Integration Through Disintegration? The Distinction Between Deserving and Undeserving Refugees in National and Local Integration Policies in Germany

Whereas refugees with an insecure residence status have long been excluded from integration measures in Germany, they have recently become the target of integration policies at both the national and the local levels, especially in cities. This chapter compares these policies through category analysis. The core argument is that there is a difference between the logics underlying the policies at the two levels: the national Integration Bill is mainly marked by an ethno-national framing of integration which contributes – through the introduction of the notion of ‘likely or not to stay’ – to a further fractioning of the refugee label and thus the deterioration of rights for many asylum claimants. While it posits integration as a privilege and duty for ‘genuine’ refugees, it aims to undermine the integration of those not deemed to be deserving, following the logic that the disintegration of the latter is necessary to reserve integration capacities for the former. At the local level, in contrast, participation matters more than legal status and refugees are increasingly viewed as a potential resource for and part of a heterogeneous urban society. Yet, also at the local level, integration is ultimately tied to disintegration, as local authorities attempt to select who comes to the city in the first place.


Introduction
Until recently, refugees 1 with an insecure residence status were, just like undocumented migrants, not a target group for integration policies in Germany. What is more, they were explicitly excluded from national integration provisions, such as German language and integration courses. They were largely prohibited from taking up employment, from moving out of accommodation centres into private flats and from leaving their assigned locality of residence. These and other measures had been introduced by the German government in the 1980s to make the stay of (rejected) asylum-seekers in Germany as unattractive as possible and to deter other potential asylum-seeking persons. Even though municipal authorities generally adopted a much more pragmatic and inclusive approach knowing that many of the protection-seekers would end up staying for many years they often did not explicitly include them in integration plans and, in some cases, also explicitly excluded them (Aumüller 2009;Bommes 2012).
However, the perception of and take on the integration of refugees with an insecure residence status seems to have changed. In the 2000s, a few city municipalities started to make plans to decentralise the accommodation of refugees -that is, to move away from mass accommodation to providing access to private flats. Since 2013, it has also become increasingly common for local authorities to adopt encompassing strategies or concepts for the integration of refugees, including those without a secure residence status. Some of the Länder (federal states), too, have adopted measures to improve the living situation of asylum-seekers -e.g. by financially or otherwise supporting the decentralisation of accommodation and introducing quality standards for collective accommodation centres (Aumüller 2018). The national government has relaxed some of the measures of discomfort and deterrence, like the residency obligation (in 2015) and the employment prohibition (since 2014). With the adoption of the Integration Bill in July 2016, refugees without a secure residence permit even became the main focus of national integration policies.
This chapter enquires into the functioning and underlying logics of recent integration policies in Germany. In line with the literature on integration policies that notes clashes between different levels of policy-making (Chauvin and Garcés-Mascareñas 2012;Mügge and van der Haar 2016), it especially explores the links and tensions between integration measures at municipal and the national levels. The integration policies and measures of the Länder are beyond the scope of this paper. The analysis of the local level is based on a case study of one German city. The indepth analysis of one local case allows to go beyond an analysis of integration policies as they are described in official documents and to trace the implementation of policies over time (Penninx and Garcés-Mascareñas 2016). Looking at both levels, this chapter asks: How do the policies construct and distinguish between the different target groups? Who is (not) to be integrated? And in how far are the national framings of integration reproduced or contested at the local level?
In order to grasp the logics of integration policies at different levels, the chapter places special emphasis on their explicit and implicit category structures. The following section lays out some of the principles of category analysis and its relevance for understanding integration policies. The third section presents the methods employed in this study. The fourth section turns to the national Integration Bill, which will then be contrasted, fifth, with an analysis of local integration policies and practices in the city of Osnabrück.
While the lack of integration of immigrants is often the proposed problem, one can also consider disintegration measures as part of the policy repertoire of states to attempt to control immigration and (post-)migration social relations. Vicki Täubig (2009) has used the concept of organised disintegration to describe the living situation of refugees with an insecure residence status in Germany. Building on Erving Goffmann's (1973) concept of a total institution, Täubig (2009 pp. 45-54) describes how the German state undermines asylum-seekers' right to a self-determined life and integrated social relations by obliging them to reside in collective accommodation centres, where their daily lives are subjected to strict bureaucratic regimentation. Sieglinde Rosenberger (2012) ties in with Täubig's work by distinguishing between residential segregation on the one hand and material disintegration on the other, the latter referring to the erecting of barriers to asylum-seekers' access to resources and institutions. In line with the conceptualisation of (dis)integration as spelled out by Collyer et al. (2020), this chapter looks for connections between integration and disintegration. It seeks to contribute to the argument that disintegration is not only an aim of policy-makers but is also, in fact, legitimised within a broader integration framework (Ibid.).
Categories are, in many ways, at the heart of migration and integration policies, as they define 'who is a wanted and who is an unwanted migrant and who requires integration and who does not' (Mügge and van der Haar 2016, p. 77). The desirability or deservingness of immigrants is framed in different and sometimes contradictory ways, as Sébastien Chauvin and Blanca Garcés-Mascareñas (2014) have highlighted. For refugees, vulnerability has been a major criterion for deservingness but, as I argue below, other framings such as economic performance or cultural deservingness also (increasingly) play a role. We can see in the shift of categories and the underlying framings of deservingness over time, as well as in diverging practices at the national and local levels, that the categorisations on which integration policies rest are always due to change. Even though, especially in policies, they appear as if they were fixed and natural, they are always social constructions. Multiple actors at multiple levels engage in the construction of categories, including those who are categorised.
From a critical perspective, it is especially important to reflect how we, as researchers, (co-)produce categorisations. For example, the use of a presumably neutral legal category such as 'asylum-seeker' is, in fact, highly problematic if we consider that it was introduced by Northern governments to mark the distinction between asylum claimants and those who are granted asylum, thus preventing access to the label 'refugee' and entailing the 'wholesale withdrawal or reduction of established rights' (Zetter 2007, p. 181). In a similar vein, the use of 'refugee' as an analytical term has been questioned, both because it is intimately tied to a specific legal status and because it seems to confirm the political differentiation between forced and voluntary migration (Fiedler et al. 2017). I have nevertheless opted to use the term because the constitution of the (non-)refugee subject in and through integration policies is the very focus of this chapter. In what follows, categories are mainly regarded as a construction of states and policy-makers. State actors and the policy documents they produce are of special relevance due to their power position and because their formal systems of categorisation are particularly apt for deconstruction (Martiniello and Simon 2005, p. 8).
Category analysis can help 'to identify the architecture of the argument that underlies a policy issue and that, while often not discussed explicitly in policy debates, nevertheless is part of policy-relevant publics' sense-making' (Yanow 2000, p. 55). In order to reveal the organising principle of category systems, it helps to question their supposedly exclusive, exhaustive and neutral nature (Yanow 2000;Mügge and van der Haar 2016). Category systems rest on the assumption that categories are mutually exclusive, an assumption which, however, can be questioned from an analytical point of view -people may move between and fit into several categories. The supposed exhaustiveness of category systems can be deconstructed if we look beyond those groups or individuals highlighted by the policy and ask who is obscured or only targeted implicitly by a policy and why. Integration policies, for example, rarely or never mention undocumented migrants, just as they remain silent on those groups deemed automatically integrated. Intended or not, category systems often (re-)produce stereotypes, prejudice and inequality (Mügge and van der Haar 2016). While some elements or groups are presented as problematic or deviant, others are constructed as normal (Yanow 2000, p. 52). Category analysis explores this as the marking of categories (Ibid.).
In the literature on integration policies, differential organising principles or markings have been noted between policy levels. For example, according to the institutional discourse of the European Union, EU citizens are integrated in all EU member-states and should therefore not be considered as specific targets of integration policies. However, the issue is handled quite differently by policy-makers at the national level, as the discrimination against some EU citizens -especially the Roma and citizens of Eastern European countries -in other EU member-states shows (Lind and Persdotter 2017;Magazzini 2020). Differences have also been observed between national and local levels, especially towards immigrants without a secure residence permit (Chauvin and Garcés-Mascareñas 2012; Mügge and van der Haar 2016; Schweitzer 2020).

Methodology
Given the interest in categorisations, the analysis of national and local policy documents will focus on their diagnostic parts as well as the solutions -i.e. the actual measures -they propose. For the (national) Integration Bill, this means that I also take into account the Draft Bill, the executive order and other statements connected to the bill. The local case study consists of an analysis of policy documents, ethnographic data and interviews generated during my PhD project on local asylum practices in Osnabrück between 2014 and 2016. I chose Osnabrück, a city of about 160,000 inhabitants in the German state of Lower Saxony, as the entry point for my fieldwork because the city has a long history of accommodating refugees and thus an established 'asylum landscape' ) with a diversity of relevant actors and sites, yet not as many as in bigger cities. I conducted interviews with key actors in the municipal administration and government, as well as local NGOs and initiatives, in order to understand how asylum is negotiated in a particular local setting and how this changes over time. Given the focus of this chapter on comparing the national and local levels, a differentiated analysis of conflicts and contradictory logics within and among local institutions has to be omitted. In line with category analysis, as laid out above, I look at the construction, demarcation and markings of the (non-) targets of recent integration policies, as well as the understanding of integration and (urban) society that these policies convey.

The National Integration Bill
The Integration Bill of 2016 concerns, despite its general framing, only a specific group: refugees. In fact, most of its regulations address asylum applicants and rejected asylum-seekers with a Duldung (certificate suspending their deportation). This presents a rupture with earlier national integration provisions, 2 from which refugees without a secure residence permit were, more or less explicitly, excluded. The introduction of the Integration Bill has to be read as part of a series of legal changes that were introduced in 2015 and 2016 in reaction to the heightened number of persons seeking asylum in Germany. As the draft Bill clearly states, the augmented number of refugees and their supposed lack of integration are defined as a problem: Only last year, 476,649 persons have applied for asylum in Germany […] A lack of integration does not only lead to social problems in the medium and long term, it also leads to high costs (Gesetzentwurf für das Integrationsgesetz 2016, p. 1).
To counter the influx of asylum-seekers and their lack of integration, the bill introduces a legal division between putative 'genuine' and 'bogus' refugees through the notion of strong or weak 'likelihood of staying' (Bleibeperspektive). For the asylum applicants considered as being likely to stay, it establishes fast-track integration and, for those who are not, accelerated asylum procedures and deportations. The following sections look at the explicit and implicit categories constructed in and through the Integration Bill.

'We' and the 'Others': An Ethno-National Framing of Integration
In addressing only refugees and the problem of their (non-)integration, the German government marks them as 'others' -that is, as persons who do not (yet) belong to German society. The distinction between 'us' and an asylum-seeking 'other' rests on the imagination of German society as a homogenous ethno-cultural entity. This container model of society and culture comes to the fore in the Integration Bill in several ways: First, it posits refugees as a potential threat to social cohesion and stability. While it points to the humanitarian obligation of the German state to protect asylum-seekers, it also underlines the national government's duty 'to maintain a peaceful, liberal and communal society' (Gesetzentwurf für das Integrationsgesetz 2016, p. 23). The arrival of asylum-seekers is thus constructed as a menace to a supposedly cohesive and conflict-free German society. Second, it frames integration as an obligation on behalf of the asylum-seeker and not as a two-way or even a threeway process (Penninx and Garcés-Mascareñas 2016). That is, it mainly focuses on the duties of refugees and foresees sanctioning mechanisms in cases of noncompliance but barely targets discrimination and other barriers to their equal participation in the institutions of the receiving society. The only exception is the removal of barriers to the labour market and education schemes for refugees considered as likely to stay. Third, it places special emphasis on 'cultural integration', which is understood as the learning of German and the acceptance of 'German values'. As the executive order to the Integration Bill explicates, 'The content of the orientation courses [for asylum-seekers] will be extended and will focus primarily on the conveying of values' (Verordnung zum Integrationsgesetz 2016, p. 9). The integration courses can be made compulsory and, in cases of non-compliance, the living allowance cut. This approach to integration ties in with the idea of the German nation as a community of descent and culture as well as with debates about a German 'leading culture'. What exactly is to be understood as German culture and values is not spelled out by the government. It also remains unclear who is assumed to be part of German society and who is not. In Germany, as in other Western countries, Muslims in particular and other persons associated with non-Western countries have been defined as target groups of integration measures (Lanz 2016), whereas migrants from the 'global North' as well as so-called 'expatriates' and their families are usually exempt from integration requirements (Hess and Moser 2009, p. 18). While integration is considered unnecessary for most and an obligation for some, it is also constructed as an exclusive privilege. Not everyone is supposed to integrate. While the Integration Bill includes some asylum-seekers without a secure residence permit, it excludes others.

Differentiating Between 'Genuine' and 'Bogus' Refugees
To a certain extent the Integration Bill breaks with the differentiation between refugees with and without a secure residence status, as it grants some asylum applicants and persons with a Duldung easier access to the labour market and job training. Whether asylum applicants are included or excluded from integration provisions depends on their assumed 'likelihood of staying'. Asylum applicants assumed to be 'likely' to be granted a secure residence status can take part in language and integration courses before a decision in their asylum procedure is taken. This right was hitherto reserved for persons with a secure residence status. According to the Integration Bill, 'being likely to stay' (gute Bleibeperspektive) means that 'a regular and permanent stay is to be expected' (Integrationsgesetz 2016, Art. 1). The distinction between asylum applicants who are likely to stay and those who are not is based on their nationality. Only if a considerable number of persons of one nationality ask for asylum in Germany and if the unadjusted protection rate for persons of this nationality exceeds 50%, they are assumed as being likely to stay (Bundesamt für Migration und Flüchtlinge 2019). This label has been attributed to persons from Syria, Eritrea, Iraq, Iran and temporarily also to Somalian nationals. Every 6 months the attribution of this label is reconsidered (ibid., 2019).
Critics have pointed out that the distinction between asylum applicants on the basis of the notion 'likely or not to stay' is inconsistent because it does not include all those who have a high probability of being granted asylum in Germany, as it is based on the unadjusted protection rate, which encompasses all asylum decisions, including those rejected on formal grounds (Pro Asyl 2017; Voigt 2016). This way of arguing, while making the case for a much larger number of asylum-seekers, follows the same logic as the Integration Bill as it holds that the line between those to be integrated and those to be excluded can be drawn on the basis of their likelihood of staying. In a more substantiated way, it can be argued that the Integration Bill and the notion of being 'likely or not to stay' contradicts the very principle of asylum as a right that is accorded to individuals on the basis of a proper hearing.
Among the asylum applicants with a poor likelihood of staying are all those whose countries of origin have been white-listed. 3 Persons from these so-called 'safe countries of origin' are not only subjected to material disintegration but also to residential segregation. Unlike other asylum applicants, they are not transferred from the so-called 'reception facilities' run by the regional authorities to municipal accommodation but they have to stay until the end of their procedure (and thus their deportation) in these facilities. Some Länder, such as Bavaria, have even established special camps for persons considered unlikely to stay since 2015. 4 The distinction between asylum applicants who are or not likely to stay is related but not equal to the discourse on economic refugees. The former notion seems more technical and less stigmatising than the latter. Economic refugees are marked as less deserving than other refugees and assumed to be moving voluntarily and in order to attain better working and living conditions, which is seen as 'asylum abuse' (Bade 2015, p. 6). 'Genuine' refugees are thought to be moving due to war, political persecution or a well-founded fear of persecution. The label thus serves to devalue the motives of persons seeking asylum and the persons themselves (Flüchtlingsrat Niedersachsen 2017, p. 33). The notion of '(not) likely to stay' is more open to interpretation, as it does not entirely rule out the possibility of a person eventually and rightfully obtaining asylum. However, it strongly suggests that this will not happen and it is often used as a synonym for economic refugee. It also suggests that voluntary migration can be clearly distinguished from forced movement and comes with certain representations and ideas of what constitutes violence and who deserves protection and who does not. Both notions are therefore in opposition to the findings of empirical studies, which have shown that migration flows are usually mixed and that migration motives are complex and may change over time (e.g. King 2002, pp. 92-93;Zetter 2007, p. 175).
The a priori distinction among asylum applicants and their differential treatment through the notion of (not) likely to stay is legitimated by the argument that the country has a limited capacity for reception and integration. The government has claimed that 'We have to concentrate our efforts on those people who flee from war and political persecution and really need protection' (Bundesregierung 2016a). In other words, the disintegration of some is legitimated by the need to reserve integration capacity for others. This logic is in line with the model of society as a container, thus constituting a finite space with limited integration capacity (Nimführ et al. this volume). Along with the humanitarian legitimation of the preferential treatment of some refugees on the basis of their nationality, a relatively new logic can be noted that ranks asylum-seekers according to their assumed usefulness (for the German labour market) and thus links protection to economic performance.

'Promoting and Demanding' the Integration of Entrepreneurial Subjects
The expansion of the integration dispositive to include asylum applicants considered 'likely to stay' and those with a Duldung has to be read above all as the triumphant success of a workfare approach to integration (Lanz 2009). In line with the workfare principle, the Integration Bill asks asylum-seekers to actively look for a job and become independent of social benefits. To enable their fast integration into the German labour market, barriers such as the interdiction to work and the proof of precedence 5 have been relaxed. Moreover, asylum applicants and those with a Duldung can -after a certified period of stay in the country -claim educational and vocational grants. Besides the easing of access to the labour market, the integration law includes a number of activating measures. In addition to the mandatory language and integration courses, asylum-seekers may be obliged to take part in socalled 'refugee integration measures' -low-paid jobs (with a remuneration of 80 cents an hour) which are supposed to serve as a 'meaningful occupation' during the asylum procedure and to make asylum-seekers fit for the German labour market (Bundesregierung 2016b).
In the workfare state, social benefits are only attributed under certain conditions. If a welfare recipient fails to (re-)enter the labour market or contribute to society by engaging in some form of work scheme, social benefits may be cut or other sanctioning mechanisms put into place. The slogan of 'promoting and demanding' labour market integration has been at the heart of such approaches and also takes a central place in the Integration Bill. As explained by the German government: 'They [asylum-seekers] are [however] obliged to also make an effort to integrate. If asylum-seekers refuse integration measures or the obligation to cooperate, benefits will be cut' (Bundesregierung 2016b). Not only social benefits but also residence permits are tied to participation in integration measures. Those who do not fulfil the requirements may have their settlement permit withheld, while those who can prove advanced German skills and secure their own livelihood may be rewarded with a permanent 'settlement' permit after 3 years. These regulations imply that (non-) participation in integration measures has to be controlled, which can be challenging for local institutions and individuals responsible for controlling and informing the national ministry (Schweitzer 2020).
While workfare regulations are not new -they have marked the treatment of unemployed persons in Germany since the 1980s (Lanz 2009, p. 111) -the extension of this logic to the area of asylum is relatively recent. By promoting and demanding asylum-seekers as entrepreneurial subjects, the government seeks to kill two birds with one stone. One aim is to prevent a lack of integration and long-term dependence on social benefits. The other is to have refugees contribute to meeting the challenge of demographic change and skills shortages in certain sectors (Gesetzentwurf 2016). As several authors have pointed out, the 'refugee problem' has partly been made up not to confront the challenge of how to deal with international migration in a globalised world (Zetter 2007). As Castles (2003) has put it, Northern governments 'tacitly use asylum and undocumented migration as a way of meeting labour needs without publicly admitting the need for unskilled migration' (p. 16).
In short, the national Integration Bill partly breaks with the distinction between persons recognised as refugees and asylum applicants, in the sense that (some) asylum applicants and persons with a Duldung now also have access to integration courses and the labour market. Yet the binary approach to deservingness is not in fact challenged and becomes even more strongly enmeshed within a frame of performance and utility. In line with a workfare approach to integration, some refugees may be obliged to take part in integration schemes. If they refuse, they can be punished with a reduction of their social benefits and the prolongation of their legal insecurity. While, for this group, integration becomes an obligation, other refugees -namely those considered as 'not likely to stay' -are deliberately disintegrated, in terms of both residential segregation and material disintegration.

A Different Narrative? Local Integration Policies and Practices
Having explored the way in which disintegration is entangled with integration policies at the national level, this section now turns to (dis)integration policies and practices at the local level. 'Integration takes place locally' has been a much-repeated phrase in both political and academic debates on integration (Bommes 2012). While Germany was not perceived as an 'immigration country' by the national government until the 2000s, many municipalities, especially cities, 6 have been developing integration plans for their immigrant populations at least since the 1980s (Gesemann and Roth 2009). While these integration concepts rarely mentioned persons with an insecure residence status, municipalities often provided some services to all residents no matter their legal status (Aumüller 2009). The first integration documents explicitly targeting refugees (with various legal statuses) were developed in the late 1990s and early 2000s and mostly focused on the area of housing. City authorities found that decentralised accommodation was better not only for those directly concerned but also for urban society as a whole as well as the public budget (Wendel 2014, p. 79). With their decentralisation plans, some city municipalities contradicted the national and Länder regulations, which stipulated (and in some cases still do) that refugees with an insecure resident status should be housed in accommodation centres. Fully fledged integration concepts explicitly targeting refugees were adopted by many German cities in 2015 and 2016 as a reaction to the heightened number of refugees arriving in the country. In Osnabrück, such policies were first developed in 2013, thus preceding both the discourse of a 'refugee crisis' in 2015 and the official re-framing of integration by the 2016 Integration Bill. The 2013 'Plan for the Integration and Accommodation of Refugees in the City of Osnabrück' was a reaction to increasing allocations of refugees and the desire to organise their accommodation in a way that corresponded with their needs, following the example of other city municipalities (Stadt Osnabrück 2013). It officially established decentralised housing and introduced 'proactive' social work. The Osnabrück Integration Plan was the first of its kind in Lower Saxony and was soon copied by other municipalities. However, it was quickly outmoded by the dynamics of the long summer of migration (Kasparek and Speer 2015): Between 2013 and 2017, more than 4000 refugees were allocated to the city (Stadt Osnabrück 2017) and both a regional 'initial reception centre' and several new municipal accommodation centres were opened, some of which were closed again in 2017 when fewer refugees were allocated to the city due to the renewed success of European and German authorities in keeping refugees out. In 2018, the municipality published an updated Integration Plan for refugees. As I show below, the changing local integration policies differ (increasingly) from the national policies in terms of how they understand integration and construct refugee subjects and their deservingness.

Refugees as Part of a Heterogeneous Urban Society
With the 2013 Plan, the Osnabrück municipality stated officially, for the first time, that the integration of refugees with insecure legal status was a policy aim and that the long-established differentiation between persons with a 'migration background' -hitherto the main target group of integration measures -and 'refugees' or 'asylum-seekers' who had been officially excluded from such measures, was to be discontinued 'as far as possible' (Stadt Osnabrück 2013, p. 3). As the Municipal Commissioner for Integration in Osnabrück underlined in an interview on 6 July 2015: That we decided to focus on this [the integration of refugees] and that we even used the word 'integration' was completely new, because the Asylum-Seekers Benefits Act actually forbids this; integration should not take place [for persons with an insecure residence status].
The 2013 Plan identified refugees as persons who have suffered and are in need of special guidance and care (Stadt Osnabrück 2013, p. 3). This served, on the one hand, to justify the expenditure on additional social workers; on the other hand, it legitimised the continued accommodation of refugees with an insecure residence status in centres, where social workers could more easily intervene. The city authorities established decentralised housing only for those refugees who had stayed for two or more years in the city, with the exception of persons -like families or the elderly -for whom a prolonged stay in an accommodation centre was considered inappropriate. The decision to link access to private housing to the length of stay in the city or the migrants' supposed vulnerability was harshly criticised by some civil society initiatives and the City's Migrant Advisory Board, which had been involved in developing the Integration Plan and demanded access to decentralised housing for all refugees (Migrationsbeirat der Stadt Osnabrück 2013;No Lager 2013).
Throughout the long summer of migration, the city authorities often reverted to the image of refugees as victims in order to mobilise support and sympathy for the newcomers in the local community. At the same time, refugees were increasingly framed as an integral part of urban society. In 2015, the city administration decided that all refugees, regardless of their legal status and length of stay in the city, could move into private flats. In the same year, the Municipal Integration Department presented an integrated activity report, which no longer differentiated between activities targeting refugees and those aimed at other migrants. The report simply highlighted that the urban community was growing and becoming more diverse: Migration is becoming more and more important in the city -today every fourth person has already a so-called migration background. Among children (younger than six) it is already 48 per cent (Stadt Osnabrück 2016, p. 4). By underlining that the local population consists to a large extent of migrants and their children, local authorities portray migration as the norm and migrants as members of the local community. This also counts for refugees, who are addressed as '(new) Osnabrückers' in the latest Integration Plan . This perception of migrants as rightful residents or 'citizens' is also reflected in the local authorities' concern to facilitate the transition of migrants with different legal statuses as soon as possible from immigration reception and orientation services to regular municipal services .
In addition, and in line with a more general shift from a problem-oriented to a potentiality-focused perspective on migration-based diversity among German city municipalities (Pütz and Rodatz 2013), (refugee) migration to Osnabrück is increasingly referred to as a potential advantage. The authorities have underlined time and again that Osnabrück is growing and flourishing and that this is at least partly thanks to (refugee) immigration. A recent municipal demographic forecast for 2017-2030 states: Compared to the composition of the population in 2016, the population will become older and more diverse as a result of international immigration and the integration of refugees. This development fits in with the picture of the expected population in other German cities, which attracts (young) people with a good infrastructure and qualified education and job offers. A particular challenge is to bind this population group in the long-term, especially once they have completed their education and during the family phase (Stadt Osnabrück 2017, p. 55).
While the positive discourse on migration-based diversity is challenged by some (within both local institutions and the wider urban community), it still dominates local political debates and practices, as an incident in 2014 illustrates. As part of a publicity campaign to promote diversity in the city, the municipality had displayed a large poster which read 'Diversity is our strength' and showed images of several Osnabrück residents with a 'migration background'. When the poster was deliberately destroyed in 2014, the local authorities reinforced their statement by selling the remaining paper shreds to locals and replacing the poster. While certainly also driven by a desire to make a stance against xenophobic and racist violence, the positive take on migration and diversity by local governments can at least partly be explained by inter-locality competition. In their quest for financial and human resources, city governments seek to position themselves as entrepreneurial cities using migration-based diversity and the way in which they are managing it as a location factor (Desille 2020;Pütz and Rodatz 2013;Schmiz 2017). In Osnabrück, for example, refugee accommodation and integration in the city have been used to reinforce the city's image as the 'City of Peace' (a reference to the signing of the Treaty of Westphalia in 1648): The City of Osnabrück already paved the way, in 2013, […] for a culture of welcome so that we did not experience the arrival of refugees in 2014/15 as a 'refugee crisis'. Since then, more than 4,000 persons, who fled from war, persecution and paucity of prospects, have found a safe haven in the City of Peace of Osnabrück (Stadt Osnabrück 2018, p. 6).
The above-mentioned 'paradigm shift' to a potentiality-oriented perspective entails the risk of reducing migrants to 'human resources' and reformulating deservingness on the basis of economic concerns in a similar way to the workfare approach to integration at the national level. However, it also has the potential to break with ethno-centric models of citizenship, particularly if participation and integration are not only framed as a duty on behalf of the newcomers.

We All Need to Integrate: Integration as a Two-Way Process
While the Integration Bill frames integration above all as a duty on behalf of the individual refugee, the local integration policies analysed here frame integration as a two-way process which concerns the whole of urban society and its institutions. In quite explicit opposition to the national discourse on 'integration', the local Integration Department states: Any demands for 'integration' are based on the erroneous assumption that integration processes are shaped above all by the immigrants themselves, since they have to integrate into the 'host society'. However, if we understand integrationas a task for society as a whole, including the opening up of established social institutions, it becomes clear that the concept of integration must be based on a broader conceptual foundation (Stadt Osnabrück 2018, pp. 9-10).
For the municipal authorities, the problem is not a supposed unwillingness or incapacity to integrate on behalf of certain individuals but an inequality of access to and participation in social systems. The aim of local integration policies is, accordingly, to establish equality of access to and participation in different spheres of social life (like health, housing, employment, law, politics, religion and so on). One of the main barriers to equal participation in these spheres, from the perspective of the municipality, is the legal insecurity and differential access tied to the different legal statuses of their residents. Discrimination on the basis of legal status is not only an additional administrative burden and cost but also a factor that contributes to inequality and insecurity in the urban community, as a representative of the Integration Service in Osnabrück explained in an interview: […] the municipality takes on the expenses, for example, for health and accommodation, and that is really insane compared to the normal system, isn't it? Why are they [refugees] not covered by statutory health insurance? Why do they not have access to integration courses? These are all federal funds. So this is the pecking order; in the pecking order we are at the very bottom as a municipality, we have to take responsibility for public order regulations like Dublin […].
There is thus a clash of interests and logics between national and local government insofar as the latter 'are obliged to make available to their inhabitants and thus also to foreigners -that is, migrants -the required economic, social and cultural institutions and services' (Bommes 2012, p. 128), whereas the former seek to protect their borders and social systems from non-citizens. Another reason for the different approach of local authorities to migrant integration is the strong engagement of civil society initiatives, and the fact that local politicians often feel more pressure to react to demands of local initiatives and migrant organisations than politicians at other levels of government (Cantat 2020;Ellermann 2009). In the US and Canada, opposed interests between city and national authorities concerning undocumented residents led to the development of 'sanctuary practices' (Bauder 2017) or even forms of 'local citizenship' (Varsanyi 2006) and also in many German cities, including Osnabrück, so-called 'solidarity-city' initiatives have formed. While the German initiatives, unlike the American and Canadian Sanctuary-City movement, are above all bottom-up initiatives, they seek dialogue with local authorities and often obtain political approval: The Osnabrück City Council, for example, endorsed the local initiative to resettle 50 refugees from camps in Greece to Osnabrück and, more recently, followed the demands of several civil society organisations to become part of an alliance of cities across Europe which declare themselves as 'safe havens' in order to protest against the (supra-)national politics of deterrence (Dörn 2018). In turn, the local government also relies on civil society initiatives to support newcomers. In Osnabrück, the Municipal Service for Refugee Integration is, in fact, a collaboration between the municipality and several NGOs. The municipality also relies on informal initiatives, particularly where integration support -like language courses for refugees who are considered 'unlikely to be able to stay' -is not funded by the regional or national government.
Yet the cooperation between local government and the administration with local activists also has its limits, as became apparent in Osnabrück in 2017, when activists asked the local council to take a stance against deportations. A broad alliance between different groups had prevented more than 36 Dublin deportations in the city in 2014/15, before changes regarding the deportation procedure by the Land rendered their prevention more difficult (Hinger et al. 2018). While part of the City Council supported the demand to 'avoid deportations whenever possible', the conservative parties emphasised their political backing of European and national legislation stipulating that rejected asylum-seekers should be deported (Kröger 2017). This shows the contested nature of the local asylum regime and highlights the fact that local governments also (re-)produce certain framings of belonging and deservingness based on legal status.

The Distinction Between 'Our' and Other Refugees
Although local integration measures thus tend to follow a different logic than national legislation, (city) municipalities sometimes also distinguish between those who (ought to) belong to the urban community and those who do not. Precisely because the discrimination of residents on the basis of their legal status is highly contested at the local level, municipalities demand selection of refugees before they become 'their' residents. As Bommes (2012, p. 128) noted, '[f]or municipalities, unless they are in a position to reject migrants, there has been no alternative but integration'. In other words, municipalities, like national governments, attempt to attract some immigrants and to reject others. In fact, German municipalities cannot reject refugees allocated to them by the Land government but can, at least to some degree, influence dispersal and accommodation policies and processes. First, municipal administrations can (try to) negotiate who is allocated to them in dialogue with the responsible service at the level of the Land. An employee of the Osnabrück Social Services, responsible for the reception and accommodation of refugees in the city, explained to me that he could express his preferences regarding the nationality of newcomers -namely, Syrians -and that, most of the time, persons were allocated in line with this preference, because it was 'in everyone's interest that there were no frictions in the allocation process'.
Second, municipalities negotiate with the Länder and the national government about the distribution of responsibilities and the cost of refugee accommodation. Even though individual municipalities or services might distance themselves from national asylum policies, as we have seen in the case of Osnabrück, city municipalities have, in fact, played a vital role in shaping migration and integration policies at the national level, including the Integration Bill. When the number of refugees arriving in German municipalities rose significantly in 2015/2016, it was also city representatives who pressed for accelerated asylum procedures and an early selection. The President of the Association of German Cities, for example, demanded that only 'genuine' refugees should be allocated to municipalities: Besides international efforts [to curb the number of asylum-seekers] the government and the Länder have to quickly implement the accelerated asylum procedures and then consistently return persons with no likelihood of being allowed to stay to their countries of origin. This is necessary, so that only those refugees who need our protection as civil war refugees and the politically persecuted are transferred into the municipalities (Lohse, cited in Deutscher Städtetag 2015).
This quote illustrates how city representatives contribute to the fractioning of the refugee label through the use of the notion of (not) being likely to stay and how this notion is equalised to the binaries genuine/bogus or civil war/economic refugee. This way of categorising refugees provides a deceptively simple moral compass and way of handling an intricate problem. Instead of waiting for the outcome of long and complex asylum procedures, the authorities simply assign refugees to one or the other category on the basis of their nationality (or, more precisely, the likelihood that persons of that nationality will be granted asylum). For the municipalities, such a preliminary selection is attractive, because the accommodation of persons who later have to be deported comes at a high cost -on the one hand, because the reception and orientation phase presents a financial burden, which is at least partly carried by the municipality, and on the other, because the enactment of deportations is a contentious issue. In Osnabrück, for example, the (planned) deportations of refugees to other EU member-states according to the Dublin regulation led to a series of protests and the actual prevention of deportations by a civil society initiative, as already mentioned above. While this engagement of the local community was welcomed by (part of) the local government, it also presented a problem, given that the local administration is expected to collaborate in the enactment of deportations. In order to forego such conflicts, the Mayor of Osnabrück asked the Land authorities in 2014 to discontinue the transfer of 'Dublin cases' to the municipality and instead keep them in the regional 'reception centre'. What is more, the mayor argued that such a measure of residential segregation was necessary in order to ensure successful integration at the local level: For meaningful and successful social work with refugees as well as for the engagement of volunteers, it is extremely difficult if, shortly after becoming acquainted with a refugee, they learn that this person is transferred to another [EU] member-state for their asylum procedure. Such very short encounters are a burden for everyone and have, as you probably know, already led to public outrage (Letter to the Minister of Interior by the Mayor of Osnabrück 2014).
The mayor's letter did not lead to a change in the allocation of refugees but shows how local authorities attempt to limit and select who comes to the city and how they, like their national counterparts, link the successful integration of some to the disintegration of others.
Finally, I want to highlight that the distinction between those who belong to the urban community and those who do not is not simply a question of presence, as suggested by Bommes (2012). The mere presence of refugees in the city does not actually suffice for them to be considered rightful members of the urban community. The example of refugees living in reception centres, run by the Länder authorities but located in cities, proves this point. Those accommodated in such centres are residents of the city but their stay is considered to be only temporary -since they are still awaiting their allocation to a municipality 7 -and their integration thus not expected. My interviewees in the local administration in Osnabrück, for example, did not consider the refugees accommodated in the reception centre that was opened in Osnabrück at the end of 2014 as 'their' refugees. Interestingly, they are not only excluded from municipal integration measures, but also civil society initiatives perceived those accommodated in the reception centre as somehow not belonging to the city, as a Solidarity-City activist in Osnabrück reflected: We only started to think about the [reception centre] sometime in the beginning of this year and that was partly because of Brenda, a Roma woman who appeared at one of our meetings one day. She told us a bit about what happened inside the camp. And this also spread among the No Lager group so that they also said we have to focus more on [the reception centre]. The whole time deportations were taking place there. Also when we celebrated our deportation preventions [elsewhere], the whole time people were being deported from the reception camp. So eventually we said we have to see the whole truth and this also includes [the reception centre].
The example of Brenda shows that 'acts of integration' (Collyer et al. 2020) are possible even for persons whose integration in (urban) society is put on hold. However, their (supposedly) only temporary stay, in addition to their accommodation in camps which are often located on the outskirts of cities, renders their participation in the life and institutions of the urban community very difficult.

Conclusions
This chapter has explored the functioning and logics of recent integration policies targeting refugees in Germany. While at both local and national levels the (dis)integration of refugees has become a policy issue, the perceived problem and proposed solutions differ. At the national level, an official integration policy targeting refugees was developed in response to their rising numbers in 2015/2016 -perceived and presented mostly as a threat to social cohesion and stability. The 2016 Integration Bill introduced a two-class asylum system with possibilities of fast-track integration for asylum claimants deemed 'likely to stay', and systematic disintegration -involving both residential segregation and material exclusion -for those considered 'unlikely to stay'. This categorisation, which is based on protection rates for different nationalities, conveys the message that decisions in asylum procedures are infallible and that a clear distinction between deserving and undeserving asylum claimants can be made. While economic performance also plays an increasing role in the determination of deservingness of refugees at the national level, the integration bill remains grounded in an ethno-national understanding of integration.
On the contrary, municipalities -especially city municipalities -developed integration measures for refugees with different legal statuses well before 2015, with the aim of reducing both financial and social costs. The distinction of residents on the basis of their legal status and the disintegration of some of them turned out to be impractical and against the very interests of municipalities in maintaining stable and flourishing local communities. The study of integration policies and practices in Osnabrück showed that migrants in general and refugees in particular, are still distinguished from other residents, however, not to mark them as non-members but to detect social inequalities. To overcome these and ensure equal participation in social systems is the main aim of local integration measures. Contrary to the national logic of integration as a duty and privilege, integration is here understood as a process also involving the opening of local institutions. Moreover, the integration of immigrants, including refugees, is seen not only as a challenge and burden but also as a chance and potential in the inter-locality competition for talent and resources.
At the same time, however, local governments also pursue integration through disintegration as they seek to select who comes to the city in the first place and to concentrate their efforts on those newcomers who are 'likely to stay'. While this notion is not used in the Osnabrück Integration Plans, the question whether or not a refugee will stay or be transferred/deported in a near future does play a role for local decisionmakers as illustrated by the letter of the Mayor cited above. While municipalities thus break with certain logics and framings of deservingness, they (re-) produce others, partly out of the (perceived) necessity to create cost and planning security and the desire to avoid local conflict. As my analysis shows, the way in which local authorities use their room for manoeuvre ultimately also depends on the local actor constellations and dynamics, e.g. the influence of civil society initiatives. In my case study, the latter mostly lobbied for a more generous implementation of asylum provisions. It needs to be stressed, that this is not always the case -in many German localities, civil society has also mobilised against the reception of asylumseeking newcomers. The negotiations between local state and non-state actors here served as an example to illustrate that negotiations take part on the local level. While not highlighted in this chapter, this observation also regards negotiations and conflicts among local government institutions. For example, the municipal Integration Department follows quite different institutional logics than the local Foreigners' or the Welfare Office. More research is needed to further explore integration policies and practices at the local level, highlighting the competing logics and forces within and among municipal administrative and political bodies, as well as comparative analyses, also including rural municipalities.
This chapter sought to contribute to the debate on (dis)integration, by pointing out that disintegration is not only an aim of policy-makers but is also legitimised within a broader integration framework. Above all on the national level, but to some extent also on the local level, policymakers (as well as parts of civil society) stress that integration capacities are limited and that successful integration requires a selection among refugees, and a systematic disintegration of those considered undeserving.
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.