Monitoring the impact of doing nothing: New trends in immigrant integration policy

Abstract

‘Mainstreaming’ has recently been considered as a possible new strategy for advancing immigrant integration in Europe. However, policy documents and current academic literature have hardly conceptualized what we label as ‘ethnic equality mainstreaming’. In this article, we lean on the widely available research on gender mainstreaming, to provide such a conceptualization of ethnic equality mainstreaming. Once conceptualized, we verify whether there is indeed a trend towards mainstreaming in Western Europe's old immigration countries. Our results show that there is no straightforward trend towards ethnic equality mainstreaming in these countries. However, the indicators that served to detect the existence of ethnic equality mainstreaming allowed us to uncover a new double and paradoxical trend in immigrant integration policies. This ‘new style’ immigrant integration policy can be depicted as follows: increasing ‘colourblindization’, in combination with ‘ethnic monitoring’. In other words, states increasingly monitor the impact of ‘doing nothing’.

Keywords

Immigrant integration policies ethnic equality policies gender equality policies mainstreaming public policy policy change

Introduction

Immigrant integration policy development has not stood still since the start of the new millennium. Research has shown that important policy changes have occurred in Western Europe. While in the 1990s, policymakers were becoming ‘all multiculturalists’ (Glazer, 1998), in the 2000s, immigrant integration policies were said to move away from multiculturalist policies (Vertovec and Wessendorf, 2010; cf. Banting and Kymlicka, 2012) towards civic assimilationism (Joppke, 2007). Most recently, academics have referred to ‘mainstreaming’ as a possible new strategy for advancing immigrant integration in Europe (Collett and Petrovic, 2014; Scholten and Van Breugel, 2017). Mainstreaming first appeared in European policy documents in 2004, when it was defined as one of the 11 ‘Common Basic Principles on Integration Policy in the EU’. Since then, mainstreaming has been highlighted in the ‘Common Agenda on Integration’ and was further defined in the second version of the Handbook for Integration for Policy-Makers and Practitioners. In addition, recent empirical research has documented the emergence of mainstreaming strategies in the national and local governance of diversity in several European countries (Adam and Van Dijk, 2015; Scholten et al., 2016; Van Breugel and Scholten, 2017).

Despite the recent references to mainstreaming in the context of immigrant integration, policy documents and current academic literature have hardly conceptualized (what we label as) ‘ethnic equality mainstreaming’ (EEM) (Bell, 2008).¹ This is unfortunate because, without having identified the fundamental components of this policy strategy, there is a risk of EEM becoming a hollow label that is applicable to widely varying approaches to immigrant integration. The absence of a clear definition thereby risks hampering theory development, since one might be explaining very different policy trends when referring to mainstreaming. The immigrant integration literature also does not consider the rich academic reflections on gender mainstreaming. Yet, the mainstreaming strategy was first developed in gender theory and there are many similarities between gender and ethnic equality strategies.

The aim of this article is twofold. First, by relying on the widely available theory on gender mainstreaming, it will conceptualize mainstreaming as a strategy for immigrant integration and ethnic equality and build a set of indicators to empirically verify its occurrence. Second, using these indicators (which we call the ‘EEM Indicators’), it will systematically verify the emergence of this policy strategy across different countries. While it is not the core objective of this article to provide a systematic explanation for the observed policy trends, in the conclusion the paper will offer potential hypotheses to understanding the detected trends as an avenue for further research.

The article is structured in three sections. By making a parallel with the development of gender mainstreaming, the first section aims to identify the necessary conditions to be fulfilled for an immigrant integration policy strategy to qualify as ‘mainstreaming'. The second section will conduct an empirical test on the fulfilment of the established conditions for EEM in the Netherlands, France, the United Kingdom and Germany. The third and concluding section summarizes our results and advances avenues for further research. Our results show that there is no straightforward trend towards EEM in Northwestern Europe. However, the EEM Indicators, which served to detect the existence of EEM, allowed us to uncover a new double, and seemingly paradoxical, trend in immigrant integration policies. This ‘new style’ immigrant integration policy can be depicted as follows: increasing ‘colour-blindization’, in combination with ‘ethnic monitoring'. In other words, states increasingly monitor the impact of ‘doing nothing' (specific).

Before proceeding with the conceptualization of EEM, let us briefly discuss terminology. We use the term ‘ethnic equality mainstreaming’ (Bell, 2008) because we essentially regard mainstreaming as a policy strategy for promoting equality between, in this case, native populations and populations with a migrant background. Given the still-existing inequalities for second and later generations as compared to native populations, the strategy’s focus is on all people with a migrant background. In this sense, using the term ‘immigrant integration mainstreaming’ might lead to confusion since, strictly speaking, immigrant integration policy focuses solely on immigrants and not on their offspring, nor on immigrants who have become citizens. Moreover, the term ‘integration’ often implies cultural conformity (Li, 2003). In our definition of EEM, the term ‘ethnicity’ is broadly conceived and includes race as a subtype of ethnicity (see also: Wimmer, 2008: 973–974). It thus refers to a policy strategy promoting equality for all people with a migrant background who are ‘ethnicized’ because of their names, accents, skin colour, religion, cultural practices, etc. and who might therefore face specific barriers to equality.

Conceptualizing ethnic equality mainstreaming

In the social sciences, conceptualizations can be derived from earlier established theories in the academic literature, but can also be constructed inductively based on systematic empirical observations (Gerring, 1999). In addition, there is also a ‘middle way’, in which conceptualizations are built from theoretical ideas as well as definitions in the empirical reality. However, as Emile Durkheim (1982) pointed out, social science concepts do not always, or even generally, tally with those of the layman. Van Breugel et al. (2014) made a first endeavour to conceptualize mainstreaming as a governance strategy in immigrant integration. They rely on a mixture of theory and policymakers’ definitions to establish indicators of ‘immigrant integration mainstreaming’. Because of this inclusion of policymakers’ definitions in their conceptualization, their end result deviates from the mainstreaming concept in gender theory on one important point, namely the fact that generic (colour-blind) policies can be considered as an indicator of mainstreaming, while this would not be the case if the concept would be inspired by the literature on gender mainstreaming. In this article, we take a different approach to conceptualize mainstreaming in the context of immigrant integration. We develop an ideal-type (Weber, 1978) conceptualization of EEM from the literature on gender equality policies and do not take laymen’s or policymakers’ definitions into account.

Mainstreaming: The ‘third’ equality strategy

In the literature on gender equality policy, mainstreaming is considered as the ‘third’ equality strategy (e.g. Booth and Bennett, 2002; Rees, 2005). Gender equality strategies moved from an initial focus on equal rights, towards a focus on equal outcome through the use of specifically targeted gender equality policies and specialized ministries for gender equality. Gender mainstreaming is regarded as the third strategy in this sequence of policy strategies aimed at fostering gender equality. We consider the gender mainstreaming literature as a relevant heuristic tool to conceptualize EEM, because we observe a clear parallel between the two equality strategies. Both types of strategies seem to have gone through a similar evolution. Note that we do not intend to claim that adopting any of these strategies will automatically lead to more equality. The aim of this article is not to measure the effectiveness of different (gender and ethnic) equality strategies, but to develop a set of indicators on the basis of which an empirical trend towards the mainstreaming strategy can be systematically assessed. This systematic assessment can be a first step in the direction of a future measurement of the effectiveness of EEM.

From the late 1990s to the early 2000s, gender theorists have distinguished between three strategies to achieve gender equality, each of which promotes a different approach to policymaking (Squires, 2005; Verloo, 2001). The first strategy inspired the ‘first wave’ of gender equality policies and focuses on equal rights by ensuring equal treatment in legislation. The goal of this first strategy is to bring about equality in opportunity for all citizens (regardless of their gender), while leaving it up to individuals themselves to exercise these rights (Booth and Bennett, 2002; Verloo, 2001). This gender equality strategy corresponds to a similar equal rights strategy in immigrant integration. The post-war period was marked by an expansion of social, civil and political rights for immigrants. This evolution towards inclusiveness for immigrants has been labelled as a trend towards ‘postnational citizenship’ (Soysal, 1994). Besides this evolution towards ‘postnational citizenship’, the legislation concerning access to citizenship was also liberalized in the post-war era (Koopmans et al., 2012).

The second gender equality strategy moves beyond legal equality, by recognizing that not every individual can exercise his or her rights to the same degree as others. This strategy rather aims for equality in outcome through the use of specific or targeted policies, such as positive action and the accommodation of special needs. These kinds of policies are usually devised and implemented by state structures that are specifically assigned the competence for gender equality (Booth and Bennett, 2002; Verloo, 2001). From the 1990s onwards, a comparable approach to immigrant integration arose. Together with the realization that immigration is a permanent feature of European societies, immigrant integration became a stand-alone policy area. Much like the second gender equality strategy, this approach stresses the need for immigrant integration policies that recognize the specific situation of people with a migrant background: their struggles with language acquisition, their confrontation with discrimination, their difficulties in getting their diplomas recognized, etc. This evolution can also be connected to the establishment of specialized immigrant integration ministries and their respective ministers in Western Europe.

Gender mainstreaming is considered the third gender equality strategy and differs from its predecessors in that it presumes that gender inequality is reproduced throughout mainstream society by gendered norms and structures. Gender mainstreaming targets the more structural causes of gender inequality by removing these gender biases (Mazey, 2001; Squires, 2005). As opposed to making gender equality the sole responsibility of specialized state structures, the mainstreaming strategy requires that a gender equality perspective is taken up in all policy sectors and at all levels of governance. It entails the reorganization of policy processes in such a way that the needs of specific disadvantaged groups in society (such as women) are taken into account throughout the whole governing process (Verloo, 2001). However, this does not mean that gender mainstreaming necessarily entails a move away from targeted policies. In fact, gender scholars advocate using a ‘twin track strategy’ or a combination of generic and targeted equality policies which can complement each other to achieve gender equality (Rees, 2005: 559–560). Targeted policies remain especially important to address indirect discrimination² of certain groups in society (Squires, 2008: 59–60). Mainstreaming can then be considered a policy strategy that calls on actors involved in policymaking to reorganize policy processes. It demands from these actors that they integrate a gender equality perspective in all policy areas and at all stages of the policy process, and that they evaluate the outcomes that their policies produce for women (Rees, 2005; Verloo, 2001).

Whereas the debate on the remaining importance of national models in immigrant integration is still ongoing, recent empirical studies (Adam and Van Dijk, 2015; Scholten et al., 2016) hint at the start of a new phase in immigrant integration policy. They have demonstrated the emergence of deconcentrated and decentralized governance and a move away from specifically targeted policies and specialized ministries in the field of immigrant integration. Considering the similarities between, respectively, gender and ethnic equality strategies as discussed in this section, it becomes interesting to explore whether these recent developments signal the adoption of the mainstreaming strategy in immigrant integration.

Defining ethnic equality mainstreaming

To be able to identify policy strategies as (gender or ethnic equality) mainstreaming, we have to know what kind of transformations qualify as mainstreaming in the academic literature. In this section, we will try to establish the necessary conditions that a policy strategy needs to fulfil to qualify as gender mainstreaming and derive our ideal-type definition of EEM from there. Although there is considerable discussion about the essential components of the mainstreaming strategy and its subsequent effect on gender equality (see, for example, Meier and Celis, 2011), the academic literature addressing gender mainstreaming agrees on at least three essential components of the mainstreaming strategy.

First, mainstreaming is about integrating gender equality objectives in all policy areas. The strategy means to open up all policymaking (i.e. policies in the field of employment, education, housing, culture, welfare and all other relevant fields) to the specific needs of historically disadvantaged groups (Lombardo, 2005: 417; Walby, 2005: 323–324). This entails reflection on the effect of general policies on such disadvantaged populations, alongside the use of targeted policies, specifically directed at particularly disadvantaged groups, within general policy areas. Similarly defined, EEM would then comprise the integration of ethnic equality objectives in general policymaking by installing policies specifically targeted at populations with a migrant background in general policy areas and assessing the impact of general policies on people with a migrant background. An example of a targeted policy within a general policy area (education) is, for instance, the creation of a mentoring program for pupils with a migrant background. An example of an impact assessment of general policies on particular groups would be an evaluation of the increasing cost of childcare on the employment rate of migrant women. In this way, specific ethnic equality considerations become an integral part of general policymaking.

Second, mainstreaming requires transforming the organization of governing structures. By demanding that gender equality is addressed in all relevant policy sectors, gender mainstreaming calls for horizontal governance. Through this redistribution of responsibilities for equality objectives among a wider range of governing actors, it breaks away from the ‘policy ghetto’ of gender policies. Gender mainstreaming thus implies a shift from the vertical governance of gender equality by specialized governing structures to the horizontal governance of gender equality by general governing structures (Woodward, 2003: 68). Moreover, gender mainstreaming requires that the issue of gender equality is taken up at all levels of governance. Therefore, it reallocates competencies for defining policies that impact gender equality across levels of governance. Because of this twofold reorganization of the governing process (horizontally and vertically), policy coordination between policy areas and levels of governance is a central concern within the mainstreaming strategy (Lombardo, 2005: 417–418). A second component of EEM that can then be deduced is the horizontal and multi-level governance of ethnic equality, which requires coordination.

Third, mainstreaming necessitates the creation of expertise on gender equality. Since gender mainstreaming aims at making policy processes gender-conscious, the increase of gender expertise among actors involved in the development of general policies is essential to the implementation of the strategy (Beveridge and Nott, 2002: 301). This can either mean that the actors normally involved in the policy process are trained to incorporate the gender perspective in their policy field, or that gender experts are involved in the policy process. It also requires the improvement of knowledge about gender relations and the mechanisms that create gender inequalities, which is made available by research, statistics and forecasts. The monitoring of the outcomes of general policies for specific disadvantaged groups is an essential part of this process (Lombardo, 2005: 417–418; Woodward, 2003: 72). What follows is a third component of EEM: the creation of expertise on ethnic inequalities among the actors involved in general policymaking.

Based on the three consensual components of mainstreaming in the gender literature, we can now form our definition of EEM. Comparable to gender mainstreaming, ‘ideal-type’ EEM is about reorganizing the policy process in such a way that the specificities of people with a migrant background are taken into account in all policy fields and at all levels of governance. Defined as such, EEM requires a three-fold transformation of policymaking. First, the specific needs or interests of people with a migrant background must be integrated into all policy areas and at all governance levels. Second, ethnic equality must be coordinated between different policy areas and levels of government. Third, the expertise on ethnic inequalities has to be strengthened. By opening up the policy process to ethnic equality considerations, the strategy might best be classified as a ‘colour-conscious’ policy strategy (Bleich, 2003).

Is ethnic equality mainstreaming practiced in Europe?

To determine whether the mainstreaming strategy is indeed present in European policymaking as regards people with a migrant background, we will focus on policy developments in the Netherlands, France, the United Kingdom and Germany. The countries are alike in that they are all long-standing countries of immigration. We assume this to be a necessary condition for the emergence of EEM. We consider it more likely for EEM to come into being in countries in which people with a migrant background form a significant part of the citizenry. We therefore decided not to study newer immigration countries in Central and Eastern Europe or Southern Europe. While these four countries under scrutiny are similar with regard to their migration history, they are different in their state structure (federal versus unitary), the size of their immigrant electorates, the relative success of radical right wing parties in their elections, as well as the national models of immigrant integration that have been ascribed to them (Koopmans et al., 2005). These are central variables in explaining variance in immigrant integration policies. Within these cases, we concentrate on policies at the national and, where relevant, regional levels of governance. Our results may therefore be less applicable to the local level. While we are aware that local governments also have competences for integration and equality policy, a comparison between different localities, regions and countries is beyond the scope of this article.

To evaluate the presence of our three components of EEM in the policy practices of the selected countries, we check for the presence of three EEM Indicators. First, we look for the availability of policies specifically targeted at people with a migrant background within the general policy fields of education and employment – two crucial fields for reaching equality. This means that we will not consider the policies that fall under the direct responsibility of the minister responsible for immigrant integration, like integration courses, but analyze how general policy areas address equality issues for people of migrant descent. We will also not explicitly focus on policy instruments addressing the ethnic equality gap for newly arrived migrants (for example, specific reception classes for new migrants in schools), but concentrate on those policy measures that are targeted at established migrants and citizens of migrant origin. Finally, we do not verify whether the impact of general policies on people with a migrant background is assessed. We excluded this indicator, because a too broad and demanding set of indicators might lead us to reject the presence of EEM too easily. Second, we review whether policy coordination is organized between all policy areas relevant to foster ethnic equality. While vertical and horizontal policy coordination mechanisms are both important for the mainstreaming strategy to take shape, in pursuance of parsimony, we will focus on the presence of horizontal policy coordination mechanisms. Third, to check for the presence of expertise on ethnic inequalities, we will focus on the types of ethnic statistics available, as we assume this is a basic form of expertise without which the ethnic inequality gap cannot be measured. Hence, we do not verify, in this first explorative account, the existence of ethnic equality training and the involvement of ethnic equality experts in the decision-making process and policy implementation. We make this choice because we presume that policymakers and equality experts alike need data on the existence and the scale of ethnic equality gaps to assess whether a policy measure impacts ethnic equality and the extent to which a targeted policy intervention is needed.³ The choice for these three minimum indicators certainly does not lead to a perfect measurement of EEM. Yet, we propose it as a meaningful starting point for such an assessment and as an invitation to other scholars to further develop and expand our set of indicators and apply them to all relevant policy fields.

It is not the objective of this article to scale the absence, or the (stronger or weaker) presence of each of these indicators by means of quantitative indicators. Instead, we will unpack in a qualitative fashion (similar to assessments of gender mainstreaming (e.g. Mazey, 2001; Rees, 2005)) what we mean by ‘absence’ or ‘presence’ of the three EEM indicators, by extensively describing and motivating (with empirical data) whether, and to what degree, each of the indicators is present or absent. Moreover, since the presence or absence of EEM is not static, but evolved during the studied time frame (1990s–present), we will also indicate whether we observe a ‘trend towards less or more’ colour-consciousness in general policy areas (indicator 1), less or more horizontal policy coordination (indicator 2) and fewer or more ethnic statistics (indicator 3).

We build our analysis from a collection of primary and secondary data. For our first indicator, we used the data on ‘targeting needs’ in education policy and ‘targeted support’ in employment policy from the Migrant Integration Policy Index (MIPEX).⁴ We corroborated this information with original policy documents, case study reports from the UPSTREAM project⁵ (Upstream, 2015) and other secondary literature. We decided to use the MIPEX instead of other indices that measure migrant integration policies because its indicators on targeted policy measures correspond best with ours. Unlike the Multiculturalism Policy Index (MPI) (2010) and other indices, the MIPEX considers all forms of targeted policy measures including, but not limited to, classic multiculturalist policies such as mother tongue instruction for pupils with a migrant background in schools. To evaluate the presence of our second indicator, we relied upon primary data from expert surveys⁶ as well as policy documents and the UPSTREAM case study reports, complemented by secondary literature. The assessment for our third indicator is based on interviews⁷ with policy experts in Germany and the Netherlands combined with existing secondary literature.

The Netherlands

Until the 1990s, the ‘Dutch model’ of immigrant integration policy was generally regarded as the example of multiculturalist policymaking in which cultural and religious differences were accommodated by the state (Vermeulen and Penninx, 2000). From the late 1990s onwards, however, the Netherlands clearly turned away from its multiculturalist, group-oriented integration model. While immigrant integration became increasingly politicized, the political debate on immigrant integration shifted to the right. Over the last two decades, this has resulted in severe restrictions to immigrants’ entitlements, restrictive immigration policy and the establishment of obligatory integration courses and tests (Duyvendak and Scholten, 2010).

Whereas the Netherlands strongly advocated the adoption of mainstreaming as one of the EU’s Common Basic Principles on Integration Policy in 2000, we find no indication that, after 2000, our first EEM indicator was still fulfilled. Although the Netherlands did implement specific policies towards immigrants in general policy areas from 1979 (Penninx, 1984), during the last decade specific measures have increasingly been replaced by general policies, which do not take into account the specific challenges that people with a migrant background might face (see also: Huddleston et al., 2015). A clear example from the field of education is the new weighting system that has applied since 2006, in which migration background is no longer a factor in the allocation of extra subsidies to schools (Mulder, 2013). We observe a similar evolution in employment policies. The SAMEN Act, which was implemented in 1998 and aimed to stimulate employers to hire ethnic minorities, was abolished in 2004 (Guiraudon et al., 2005). This trend towards the abolishment of targeted support continued over the period from 2010 to 2014. During this last time frame, targeted language and employment programs, which have expanded in many Western European countries over recent years, were discontinued in the Netherlands (Huddleston et al., 2015). To sum up, while general policies took the specific challenges of migrants into account up to the early 2000s, we find, at present, no indication for our first EEM indicator in the Netherlands. The authors of MIPEX (Huddleston et al., 2015: 2) arrive at a similar conclusion by arguing that the Dutch approach to immigrant integration is ‘a policy of no policy’. Concerning the second indicator we find that, in comparison to other European countries, there are hardly any mechanisms for the coordination of immigrant integration policies in the Netherlands (Eurofund, 2015: 26–29). Whereas in the past, coordination of policies related to immigrant integration and immigration between the different ministries took place through the Interdepartmental Commission for Integration and Immigration,⁸ this commission was abolished in 2008 (Staatscourant, 2008). Moreover, the different ministers who have been responsible for immigrant integration since 2010 seem to be performing less of a coordinating role than their predecessors. Now, it seems to be only the local governments that have set up coordination mechanisms (Maan et al., 2014: 18, 20, 72). In conclusion, we see a trend towards less horizontal coordination mechanisms in the Dutch case.

In comparison to other European countries, statistics on the position of people with a migrant background in the Netherlands are widely available and institutionalized (Jacobs and Rea, 2009). In the first place, ethnic data are collected through the Municipal Personal Records Database (Basisregistratie Personen, BRP) in which people with a migrant background are identified through their country of birth and that of their parents.⁹ These data are anonymized and made available to researchers and the larger public by the Central Bureau for Statistics (CBS). In addition, there are many large-scale surveys available that monitor the position of people with a migrant background in society, in particular, in education and on the labour market. An example is the Survey Integration Minorities (Survey Integratie Minderheden, SIM¹⁰), commissioned by the Dutch government and carried out by the Netherlands Institute for Social Research (Sociaal Cultureel Planbureau, SCP) and the CBS (Expert interview SCP, 16 June 2016). Thus, our third EEM indicator is still present in the Netherlands.

France

The French approach to immigrant integration has often been categorized as civic assimilationist (Brubaker, 1990; Favell, 1998). The ‘colour-blind’ French model of immigrant integration is based on the republican ideal of equal treatment of all French citizens which, in practice, entails a rejection of every form of differential treatment on the basis of ethnicity or religion. While citizenship is granted automatically to most people who are born in France, citizens are expected to endorse republican values in the public sphere and restrict specific cultural and religious practices to the private sphere (Favell, 1998).

In regard to our first EEM indicator, we find that particular barriers to equality faced by populations of immigrant origin are rarely being addressed explicitly in the general policy areas in France. Implicitly, however, general policies, including employment and education policies, do target immigrant populations. They do so by focusing on ‘priority neighborhoods’ in which these populations are overrepresented (Bénabou et al., 2009; Bozec and Simon, 2014: 5, 11, 13). Apart from language support to newcomers, education policy does not address the specific needs of pupils with a migrant background (Bozec and Simon, 2014: 22–26; Escafré-Dublet, 2014: 8–11; Huddleston et al., 2015). Also in employment policies, specific support to immigrant populations is only available for newly arrived immigrants (Bozec and Simon, 2014: 48). Compared to other European countries, targeted support for people with a migrant background is weak in France, where measures such as specially trained teachers and work-specific language courses are non-existent (Huddleston et al., 2015). Because of the absence of these colour-conscious policies, our first EEM indicator is thus clearly not met in France.

Arriving at our second EEM indicator, we notice that some mechanisms for the horizontal coordination of immigrant integration policies have recently been set up. Already in 1990, an inter-ministerial committee for immigrant integration was created. However, this committee did not have permanent status and has only met sporadically (Bozec and Simon, 2014: 14–15; Escafré-Dublet, 2014: 6). Yet, in 2014, the government established a new structure for the coordination of immigrant integration policy. The ‘inter-ministerial delegate for integration and republican equality’ has permanent status and works under the prime minister’s authority (Bozec and Simon, 2014: 19). At the regional and local levels, there are also some programs in place for the coordination of immigrant integration policy. The Programmes Régionaux pour l'Intégration des Populations Immigrée (PRIPI) and more recent Plans Départementaux d'Intégration (PDI) aim to unite different stakeholders and promote the consideration of immigrant integration across policy sectors (Bozec and Simon, 2014: 15). With the recent establishment of these horizontal policy coordination mechanisms, we thus see a trend towards more horizontal coordination (our second EEM indicator) in the French case.

Because of the republican principle of equal treatment and historic events, ethnic statistics are taboo in France (Bleich, 2003). Nationality at birth and country of birth are registered in the census. This allows the distinction between French citizens by birth and French citizens by naturalization. While this provides partial ethnic data, it ‘invisibilizes’ the study of ethnic equality since it excludes the increasingly numerous second and third generations who are French citizens from birth (Simon, 2008: 1). However, since the 2000s, discrimination prominently entered the public debate and on the political agenda. Partly because of the 2005 riots and the voice of black French citizens, the previously hegemonic colour-blind policy strategy and the absence of ethnic statistics has become questioned by researchers, anti-racist movements and policymakers alike (Simon, 2008: 8–9). This gradual change led to an increase of alternative ethnic monitoring methods, in particular, by the National Institute for Demography (INSEE, later INED), thereby circumventing the still strong opposition of political elites to introduce ethnic categories in the census or in generalized population registers (Tribalat, 2016). For example, extensive ethnic equality data have become available from 2010 onwards through a large-scale survey conducted by the French state’s institutes for statistics and demography (INED and INSEE), called ‘Trajectories and Origins’,¹¹ which does not only identify the place of birth of French residents, but also the nationality at birth and country of birth of the parents (Beauchemin et al., 2010, 2016; Tribalat, 2016). In short, while state statistics still refrain from collecting ethnic data, the gradual acceptation of alternative ethnic monitoring tools by state institutions since the 2000s can be considered as a very modest evolution in the direction of ethnic monitoring (our third EEM indicator).

The United Kingdom

The British ‘model of immigrant integration’ is often referred to as a ‘race relations model’, within which strong anti-discrimination laws are coupled with sharp curbs on immigration (Favell, 1998). Shaped from the 1960s onwards, the British approach to immigrant integration consisted mostly of multiculturalist policies that respect and promote cultural pluralism, and anti-discrimination policies that aim to combat ethnic or racial inequalities with a focus on ethnic minorities instead of newly arriving immigrants (Bleich, 2003). Yet, the 2000s were marked by an increased emphasis on ‘community cohesion’. This new focus resulted in the gradual replacement of multiculturalist policies by policies that aim to create shared bonds and values between different communities (Saggar and Somerville, 2012; Vertovec and Wessendorf, 2010).

With regard to our first EEM indicator, we consider the traditional British race relations policies to fit well with the mainstreaming strategy, as they are general policies that take into account the specific needs of people with a migrant background. However, the attention for ethnic equality issues seems to have decreased under the conservative–liberal democrat coalition of 2010 and the 2015 conservative majority government. The government abolished the (ethnic) equality impact assessments, which used to be compulsory (Huddleston et al., 2015). Policies that are specifically targeted at migrant or ethnic minority populations are increasingly being rejected (Ali and Gidley, 2014: 1, 8). For example, measures aimed at improving the labour market integration of people with a migrant background were withdrawn. In fact, while most European countries increased targeted support for people with a migrant background, in the period 2011 to 2012, many such targeted measures, including work-specific language courses, came to a halt in the United Kingdom (Huddleston et al., 2015). Also, the ‘Ethnic Minority Achievement Grant’, which specifically targeted ethnic minority students, was abolished in 2011. However, targeted measures in education policy were partly recovered again in 2013 by allocating extra funding to bilingual students (Huddleston et al., 2015). Nevertheless, the possibility of using positive action for ethnic minorities and other historically disadvantaged groups has been expanded (without becoming a requirement) under the new Equality Act of 2010 (Equality and Human Rights Commission, 2016). Altogether, we deem our first EEM indicator to still be present in the United Kingdom, but recognize that the support for colour-conscious policies is declining.

In the United Kingdom, policy responsibility for immigrant integration has always been dispersed between different governing structures. Policies that are relevant for immigrant integration are designed in isolated ‘silos’ (Expert Survey United Kingdom, 2014) and no department seems to take the lead in horizontal policy coordination, which results in fragmented policy responses (Saggar and Somerville, 2012). In 2012, the government launched its new integration strategy which delegates much of the responsibility for immigrant integration to the local level and society at large (Ali and Gidley, 2014: 3, 5; Huddleston et al., 2015) without providing any coordination mechanisms. Therefore, we conclude that we cannot confirm the presence of our second EEM indicator in the British case.

In the British context, ethnic statistics are less controversial and practiced more than in many other European countries. The census identifies people with a migrant background on the basis of self-ascribed ethnic or racial categories (Jacobs et al., 2009). Moreover, authorities are required to monitor the use of public services by different ethnic minorities (Ali and Gidley, 2014: 10). However, monitoring programs are equally the object of austerity cuts (Huddleston et al., 2015). Nevertheless, a culture of monitoring and evaluation still seems to exist. While the equality impact assessments are no longer mandatory, public administrations still seem to use them (Ali and Gidley, 2014: 10). Hereby, we find clear evidence for our third EEM indicator in the United Kingdom.

Germany

German citizenship policies have long been characterized as ‘ethno-cultural’ (Brubaker, 1990). Until the end of the 1990s, Germany mainly granted citizenship on the basis of descent (Winter, 2014). Whereas immigrant integration policies have existed in previous decades, it was only in the year 2000 that Germany officially recognized its position as an immigrant destination, reformed its citizenship act and gradually developed explicit immigrant integration policies.

Although policies specifically targeted at migrant populations have, every so often, been criticized for their stigmatizing effect, these specific policies continue to exist. Overall, there seems to exist a mix of specially targeted policies and general policies which target the population at large (Bendel, 2014: 4–5). In some regions, for example, language classes in the mother tongue are provided to pupils with a migrant background as part of the regional education policy (Expert interview Ifpol, 13 June 2016). Policy measures oriented towards the interculturalization of school curricula form another example from the policy area of education (Schönwälder, 2010: 158–160). In the area of employment, and in comparison with other EU countries (Huddleston et al., 2015), Germany offers relatively much tailored support to people with a migrant background, for instance, in the form of work-specific language courses. Yet, many of these targeted policy measures are aimed more towards newcomers than to established migrants or citizens with a migrant background. Nevertheless, some programs are also targeted towards the latter group. For example, there is the ‘Jobstarter Kausa project’, which aims to promote the participation of migrant entrepreneurs in the German apprenticeship system (Jobstarter, 2016). Moreover, some regions have recently launched initiatives to promote more diverse public administrations by trying to increase the number of people with a migrant background working in the public sector (Huddleston et al., 2015). Thus, the specific focus on immigrant populations in general policy areas (our first EEM indicator) is present in Germany’s immigrant integration policy, albeit to a limited and varying degree (and depending on the region).

Regarding our second EEM indicator we remark that, compared to other European countries, Germany has established (relatively) numerous structures for the coordination of policies affecting immigrant integration (Eurofund, 2015: 26–29). At the federal level, since 2005, a special Commissioner for Migration, Refugees and Integration has been appointed as a Minister of State in the chancellery (Expert Survey Germany, 2014). This Commissioner, who was formerly associated with the ministries of labour and family, is responsible for the coordination of all integration efforts across different policy sectors and the monitoring of regional and local governments’ accomplishments in immigrant integration. Moreover, the Inter-Ministerial Working Group on Migration and Integration (IMAG) was also tasked with this coordination function. However, this working group has not been functioning since 2009 (Bendel, 2014: 5–7; Collett and Petrovic, 2014: 9). At the regional level, there are inter-departmental committees for immigrant integration in most länder. In short, we can conclude that our second EEM indicator is clearly present in the German case.

With respect to the final component of EEM as defined in this article – the creation of expertise – it should be noted that, traditionally, the collection of ethnic data is just as controversial in Germany as it is in France (Givens and Evans Case, 2014; Jacobs and Rea, 2009). Consequently, the German population register distinguishes only between people on the basis of their nationality and country of birth. However, in order to gain data on the second generation, the German government added a question to the microcensus in 2005 (Rühl, 2009: 133). The microcensus survey records information on the migration background of a representative (1%) sample of the German population, by registering the place of birth of people’s parents (Simon, 2012: 1371). Another large-scale survey is the socio-economic panel or ‘SOEP’, which includes immigrant populations as from 1994, but does not identify people of immigrant origin who have attained citizenship. Furthermore, the Ministry of Education collects data from school pupils of immigrant origin on the language spoken at home (Expert interview Ifpol, 13 June 2016). Official statistics on ethnic inequality in the labour market are, however, non-existent and few research studies have been carried out so far (Simon, 2007: 58–60). Therefore, for our third EEM indicator, similarly as for France, we can only speak of a modest trend towards more ethnic statistics in Germany.

Mainstreaming practices compared

The foregoing discussion on developments in immigrant integration policymaking across different European countries shows that there is no clear presence, nor a trend towards EEM in the countries compared. For our first EEM indicator, the analysis reveals that there is a trend towards less support for policies that are sensitive to the particular thresholds to equality for people with a migrant background within general policy areas. Instead of policy measures specifically targeted at people with a migrant background, we rather perceive a shift towards colour-blind policies. Thereby, the special attention to ethnic inequalities within the mainstream policy process, which is required by the mainstreaming strategy, appears to be largely absent. Moreover, even in former multicultural countries such as the Netherlands and the United Kingdom, we observe a trend towards more colour-blind policies with an increasing reliance on area-based policy measures and a broadening of target groups of specific policies to socio-economically disadvantaged groups in general. Our second EEM indicator can also not univocally be confirmed in our cases. Horizontal policy coordination mechanisms are limited, although there exists some variation between the different countries and levels of governance. While in Germany (relatively) many such mechanisms are in place, in the Netherlands and the United Kingdom, there rather seems to be a move away from horizontal policy coordination on immigrant integration. Concerning our third EEM indicator, we notice a slight trend towards more ethnic monitoring, not so much through the introduction of ethnic categories in the census (which still exist in the United Kingdom) or population registers (like in the Netherlands) but through the set-up of large-scale ethnic minorities surveys, even in formerly reluctant countries such as Germany and France.

To conclude, whereas we do not find a trend towards EEM in these countries, our EEM indicators allowed us to uncover the existence of a double and seemingly paradoxical trend in immigrant integration policies in Western Europe. This ‘new style’ immigrant integration policy can be depicted as follows: increasing colour-blind policies (particularly in the Netherlands and the United Kingdom) coupled with increased ethnic monitoring (or the continuity of ethnic monitoring in the Netherlands and the United Kingdom, and a trend towards more ethnic monitoring in countries that were previously opposed to it, namely France and Germany). In other words, states increasingly monitor the impact of ‘doing nothing’. Our comparative findings are summarized in Table 1.

Table 1.

Results empirical analysis of ethnic equality mainstreaming in Europe.

	Integrating the specific in the general: Specific policy measures in general policy fields	Transforming governance structures: Horizontal policy coordination	Creating expertise: Availability of ethnic statistics
The Netherlands	Trend towards less	Trend towards less	Present
France	Absent	Trend towards more	Trend towards more
The United Kingdom	Trend towards less	Absent	Present
Germany	Present (but modestly and varying per region)	Present	Trend towards more

Conclusion and discussion

Immigrant integration policies in Europe are constantly evolving. After the ‘multiculturalism backlash’ and the ‘assimilationist turn’, it is mainstreaming that increasingly pops up in policy documents and literature as a seemingly new policy trend. In this article, we tried to contribute to this awakening debate by developing a conceptualization of EEM and proposing a set of ‘EEM Indicators’, built from established theory in the literature on gender mainstreaming. We are convinced that this contribution to the conceptualization of EEM will enhance theory development across different fields of equality policies. Political science communities on ethnic or racial equality, on the one hand, and on gender and other forms of equality on the other, have traditionally sat at different tables. This paper relied on the gender equality literature to develop a conceptualization of EEM. We hope to have set the stage for further cross-fertilization between these distinct research communities. Future research could enhance theory building on equality policies by comparing similarities and differences in policy evolutions in different types of equality policies. While there seem to be similar backlashes against gender equality and mainstreaming as against ethnic equality and mainstreaming (Hubert and Stratigati, 2016; Kuhar and Paternotte, 2017), scholars have paid little heed to comparing underlying mechanisms. Before doing so, however, in order not to hamper theory-building, similar indicators of ethnic and gender equality policies are needed. We hope to have taken a first step in that direction.

Moreover, this article also shows the increasing need of applying the North-American categorization of racial equality policies, namely the distinction between colour-blind and colour-conscious policies, to Europe (Bleich, 2003). European immigration countries count a growing share of citizens with a migrant background. These citizens cannot be considered immigrants anymore. However, they still face barriers to equality. Therefore, there is a need for different categorizations of ethnic equality policies than those used to distinguish between types or ‘models’ of immigrant integration policies.

In the second part of the article, we empirically verified the absence or presence of an EEM strategy in four Western European immigration countries. Even though mainstreaming is inscribed in the EU’s Common Basic Principles of integration policy, we demonstrated that there is no clear trend towards EEM in these countries. However, our conceptualization of EEM allowed us to detect a different, and new, trend in immigrant integration policies. This trend involves a seemingly paradoxical combination of an increasing reliance on colour-blind policies and a rise in ethnic monitoring. Given that colour-blind policies reject targeted policy measures for ethnic minorities, and that ethnic statistics are such a targeted policy tool, the combination of both trends seems surprising.

Recent research supports the accuracy of our conclusions. The MIPEX (Huddleston et al., 2015) also detects a withdrawal of targeted support for migrant populations in formerly multiculturalist countries, such as the United Kingdom and the Netherlands. Other research (Adam, 2013; Adam and Van Dijk, 2015) points at similar trends in Belgium, with the withdrawal of targeted support programs for migrants (in employment) in formerly multiculturalist Flanders and the formerly colour-blind Walloon region increasingly opening up to ethnic statistics. While previous research has argued that the ‘multiculturalism backlash’ affects discourse more than policies (Vertovec and Wessendorf, 2010), more recent policy evolutions such as the ones described earlier seem to show that the backlash, in particular, in the United Kingdom and the Netherlands, has moved from discourse to concrete policy change. Other than ours, these related research projects, however, measure the evolution regarding multiculturalism (MPI, Multiculturalism Policy Index, 2010) or policy inclusiveness (MIPEX, Huddleston et al., 2015). These measurements partially overlap with our first EEM indicator (the existence of colour-conscious policy measures within general policies), but do not include evolutions regarding policy coordination (EEM indicator 2) or ethnic statistics (EEM indicator 3). The added value of our measurement of a possible trend towards EEM is twofold. Firstly, as highlighted above, it allows comparison of similar trends over different types of equality policies. Secondly, it triggers new avenues for research, trying to explain the seemingly paradoxical policy trend, of less targeted support and more ethnic statistics, that we detected.

Because of our focus on four old immigration countries in Western Europe, we cannot come to a definitive and generalizable conclusion for all of Western Europe. Future research should investigate whether a similar trend can be observed in other and newer Western European immigration countries like Italy and Spain. Our research findings for the older immigration countries in Northwestern Europe can also not be extended to all policy levels. British research on racial equality policies from the 1990s (Ball and Solomos, 1990) and immigrant integration research from the 2000s onwards (Caponio and Borkert, 2010; Zapata-Barrero et al., 2017) have shown that policy frames at the local level can diverge from, and even contradict, national policies. Future research should thus investigate if similar trends as those detected and reported in this article at the national (and regional) levels can also be found at the local, and possibly international levels. Research on local (Jensen, 2015; Van Breugel et al., 2015) and European immigrant integration policies (Veloso Leão et al., 2017) seems to indicate that EEM does survive at the local level, and is initiated at the EU level.

Our empirical results also show the need for a more nuanced theorizing on the evolution of immigrant integration policies in Europe. They seem to confirm neither the ‘national models’ (Brubaker, 1990; Koopmans et al., 2012) nor the ‘convergence’ (Joppke, 2007; Soysal, 1994) hypotheses familiar in the immigrant integration policy literature. They are nuanced, as they show that both theses are worth defending. Our detected trend towards more colour-blind policies and more ethnic monitoring reveals that countries gradually move away from their former national models (with the Netherlands and the United Kingdom becoming more colour-blind; and France and Germany getting acquainted with ethnic statistics), but are at the same time characterized by path dependency (with the continuous use of ethnic statistics in the Netherlands and the United Kingdom as well as the continuous colour-blind policies in France). This non-straightforward policy evolution asks for complex explanations, which cannot be developed in this article.

However, to spark future academic debates, we would like to very briefly suggest one amongst many rival explanations as a new avenue for future research. Our interpretative key goes beyond a combination of existing explanations for policy shifts in the field of immigrant integration (for an overview, see: Koopmans et al., 2012). In order to explain both elements of our observed trend together, namely the concurrent trend towards colour-blind policies and ethnic monitoring, and not as two separate trends, it might prove fruitful to distinguish between ‘ideological colour-blindness’ and a more ‘strategic colour-blindness’. Whereas an ‘ideological colour-blind’ policy frame would reject ethnic categories as a legitimate factor in equality policymaking, since it is believed that ethnic categories reify differences, a ‘strategic colour-blind’ policy frame would not reject these ethnic categories on ideological grounds. Instead, strategic colour-blindness, which we find to be increasingly present in the Netherlands and the United Kingdom, rejects policies targeting ethnic minorities for strategic reasons. Policymakers do not want the public to think that specific efforts are made to promote ethnic equality. The detected trend towards the retreat of policies that are specifically targeted at people of immigrant origin coupled with an increase in ethnic monitoring, might then be regarded as a trend towards ‘strategic colour-blindness’, in which ethnic monitoring is still considered a relevant practice, but, at the same time, politically contested colour-conscious policies form too much of a political risk. Explained along these lines, our detected trend might not be so paradoxical after all in a context of intense politicization of immigrant integration.

Footnotes

Acknowledgements

The authors would like to thank the three anonymous reviewers for their thoughtful and constructive comments. We also acknowledge the excellent feedback we received during the MAM (Migration Asile Multiculturalisme) Seminar of 29 April 2016 at the Université Libre de Bruxelles (ULB), in which a first version of this article was discussed. Finally, our thanks go to Saskia Bonjour, Maurice Crul, Dirk Jacobs, Christof Roos and Augusto Veloso Leão for their valuable comments on earlier versions of this article.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

Notes

References

Adam I (2013) Immigrant integration policies of the belgian regions: Sub-state nationalism and policy divergence after devolution. Regional & Federal Studies 23(5): 547–569.

Adam

Van Dijk

(2015) Renforcer la Coopération Institutionnelle pour améliorer l’Accès des Personnes issues de l’Immigration au Marché du Travail. Brussels: King Baudouin Foundation.

Ali

Gidley

(2014) Advancing Outcomes for all Minorities: Experiences of Mainstreaming Immigrant Integration Policy in the United Kingdom. Brussels: Migration Policy Institute Europe.

Ball

Solomos

(1990) Race and Local Politics. Basingstoke: Palgrave Macmillan.

Banting

and Kymlicka

(2012) Is There Really a Backlash against Multiculturalism Policies? New Evidence from the Multiculturalism Policy Index. GRITIM Working Paper Series 14. Barcelona: Pompeu Fabra University.

Beauchemin

Hamel

Simon

(2010) Trajectories and Origins: Survey on Population Diversity in France. Paris: Institut National d'Études Démographiques.

Beauchemin

Hamel

Simon

(2016) Les Discriminations Racistes Existent et Résistent. Paris: Libération.

Bell

(2008) Racism and Equality in the European Union. Oxford: Oxford University Press.

Bénabou

Kramarz

Prost

(2009) The French Zones d’Education Prioritaire: Much Ado about Nothing? Economics of Education Review 28: 345–356.

10.

Bendel

(2014) Coordinating Immigrant Integration in Germany: Mainstreaming at the Federal and Local Levels. Brussels: Migration Policy Institute Europe.

11.

Beveridge

Nott

(2002) Mainstreaming: A case for optimism and cynicism. Feminist Legal Studies 10(3): 299–311.

12.

Bleich

(2003) Race Politics in Britain and France: Ideas and Policymaking Since the 1960s. Cambridge: Cambridge University Press.

13.

Booth

Bennett

(2002) Gender mainstreaming in the European Union: Towards a new conception and practice of equal opportunities. The European Journal of Women's Studies 9(4): 430–446.

14.

Bozec

Simon

(2014) The Politics of Mainstreaming Immigrant Integration Policies. Case Study of France. Paris: Institut National d'Études Demographiques.

15.

Brubaker

(1990) Immigration, citizenship, and the nation-state in France and Germany: A Comparative Historical Analysis. International Sociology 5(4): 379–407.

16.

Caponio

Borkert

(2010) The Local Dimension of Migration Policymaking. Amsterdam: Amsterdam University Press.

17.

Collett

Petrovic

(2014) The Future of Immigrant Integration in Europe: Mainstreaming Approaches for Inclusion. Brussels: Migration Policy Institute Europe.

18.

Durkheim

(1982) The Rules of Sociological Method. New York: The Free Press.

19.

Duyvendak

Scholten

(2010) Beyond the Dutch ‘Multicultural Model’. Journal of International Migration and Integration 12: 331–348.

20.

Equality and Human Rights Commission (2016) Positive cction – Your opportunity to advance equality. Available at: www.equalityhumanrights.com/en/legal-work-scotland/legal-work-scotland/scottish-legal-articles/positive-action-–-your-opportunity (accessed 7 December 2016).

21.

Escafré-Dublet

(2014) Mainstreaming Immigrant Integration Policy in France. Brussels: Migration Policy Institute Europe.

22.

Eurofund (2015) Challenges of Policy Coordination for Third-country Nationals. Luxembourg: Publication Office of the European Union.

23.

Expert Survey Germany (2014) Country Expert Survey for the Multi-Level Governance of Integration Project (KBS MLG). Survey Conducted With Dietrich Thränhardt, University of Münster. Brussels: King Baudouin Foundation.

24.

Expert Survey United Kingdom (2014) Country Expert Survey for the Multi-Level Governance of Integration Project (KBS MLG). Survey Conducted With Eve Hepburn, University of Edinburgh. Brussels: King Baudouin Foundation.

25.

Favell

(1998) Philosophies of Integration: Immigration and the Idea of Citizenship in France and Britain. Basingstoke: Palgrave Macmillan.

26.

Gerring

(1999) What makes a concept good? A criterial framework for understanding concept formation in the social sciences. Polity 31(3): 357–393.

27.

Givens

Evans C

(2014) Legislating Equality. The Politics of Antidiscrimination Policy in Europe. Oxford: Oxford University Press.

28.

Glazer

(1998) We Are All Multiculturalists Now. Cambridge: Harvard University Press.

29.

Guiraudon

Phalet

Ter

(2005) Monitoring Ethnic Minorities in the Netherlands. International Social Science Journal 57(183): 75–87.

30.

Hubert

Statigati

(2016) Twenty years of EU gender mainstreaming: Rebirth out of the ashes? Femina Politica – Zeitschrift Für Feministische Politikwissenschaft 2: 21–36.

31.

Huddleston

Bilgili

Joki

et al . (2015) Migrant Integration Policy Index 2015. Barcelona/Brussels: CIDOB, MPG.

32.

Jacobs

Rea

(2009) ‘Allochthones’ in the Netherlands and Belgium. International Migration 50(6): 42–57.

33.

Jacobs

Swyngedouw

Hanquinet

et al . (2009) The challenge of measuring immigrant origin and immigration-related ethnicity in Europe. Journal of International Migration and Integration/Revue De L'integration Et De La Migration Internationale 10(1): 67–88.

34.

Jensen

(2015) Mainstreaming in Practice: The Efficiencies and Deficiencies of Mainstreaming in the UK. Oxford: COMPAS.

35.

Jobstarter (2016) Koordinierungsstelle Ausbildung und Migration: Ausbildung jetzt! Available at: www.jobstarter.de/kausa (accessed 7 December 2016).

36.

Joppke

(2007) Beyond national models: Civic integration policies for immigrants in Western Europe. West European Politics 30(1): 1–22.

37.

Koopmans

Michalowski

Waibel

(2012) Citizenship rights for immigrants: National political processes and cross-national convergence in Western Europe, 1980-2008. American Journal of Sociology 117(4): 1202–1245.

38.

Koopmans

Statham

Giugni

et al . (2005) Contested Citizenship: Immigration and Cultural Diversity in Europe. Minneapolis: University of Minnesota Press.

39.

Kuhar

Paternotte

(2017) Anti-Gender Campaigns in Europe Lanham: Rowman & Littlefield.

40.

Lombardo

(2005) Integration or setting the agenda? Gender mainstreaming in the European constitution-making process. Social Politics: International Studies in Gender, State and Society 12(3): 412–432.

41.

(2003) Deconstructing Canada’s discourse of immigrant integration. Journal of International Migration and Integration / Revue De L'integration Et De La Migration Internationale 4(3): 315–333.

42.

Maan

Van Breugel

Scholten

(2014) The Politics of Mainstreaming Immigrant Integration Policies Case Study of The Netherlands. Rotterdam: Erasmus University Rotterdam.

43.

Mazey

(2001) Gender Mainstreaming in the EU: Principles and Practice. London: Kogan Page Limited.

44.

Meier

Celis

(2011) Sowing the seeds of its own failure: Implementing the concept of gender mainstreaming. Social Politics: International Studies in Gender, State & Society 18(4): 469–489.

45.

Mulder

(2013) Allochtone Leerlingen en Onderwijsachterstand. De Ontwikkeling van Specifiek naar Generiek Beleid. In: Coello

Dagevos

Huider

et al . (eds) Het Minderhedenbeleid Voorbij: Motieven En Gevolgen. Amsterdam: Amsterdam University Press, pp.91–113.

46.

Multiculturalism Policy Index (2010) The multiculturalism index for immigrant minorities. Available at: www.queensu.ca/mcp/ (accessed 16 August 2017).

47.

Penninx

(1984) Research and policy with regard to ethnic minorities in the Netherlands; An historical outline and the state of affairs. International Migration (Geneva, Switzerland) 22(4): 345–366.

48.

Rees

(2005) Reflections on the uneven development of gender mainstreaming in Europe. International Feminist Journal of Politics 7(4): 555–574.

49.

Rühl

(2009) Germany. In: Fassmann

Reeger

Sievers

(eds) Statistics and Reality: Concepts and Measurements of Migration in Europe. Amsterdam: Amsterdam University Press, pp.131–150.

50.

Saggar

Somerville

(2012) Building a British Model of Integration in an Era of Immigration: Policy Lessons for Government. Brighton: University of Sussex and the Migration Policy Institute.

51.

Scholten

Collett

Petrovic

(2016) Mainstreaming migrant integration? A critical analysis of a new trend in integration governance. International Review of Administrative Sciences 83(2): 283 - 302.

52.

Scholten

Van Breugel

(2017) Mainstreaming Integration Governance New Trends in Migrant Integration Policies in Europe. Basingstoke: Palgrave Macmillan.

53.

Schönwälder

(2010) Integration policy and pluralism in a self-conscious country of immigration. In: Vertovec

Wessendorf

(eds) The Multiculturalism Backlash. European Discourses, Policies and Practices. London: Routledge, pp.152–169.

54.

Simon

(2007) ‘Ethnic’ Statistics and Data Protection in the Council of Europe Countries. Study Report. Strasbourg: Council of Europe.

55.

Simon

(2008) The choice of ignorance: The debate on ethnic and racial statistics in France. French Politics, Culture & Society 26(1): 7–31.

56.

Simon

(2012) Collecting ethnic statistics in Europe: A review. Ethnic and Racial Studies 35(8): 1366–1391.

57.

Soysal

(1994) Limits of Citizenship: Migrants and Postnational Membership in Europe. Chicago: University of Chicago Press.

58.

Squires

(2005) Is mainstreaming transformative? Theorizing mainstreaming in the context of diversity and deliberation. Social Politics: International Studies in Gender, State and Society 12(3): 366–388.

59.

Squires

(2008) Intersecting inequalities: Reflecting on the subjects and objects of equality. The Political Quarterly 79(1): 53–61.

60.

Staatscourant (2008) Regeling van de Ministerie voor Wonen, Wijken en Integratie van 28 November 2008, nr. BJZ2008111822, tot intrekking van diverse regelingen. Nr. 241, 11 December 2008. Den Haag: Koninkrijk der Nederlanden.

61.

Tribalat

(2016) Les Statistiques Ethniques. Une Querelle Bien Française. Paris: Editions du Toucan.

62.

Tribalat

Riandey

Simon

(1992) Mobilité Géographique Et Inserion Sociale. Paris: Institut National d'Études Démographiques.

63.

Upstream (2015) Mainstreaming Integration Governance. Available at: http://project-upstream.eu (accessed 21 December 2016).

64.

Van Breugel

Maan

Scholten

(2014) Conceptualizing Mainstreaming in Immigrant Integration Governance. A Literature Review. Rotterdam: Erasmus University Rotterdam.

65.

Van Breugel

Maan

Scholten

(2015) Mainstreaming in Practice: The Efficiencies and Deficiencies of Mainstreaming in the Netherlands. Rotterdam: Erasmus University.

66.

Van Breugel

and Scholten

(2017) Mainstreaming in response to superdiversity? The governance of migration-related diversity in France, the UK and the Netherlands. Policy and Politics 45(4): 511-526.

67.

Veloso Leão

Westerveen

Adam

(2017) The first year of implementation of the EU action plan on integration: An evaluation of European policy coordination on migrant integration. IES Policy Brief Issue 2017/2: 1-4.

68.

Verloo

(2001) Another Velvet Revolution? Gender Mainstreaming and the Politics of Implementation. IWM Working Paper No. 5/2001. Vienna: IWM.

69.

Vermeulen

Penninx

MJA

(2000) Immigrant Integration:The Dutch Case. Amsterdam: Het Spinhuis.

70.

Vertovec

Wessendorf

(2010) The Multiculturalism Backlash. European Discourses, Policies and Practices. London: Routledge.

71.

Walby

(2005) Gender mainstreaming: Productive tensions in theory and practice. Social Politics 12(3): 321–343.

72.

Weber

(1978) Economy and Society: An Outline of Interpretive Sociology. Oakland: University of California Press.

73.

Wimmer

(2008) The making and unmaking of ethnic boundaries: A multilevel process theory. American Journal of Sociology 113(4): 970–1022.

74.

Winter

(2014) Traditions of nationhood or political conjuncture? Comparative Migration Studies 2(1): 29–55.

75.

Woodward

(2003) European gender mainstreaming: Promises and pitfalls of transformative policy. Review of Policy Research 20(1): 65–88.

76.

Zapata-Barrero

Caponio

Scholten

(2017) Theorizing the ‘Local Turn’ in a multi-level governance framework of analysis: A case study in immigrant policies. International Review of Administrative Sciences 83: 241–246.