Stalled integration? How experience with anti-immigrant expressions and ethnic neighbourhoods affect immigrants’ political integration

Aim

The general aim of this project is to test the influence of the surrounding environment on immigrant integration during the “integration window”, i.e., the initial years after immigration. More precisely, we test the impact of a) anti-immigrant expressions (in particular electoral support for anti-immigrant parties), and b) ethnic neighbourhoods, on immigrants’ political integration.

Immigrant integration is a key concern in contemporary European societies. Successful integration makes it possible to harness the opportunities related to migration, while failed integration is related to a number of societal problems. Yet, we have surprisingly little knowledge about to what extent the society in which migrants are expected to integrate assists or, rather, prevents successful integration.  Previous research suggest that the period after arrival represents an “integration window” in which immigrants may be open to habit change, and that the surrounding context can play an important role during this period (Ferwerda et al., 2020; Hainmueller et al., 2017).  Despite this, current research largely fails to capture the experiences of immigrants during the “integration window”. Moreover, despite evidence that the surrounding contexts plays an important part in the integration process, integration research remains focused on individual rather than contextual explanatory factors. The purpose of our 3-year multidisciplinary project is to fill this gap by answering how political integration in Sweden is influenced by the immigrants’ surrounding environment during this initial integration window.

While most previous research focus on the economic integration of immigrants, our project focuses on political integration. Political integration ranges from a sense of belonging to the new national community (national identity), to willingness to become a citizen of the new country (naturalization), to engaging politically in the new country (e.g. voting). Political integration is crucial for social cohesion and democratic legitimacy. Indeed, it has been argued that national identities can function as a 'glue' that holds culturally diverse societies together (Miller and Ali, 2014). In addition, prior work find citizenship to be an important catalyst for long-term economic and social integration (Hainmueller et al, 2017; 2019). Political integration is thereby essential also in order to increase other types of integration.

We test the effect of two very different contextual factors on political integration: anti-immigrant expressions and ethnic neighbourhoods. These areas have attracted significant attention in the societal and political debate, but when it comes to their effect on integration we still lack systematic knowledge on key issues. For example, how do newly arrived immigrants react if met with hostility in the new country: do they hold on even closer to their old national identity – either emotionally (sense of belonging, reluctance to engage with the new society) or formally (citizenship) - or does it cause them to embrace the national identity of the host society, to avoid being the targets of hostility? And how does it affect the integration of immigrants to live in areas with many co-nationals the first years after arrival; does this function as a bridge to the new society, or is it rather an obstacle in the way of engaging with the new society?

The lack of systematic investigations of, and firm answers to, these questions is all the more surprising given the fact that both anti-immigrant sentiment and ethnic neighbourhoods feature prominently in the political debate, and given that previous research consider them as two of the key social contextual factors related to immigrants’ political integration. To take one example: while many have studied the rise of anti-immigrant parties, few have studied how this expression of anti-immigrant sentiment among voters affect the targets – migrants. Moreover, while some – including one of the members of this project – have studied the effect of ethnic neighbourhoods on economic integration (Martén et al. 2019), we know very little about their effect on political integration and therefore their impact on social cohesion and democratic legitimacy.

The research literature, so far, displays conflicting theoretical expectations and inconclusive empirical findings. We suggest that conflicting theoretical expectations may, to a large extent, be due to a lack of attention to the different characteristics of different types of political integration, which range from the very emotional to the very practical, and are surrounded by very different incentive structures. As for the conflicting empirical results in the literature, we suggest that these are, to a significant extent, a result of inadequate data and research designs. In particular, the field suffers from a severe lack of information about the most relevant group – newly arrived immigrants. Moreover, most research designs rely on observed correlations, which makes it difficult to draw conclusions about causality.

Our multidisciplinary project, combining political science and economics, uses novel research designs and modes of data collection – previously used successfully by project members – to address these gaps in the literature. The project includes a unique and extensive data collection - including original survey data and register data – which allows us to reach migrants in different contextual settings with various integration levels. We capitalize on the significant local variation in both anti-immigrant party support and ethnic neighbourhoods, making use of the 2016 Swedish settlement reform as a natural experiment to test the causal effect of these factors.

Background

The research literature on immigrant integration trajectories focuses heavily on individual-level predictors such as age, gender and education (e.g. Bratsberg et al 2014). More recently, a growing research field have emphasized the importance of contextual factors that are specifically relevant for immigrants, including institutional conditions that can hinder or promote inclusion, such as legal access to citizenship and voting rights  (Hainmueller 2017), and the surrounding environment of the host society (Just and Andersson, 2014).

There are two contextual factors with potentially large, but yet largely unknown, effects on political integration: anti-immigrant expressions and ethnic neighbourhoods.

Starting with anti-immigrant expressions, this includes a variety of social and political expressions, such as attitudes towards immigration, xenophobic political communication, discrimination, and – which is the main focus in this project, electoral support for anti-immigrant parties. The research field on anti-immigrant expressions’ effects on political integration is a field still in its infancy, and, so far, displays conflicting results regarding the effect of anti-immigrant expressions on political integration. Prior work suggests that exposure to anti-immigrant expressions can either increase or decrease political participation, as well as naturalization (e.g. Pantoja et al 2001; Simonsen, 2020).

A major problem with the empirical literature on anti-immigrant expressions, as it stands now, is that despite the fact that previous research has shown that the period after arrival represents an “integration window” in which immigrants may be open to habit change, current research largely fails to capture the experiences of immigrants during the “integration window”. Larger surveys tend to mainly include immigrants who are already highly integrated into the society, which severely impedes our ability to understand the integration process (e.g. Feskens et al., 2006, but see Esaiasson and Sohlberg 2018 for exception). Qualitative studies, in turn, provide a deeper understanding of integration but does not enable generalizations to the broader population. Moreover, empirical studies rarely provide evidence of a causal relationship.

The diverging empirical results in the field are in line with the diverging theoretical assumptions; individuals are expected to react to anti-immigrant expressions either by withdrawing from the society or engaging, in order to raise the group status or secure material interests (Ellemers, 2002; Simonsen 2020).

We suggest that more attention to the different incentive structures framing different types of political integration is necessary in order to develop clearer theoretical propositions about how and when anti-immigrant expressions can be expected to engage or disengage. The empirical inconsistencies, on the other hand, demand novel research designs and innovative ways of collecting data, as will be described below.

Turning to ethnic neighbourhoods, their assumed effects on economic and political integration have featured prominently in the political debate, and more recently also in the academic debate. Some argue that ethnic neighbourhoods foster integration by providing networks that enable easier access to the labour market and strengthen immigrants’ engagement with the host society. Others, in contrast, argues that ethnic neighbourhoods hinders integration by isolating immigrants further from the host society.

There is a small but growing research field testing the effect on ethnic neighbourhoods on political integration. On political participation, some find that ethnic neighbourhoods hinder immigrants’ inclusion by isolating immigrants further from the host society (Cho et al., 2006), whereas others find that they can provide networks that strengthen immigrants’ political engagement (Bhatti and Hansen, 2016). More recent work suggests that the effect may be conditional, so that ethnic neighbourhoods influence voting among refugees only in those cases where they live among co-nationals who are politically active (Andersson et al, forthcoming). On citizenship, there are some cross-sectional studies showing opposite conditional effects:  that ethnic neighbourhoods correlates positively with naturalization rates in poorly assimilated neighbourhoods (Duncan and Waldorf, 2009).

Research on the effects of ethnic neighbourhoods on political integration suffers from the same weaknesses as the research on the effects of anti-immigrant expressions described above: a lack of attention to the incentive structures surrounding different types of integration; a lack of information about the experiences of immigrants during the “integration window”; and an inability to draw conclusions about causality. Causality is a particularly pressing problem for research on ethnic neighbourhoods, since many observed empirical patterns could be a result of immigrants’ self-selection into particular types of neighbourhoods, making inferences about the effects of that neighbourhood very difficult to sustain. Moreover, many existing studies use old data that do not reflect the immigrant population arriving in Europe today.

Significance and scientific novelty

In the short term, the project will be significant by testing the impact of a) anti-immigrant expressions (in particular electoral support for anti-immigrant parties), and b) ethnic neighbourhoods, on immigrants’ political integration in Sweden. While doing so, we address both theoretical and empirical inconsistencies in previous research:

• We test the impact of our independent variables on integration that takes place during the “integration window” – which is most relevant, but least researched
• We will collect survey data on a, so far, largely neglected group of individuals that constitute a large share of the immigrant population in Sweden today: newly arrived migrants. So far, opinion surveys – in Sweden and Europe at large – have mostly captured the experiences of already well-integrated migrants.
• We provide causal evidence on the influence of anti-immigrant expressions and ethnic neighbourhoods on political integration, while previous research mainly provide observational data
• We address, both theoretically and empirically, the various characteristics of different types of political integration In the longer term, we hope to influence the field also in other ways, by using innovative research designs and developing novel ways of collecting data that could be used also in other contexts.

Turning to the project’s significance for society, the results from our project will be highly relevant for policy makers when considering appropriate policy measures to increase integration. In this respect, the significance of the project can hardly be overstated: successful integration makes it possible to harness the opportunities related to migration, while failed integration is related to a number of societal problems. On the individual level, political integration is an important predictor of wellbeing (Huo et al., 2010). On the societal level, political integration is key in order to reach a cohesive political society. Theorists of liberal nationalism argue that national identities  function as a 'glue' that holds culturally diverse societies together (Miller and Ali, 2014). Moreover, immigrants make up a significant part of the population in many of the receiving countries, but have considerably lower turnout levels than natives (Statistics Sweden 2019). Such substantial inequality in participation may bias public policy against immigrant preferences (Vernby, 2013), indicating the importance of integration to ensure democratic legitimacy.

Research group

• Frida Boräng, Associate Professor, Department of Political Science, University of Gothenburg (Pricipal Investigator)
• Linna Martén, Researcher, Deptartment of Economics, Uppsala University
• Maria Tyrberg, Post Doctor, Department of Political Science, University of Gothenburg.