing the discourse element with the type of claim. Consider for example the following statement: All EU member states

should be obliged to take in an – yet to be determined- amount of asylum seekers [trans.]

Because it makes a claim about what should be done, I categorised this statement as Advocative Agency (box 14). After sorting all 158 statements, I selected 4 statement from each box, while maximizing the variety of statements, resulting in a Qset of 64 statements (see appendix IV).

The procedure as described above is still an interpretative exercise as the researcher decides how to categorise ambiguous statements and ultimately selects the statements for the Qsort. However, if the statements are categorised consistently, the method does lead to a varied set of statements.

III. P-set

Due to the intensive nature of the sorting in qmethodology, the number of people that can be interviewed is limited. This does not mean that the research method is not robust or accurate. As mentioned, Q methodology does not aim to be representative of the population. Rather, it is used to distil the discourses present in society (Dryzek, & Holmes, 2002). One of Q methodology’s prime assumption is, therefore, that there are a limited amount of discourses

(35)

present in society and that research can be considered successful when those discourses have been uncovered. Thus, rather than representativeness, the p-set should lead to data saturation; that is, if more people than included in the research are interviewed, no new discourses will be discovered (Dryzek & Holmes, 2002). A p-set should, therefore, include as much diversity as possible on variables that are likely to lead to different perceptions on the EU policy.

To determine my p-set, I looked at quantitative studies relating individual characteristics to attitudes towards immigration and European integration to determine which variables matter. Due to the limited scope of this study, I only used the most recurrent and influential variables I found in the literature to determine the p-set: age, gender and level of education.

Level of education

According to Wilkes, Guppy & Farris (2008), the literature on the impact of individual

characteristics on migration attitudes, can be broken down into two main narratives: economic and ideological explanations. Explanations focused on ideology concentrate on beliefs. The idea here is that “anti-immigration views reflect a larger set of beliefs about the native-born being superior to people who are culturally and/or racially different” (Wilkes et al., 2008, p.304). According to this logic, when migrants are culturally more similar to nationals, the host country will perceive them more positively. For determining the p-set, ideological explanations are not very useful as

Economic explanations focus on the idea that when migrants are competing with a person’s job position, thus if they are part of the same labour market, this person is more likely to evaluate migrants as threatening and as a result develops anti-migration attitudes (Card, Dustmann, & Preston, 2005). An important factor here is not the actual position one has in the job market, but a person’s perceived place in the job market (Tuci, 2005). According to the described logic, people with higher education, or socio-economic status, will be more positive towards migrants if those migrants are from a lower socio-economic background, or are unable to transfer their human capital to the host country. According to the same logic, in such a situation, people with a lower socio-economic background will perceive migrants more negatively(Wilkes et al., 2008).

The interests vs. ideology based explanation also comes back in the literature on

Euroscepticism. For instance, Hakhverdian, Van Elsas, Van der Brug, & Kuhn (2013) refer to the idea that there are “winners” and “losers” of globalisation, the former often having a higher level of education and the latter a lower level. As a result, lowly educated people tend

Understanding “what” people think about European Integration: A Qmethodological Study of Dutch’ Discourses on the Freedom of Movement of Persons