In a press release outlining its “vision for the area of freedom, security and justice” the European Commission (2009) proposed that Member States should “[e]nsure a flexible immigration policy that is in line with the needs of the job market whilst at the same time support the integration of immigrants and tackle illegal immigration”. At least two questionable assumptions underlie this vision: that a neat distinction exists between those individuals whose integration should be supported and those whose immigration must be ‘tackled’; and that both goals can be achieved without respective measures interfering with each other or the demands of an increasingly transnational economy. What the proposal conceals are the many conflicts and contradictions between the very different interests, norms and logics that underlie these as well as other important functions of the state, which also include the provision of welfare services to its population. The various policies and practices aimed at regulating the movement of people across state borders are often subsumed under the term ‘migration management’, which provides “a convenient umbrella under which very different activities can be regrouped and given an apparent coherence, thus also facilitating cooperation between actors who would otherwise have little in common” (Geiger & Pécoud, 2010, p. 3). While many different actors and various levels of government are effectively involved in the management of international migration (Hepburn & Zapata-Barrero, 2014), the overall responsibility and formal decision-making power lies with the (migrant receiving) nation-state, which in order for migration to occur, “must be willing to accept immigration and to grant rights to outsiders” (Hollifield, 2004, p. 885). This does not mean, however, that by not accepting it, governments can prevent immigration from happening; but they can render it ‘irregular’.

The making of the distinction between regular and irregular migration is an important articulation of the nation-state’s sovereign power over its territory and population. It is the basis upon which states control and limit access to their territory and can deport unwanted foreigners from within it. At the same time, state governments can also effectively unmake this distinction by establishing paths to regularisation. In practice, however, no liberal state has ever been able to prevent all illegal entry, nor to deport or regularise all those unlawfully present. In fact, it is precisely their inability to avoid irregular migration that requires liberal states to deal with this phenomenon and thus to respond to its complexity. Part of this response is what I call the micro-management of irregular migration. That is, the formal or informal consolidation of a governmental logic that officially demands the exclusion of a person from the national territory where s/he is an irregular immigrant, with the various subjacent pressures for the same person’s inclusion as a local resident, worker, patient, student, and so on. Much of this micro-management takes place within the institutions of the welfare state, which often struggle to meet (or otherwise deal with) irregular residents’ fundamental needs and most legitimate claims. It is this rather indirect and obscure aspect of migration management that I am concerned with throughout this book.

By aiming or claiming to ‘effectively manage migration’, politicians arguably evoke the idea that immigration is (or at least can be brought) under their control. It is important to note, however, that the original meaning of the verb ‘to manage’ was not to be fully in control of something or someone but instead ‘to handle or train a horse’;Footnote 1 and that precisely because the untrained horse cannot be fully controlled, its handling used to take place in the manège – the etymological precursor of the term. As a place initially created to maximise the safety of both the horse and its trainer, the manège later also became a site of spectacle where riders display their horsemanship as well as the discipline of their horses and where circuses exhibit their spectacular or exotic performances for the entertainment of their audiences. This can be related to certain practices of migration management, which are also – even if that is not always made explicit or presented in public – the response to what effectively is a lack of control. And they too, involve the handling and sanctioning of certain irregularities and (miss-)behaviours. The more or less visible display of physical violence is thereby used as evidence of the government seemingly ‘being in control’. Nicolas De Genova (2013) referred to this aspect of immigration control as the ‘spectacles of migrant ‘illegality’, comprising not only the ‘scene of exclusion’ but also ‘the obscene of inclusion’ of irregular migrants within the legal, social and economic structures of the societies in which they live.

At the same time, the very terminology also invites consideration of the theoretical and empirical parallels between the management of migration and that of private companies as well as public services. Both have experienced a ‘managerial revolution’ of their own, which in the case of the former has been described as the replacement of the ‘invisible hand’ of market forces by the ‘visible hand of management’ (Chandler, 1977).Footnote 2 More recently, also the public service sector in many Western European countries has seen the establishment of various administrative management positions interposing direct government oversight and control within bureaucratic structures, and thus reflecting an increasing market orientation and vision of the citizen as a consumer of public services (Walsh, 1994; Webb, 2006; Hasenfeld & Garrow, 2012). In both cases, the managers – responsible for controlling or administering a particular set of resources, processes or practices – assume an intermediary role within a certain relationship of power: they manage and are themselves managed at the same time.

Scholars critically engaged with managerial practices within businesses and other organisations (see McKinlay & Starkey, 1998), various fields of social policy (McKee, 2009) or contemporary immigration regimes (Inda, 2006; Walters, 2015) have therefore often drawn on the ideas and concepts of Michel Foucault. What makes his work so useful as the basis for such analyses is his refined understanding of the exercise of power not as direct and absolute domination but in the form of ‘governmentality’, by which he means “a conduct of conducts” as well as “a management of possibilities” and thus a way “to structure the possible field of action of others” (Foucault, 2002a, p. 341). Such power relations are characterised by significant degrees of “informed consent, autonomy, voluntary action, choice, and nondirectiveness”, rather than complete and unidirectional rule and authority (Mezzadra & Neilson, 2013, p. 174). Public service provision is one of many spheres where the state’s sovereign power to neatly define, control, punish and exclude irregularities loses at least some of its grip, and must be complemented by more ‘governmental’ logics and modalities of power.

This chapter presents the theoretical and analytical framework through which I have studied the micro-management of irregular migration. On one hand, I thereby draw on the longstanding body of literature concerned with conceptualising the role of ‘the state’ in migration policies and policymaking, as well as more recent academic work on migrant irregularity as the product but also mirror of these policies. On the other hand, I look at some of the theoretical and empirical work done in the field of organisation studies, which helps to explain how organisations themselves deal with multiple and often contradictory norms and institutionalised logics originating both inside and outside of a particular organisational or professional field. The two strands of literature are linked via a Foucauldian understanding of governmental power and the conceptualisation of migrant irregularity as a ‘code’ through which the logic of immigration control is inscribed into existing power relations within and between different organisations. This theoretical approach helps to overcome the often too simplistic understanding of ‘the state’ that characterises much of the migration studies literature (Gill, 2010) by disaggregating the agency involved in the ‘management of migration’, and accounting for the multiple interests, rationales and constraints that underlie the involvement of very different actors at various administrative levels.

2.1 The State as the ‘Manager’ of Migration?

I don’t want to say that the sate isn’t important; what I want to say is that relations of power, and hence the analysis that must be made of them, necessarily extend beyond the limits of the state – in two senses. First of all, because the state, for all the omnipotence of its apparatuses, is far from being able to occupy the whole field of actual power relations; and, further, because the state can only operate on the basis of other, already existing power relations (Foucault, 2002b, p. 122).

In principle, democratic governance means that people’s ideas and opinions are translated into formal legal frameworks and laws, which then – mediated through local implementation processes – determine actual policy outcomes (Deutsch, 1970). In this way the rule of law guides the actions of individuals as well as public and private institutions. Particularly with regard to policy-making in the field of immigration, which has become a highly politicised and much researched topic, academic debate has long circled around the question of why these regulatory processes often fail to achieve the desired outcomes or declared objectives (Soysal, 1994; Freeman, 1995; Sassen, 1996; Joppke, 1998; Sciortino, 2000; Castles, 2004; Lahav & Guiraudon, 2006). More specifically, scholars identified a ‘gap’ between the official aims of immigration policies – which increasingly reflect the rising public pressure to restrict further unwanted immigration – and their often more liberal outcome regarding not only the admission of foreigners to the country but also their access to various social and economic rights (Hollifield, 1986; Cornelius et al., 1994). Where such rights are extended to people who have not been formally admitted, the underlying conflicts are particularly pronounced.

While the claim that national governments are generally ‘losing control’ over unwanted immigration remains contested (cf. Brubaker, 1994) the identified ‘gap’ has been related to a wide range of potential causes located both within and outside the realm of receiving states. One set of explanations points to the expansion of human rights and rights-based conceptions of membership that increasingly cut across national borders and citizenship, and thereby contribute to an alleged decline of the nation-state: Yasemin Soysal’s (1994) much-disputed vision of a post-national model of citizenship – based on ‘universal personhood’ rather than national belonging – derives its legitimacy from a ‘transnational discourse of human rights’ that entails certain obligations for states towards not only their own nationals but also aliens who legally reside within their borders, such as guest workers or students. For David Jacobson (1996, p. 2) this extension of ‘rights across borders’ has also significantly altered the legal position of migrants living ‘illegally’ within the borders of liberal states, which increasingly have to accept and respond to at least some of their claims. While human rights themselves “evolve from the nation-state” (p. 3), it is through them, he argues, that “[t]he state is becoming less a sovereign agent and more an institutional forum of a larger international and constitutional order based on human rights” (p. 2/3). Others have related the state’s limited capacity to control immigration to the complex and powerful macro-dynamics driving migration processes, including transnational networks of information, people and communication, and the highly unequal distribution of wealth and opportunities (Sassen, 1996; Castles, 2004). Looking at the micro-level, scholars have also highlighted migrants’ own networks, counterstrategies and agency in more or less effectively avoiding and contesting state control (e.g. Broeders & Engbersen, 2007; Vasta, 2011), as well as the various formal or informal support structures, including non-governmental organisations (NGOs) and advocacy groups, operating within and across countries of origin and destination (Faist, 2014; Ambrosini, 2017).

On the other hand, the ‘gap’ between official policy goals and outcomes has also been related to domestic political forces in the form of either organised interests (Freeman, 1995), governments’ own ‘hidden agendas’ (Castles, 2004), or ‘self-imposed’ constraints enshrined in national constitutions (Joppke, 1998; Guiraudon & Lahav, 2000). Rather than the international human rights regime imposing limits on the ability of states to reduce immigration, Christian Joppke (1998) argues that liberal states themselves accept unwanted immigration, and thus ‘self-limit’ their own sovereignty. His account specifically refers to legal mechanisms for family reunification and the admission of refugees, both of which are unwanted in the sense that they often do not fit the needs of national labour markets. Gary Freeman (1995, 2006), in turn, tried to explain the ‘expansionary bias’ of policy regimes governing the entry and stay of (regular and irregular) migrant workers. According to his model of ‘client politics’, the making of such policies tends to be driven by powerful interest groups who benefit from large-scale immigration (as a source of cheap and flexible labour) and whose interests prevail over those of a more restrictionist but poorly organised public that bears its rather diffuse costs (in the form of depressed wage levels and increased competition for jobs and resources).

Importantly, it is not just the making of immigration policy that is underpinned by different and often conflicting interests but so is its local-level implementation. The latter hinges on the capacity and willingness of a growing number and variety of actors across society to enforce exclusionary practices towards certain immigrants (Guiraudon & Lahav, 2000; Jordan et al., 2003). Based on a detailed mapping of the various ‘actors and venues in immigration control’, Gallya Lahav and Virginie Guiraudon (2006) demonstrated that specific constraints operate either at the level of policy formation, where various policy ‘inputs’ are filtered so that particular policy choices (‘outputs’) prevail, or the implementation stage, where these ‘outputs’ are translated into actual policy ‘outcomes’. Also Joppke (1998, p. 267) suggested an analytical distinction between what he sees as “two separate aspects of sovereignty, [namely] formal rule-making authority and the empirical capacity to implement rules”. In relation to the latter, he notes that the capacity of states “to control immigration has not diminished but increased – as every person landing at Schiphol or Sidney airports without a valid entry visa would painfully notice” (p. 270). This seems to suggest that states are more constrained in establishing the rules according to which they grant entry and residence titles (as in the case of family migrants or recognised refugees, who they have to admit), than in enforcing those rules at their external borders.

A second and cross-cutting distinction can be drawn between control policies targeting those foreigners trying to enter and those who already live within the country, i.e., between constraints to external versus internal immigration control. In her case study of local immigration bureaucracy in Germany, Antje Ellermann (2006) shows that individual enforcement officers often face significant resistance – both from an organised public and elected municipal officials – against the deportation of local residents. Her analysis suggests that also here, the rationales and constraints underlying the making of these policies tend to be different from those shaping their implementation: Whereas “at the legislative stage, demands for ‘cracking down’ on immigrants are quickly established, policy debates are framed in pro-regulatory, rights-restricting ways, and little attention is paid to the costs of regulation”, these costs become drastically visible to the public and can easily turn into an obstacle for implementation where friends or neighbours face imminent deportation (Ellermann, 2006, p. 296). Put in De Genova’s (2013) terms, resistance arises where the ‘obscene of (everyday) inclusion’ gives way to the ‘scene of exclusion’, and where migrant ‘illegality’ suddenly becomes a human face that is not (anymore) that of a distant stranger.

One site that plays a significant role for the constant (re-)negotiation of (irregular) migrants’ inclusion and exclusion and has thus become an important venue for internalised immigration control, is the liberal welfare state (Bommes & Geddes, 2000). Stephen Castles (2004, p. 216) sees the latter as “a major factor driving the incorporation of immigrants […] because it follows a logic of inclusion: failure to grant social rights to any group of residents leads to social divisions, and can undermine the rights of the majority”. But the welfare state is also “a stratification system in its own right” (Esping-Andersen, 1990, p. 4) and as such – and although it was invented to reduce problems of social inequality and stratification – it also modifies or (re-)produces existing inequalities. Central to these processes is what Anne L. Schneider and Helen M. Ingram (1993, p. 336) called the “social construction of target populations”. Their model identifies four different (ideal) types of target populations: “advantaged”, “contenders”, “dependents”, and “deviants” that are differentiated by their either weak or strong political power, and their (overall) positive or negative social construction. It is a way to explain why different people become the target of different kinds of policy tools (which are also justified in different ways). Following this model, irregular migrants are generally constructed as “deviants”, that is, “negatively constructed powerless groups [that] will usually be proximate targets of punishment policy” (Schneider & Ingram, 1993, p. 337). Only in exceptional cases, like that of a child or pregnant women, will they be constructed as “dependents”; that is, more positively (and thus more deserving of certain support) but still with little or no political power (and thus unable to make substantial demands on the welfare system).

At least to a degree, such categorisations become internalised not only by the target population itself (Soss, 2005), but also by the providers and administrators of public services (Willen, 2012; Cuadra & Staaf, 2014; Landolt & Goldring, 2015). As I will show, individual welfare workers often rely on official labels and sometimes reproduce the same stereotypes that also dominate public discourse and political debates, like the often proclaimed need to protect the (national) welfare system against ‘health tourists’ or ‘benefit-scrounging foreigners’. But street-level bureaucrats also regularly contest or simply modify (some) claimants’ official categorisation. By treating some of them as “dependents” instead of “deviants”, they can effectively shift the everyday boundaries of deservingness, vulnerability and belonging. All this seems to suggest that when it comes to internal immigration control and enforcement, liberal states are more constrained in implementing than in making restrictive rules. Arguably, this has to do with the type of actors and venues involved in this kind of control, as well as the fact that those who are its target already live within, and thus in various ways form part of, the host society.

Notably, most of the explanations for receiving states’ failure to effectively control and limit unwanted immigration build on some notion of inherent contradiction or inconsistency, whether between competing (or simply different) normative principles, actors and their interests, or institutional logics. According to Christina Boswell’s (2007) influential conceptualisation of migration policy, states themselves are constantly torn between the fulfilment of their various ‘functional imperatives’, namely: (i) to promote a just distribution of resources (‘fairness’), (ii) to provide ‘security’ for its subjects as well as (iii) the necessary conditions for the ‘accumulation’ of wealth, and (iv) to respect the constitutional principles and individual liberties of those affected by its jurisdiction (‘institutional legitimacy’). While each of them constitutes an essential precondition for sustaining the state’s “legitimacy and capacity to govern” (p. 88), they tend to have contradictory policy implications and are therefore difficult (or even impossible) to realise simultaneously. In her view, the best explanation for the observed ‘gap’ between (restrictionist) policy goals and (more liberal) outcomes is that “a state unable to simultaneously meet all functional requirements may have an interest in the persistence of contradictions and inefficiencies in policy” (p. 93). For Bill Jordan et al. (2003, p. 211) it is precisely because legislation often reflects a compromise between competing interests that “the dilemmas of policy-making remain, at least partly, unresolved and are transferred to the implementation stage”. The last two sections of this chapter will therefore focus on how public organisations and the individuals acting within them deal with these ambiguities as they implement policy.

Another theoretical perspective that is helpful for the study of how immigration policies work within society is offered by political sociology, as suggested by Giuseppe Sciortino (2000). Instead of the state, he takes society itself as the basic unit of analysis and understands it, following Niklas Luhmann (1982b), as an entity that “has no head, no base and no center, but is articulated in a plurality of specialized subsystems that have their own set of symbolic codes, leading values, operational programs and regulative means” (Cvajner & Sciortino, 2010b, p. 392). Such perspective allows to take into account the different organisational cultures and logics, shared norms, professional identities, values and codes of conduct that guide the actions and shape the interests of professionals working in those societal subsystems that only recently are becoming part of the immigration regime (Jordan et al., 2003). It is thereby well suited to identify the various contradictions that arise where the particular logic of immigration control – based on the fundamental distinction between regular and irregular migration – intersects with, for example, the imperatives of local governments to manage housing and administer mainstream services to local residents, or that of a doctor to treat a patient, or a social worker to meet the needs of a homeless person.

Sciortino (2000, p. 220) asked a significant question about the role and motivation of individual actors: “Has the person who hires an undocumented immigrant really also lobbied in favour of a weak enforcement of border controls?” Since the answer will often be ‘no’, it very well illustrates the need to shift the focus of analysis: from the rather abstract idea of competing ‘powerful interests’ behind the making of immigration policies to the subsequent and much more concrete re-negotiation, bending and selective transgression of the resulting rules and regulations by implementing actors. Another question could then be: What risk does the person who hires (or provides a service to) an undocumented immigrant assume in doing so; and where does this risk come from? I thereby want to point at a potential shortcoming of conceptualising migration policy based on Luhmann’s “fully horizontal perspective, where each differentiated functional context sees ‘the world’ according to its own code and treats all the other contexts as its external environments” (Sciortino, 2000, p. 221). The danger here is to automatically assume that external influences have little or no meaning and thus authority within a particular subsystem, and hence to lose sight of the actual power relations that link the various subsystems and thereby define how (and which) meanings and logics are transferred from one to another.

What is required, therefore, is a theoretical approach that recognises the functional differentiation of society without reducing the role of the state to that of a passive and neutral ‘broker’ between competing societal interests (Boswell, 2007). First of all, such approach must understand ‘the state’ itself not as a unified and monolithic entity but a fragmented aggregation of various administrative bodies that are partly driven by their own interests and functional imperatives (Gill, 2010). Such an understanding is reflected in the notion of an ‘assemblage’ of governance (Walters, 2015) or state power (Allen & Cochrane, 2010), but also in post-Foucauldian scholarship that understands ‘the state’ not as an absolute concentration of power, but rather a “site at which power condenses” (Cowan & McDermont, 2006, p. 182, cit. in Mckee, 2009, p. 476). Also Foucault’s own interpretation of power in terms of ‘governmentality’, as Didier Fassin (2011, p. 217) has noted,

does not so much focus on the power of the nation-state as on the limits of its ideal-typical representation as coherent, impartial, and effective. On the contrary, it shows its illegality and illegibility, demonstrates its partiality and ineffectiveness, but also establishes the functionality of these apparent dysfunctions.

Secondly, then, more attention needs to be paid to the power relations at work between (and within) the various interests of central and lower levels of government as well as different state and non-state agencies; and to understand, like Boswell (2007), the influence of liberal institutions not only as a function of their relative autonomy, but also the ‘resonance’ of their own interests with (some of) those of ‘the state’. What emerges are various ‘assemblages of power’, within which – as Sandro Mezzadra and Brett Neilson (2013) show – both ‘governmental’ and ‘sovereign’ forms of power overlap and interact with each other, rather than the former having largely replaced the latter, as Foucault’s earlier work suggested.

Thirdly, it also requires a more dynamic and nuanced understanding of the varying degrees of autonomy and margins of discretion given to those individual actors who ultimately implement policy within such ‘assemblages of power’. Both are determined not only internally – by the professional identity and institutional logic dominating a particular field of work (such as healthcare), but also externally – via binding regulations through which the government tries to ensure a more effective implementation of its rule regarding other policy areas (such as immigration).

Accordingly, my study focuses on the intersection of not only sovereign and governmental forms of state power but also the internal and external imperatives that trigger individual and organisational action on the ground. Even a partial convergence of internal and external logics can thereby be expected to enhance compliance with a particular set of rules, while contradictions between the two are likely to trigger resistance and thus hinder their implementation. By highlighting these processes of (re)negotiation and contestation, my approach helps to explain local policy outcomes without framing them in terms of either ‘success’ or ‘failure’. From this perspective, receiving states appear less as the ‘managers’ of migratory processes as such, than of the challenges that migration – but also the control of migration – poses to their own functioning, legitimacy and sovereignty.

One increasingly important way for the state to manage the contradictory interests and imperatives triggered by immigration is what Lydia Morris (2002, p. 19) called ‘civic stratification’, whereby “the rights and protections afforded by the state to different ‘entry’ categories constitute a system of stratified rights closely associated with monitoring and control”. This system – which places irregular migrants at the very bottom of the hierarchy – thereby relies on various ‘dividing practices’, similar to those described by Foucault (2002a, p. 326) as the ‘objectivizing of the subject’: “The subject is either divided inside himself or divided from others. This process objectivizes him. Examples are the mad and the sane, the sick and the healthy, the criminals and the “good boys”.” As I will discuss in the following, also the control of migrant irregularity first of all requires the separation of the irregular from the regular migrant.

2.2 The ‘Unmanaged’: Irregular Migrants as the Exception to the Rule

Preventing illegal entry and residence is one of the key issues addressed within the migration management discourse, which thereby tends to suggest that within a perfectly managed migration system irregularity would simply cease to exist. Historically, however, migrant irregularity has always been directly linked to national frameworks of immigration regulation and restriction, and thus only became a major policy issue in the aftermath of World War I, when the consolidation of these regimes gave rise to the emergence of what James Hollifield (2004) called the ‘migration state’. Early examples of systematic immigration restrictions imposed by modern nation-states were usually directed against particular groups of foreigners, whose entry and presence were deemed undesirable based on rather specific characteristics (Düvell, 2006). Today, in contrast, immigration restrictions target all those who do not fulfil the ever more complex and selective requirements for legal entry, stay and employment in a particular country. In most cases, those to be excluded are thus negatively defined, so that “the contours of illegality mirror those of legality, [and] the meaning of illegality depends on that of [other] migrants’ legality” (Garcés-Mascareñas, 2010, p. 80). Hence, in their endeavour to effectively manage migration, liberal states not only create specific patterns of ‘legal’ immigration according to their economic and political needs, but they also, though less explicitly, produce ‘illegal’ immigration (De Genova, 2002; Samers, 2004; Inda, 2006; Goldring et al., 2009).

On one hand, this perspective relates the empirical increase of irregular migration to the growing restrictiveness of receiving states’ border control policies (de Haas et al., 2018). In Europe this has been the case since the 1970s, when the active recruitment of foreign workers was suddenly stopped and gradually replaced by a stricter policing and externalisation of borders, ever more restricted access to asylum as well as family-related migration, and highly selective regulations on temporary labour migration. On the other hand, the ever-increasing complexity and diversification of these various strands of policy also explain some of the conceptual and terminological difficulties surrounding contemporary migrant irregularity. For the purpose of this study, irregular migrants are defined as non-EU citizens who according to the immigration law of the country in which they reside lack the formal permission to do so. Their condition vis-à-vis the host state and its (local) institutions is thus characterised, on one hand, by their irregular immigration status and, on the other, by their being local residents. I therefore also refer to them as irregular residents and specifically speak of migrant irregularity where I want to remind the reader that the problems I describe are not caused by (the actions of) particular human beings but follow from the administrative situation that these people are in.Footnote 3

Critical migration and border scholars have intensely debated the terminology to best be used when describing and analysing the meaning of irregularity (or ‘illegality’) as well as the processes through which it is produced and imposed on individuals (see Bauder, 2014). Unlike others, I prefer the term ‘irregular’ over ‘undocumented’ or ‘illegalised’ (im)migrants: Over the former because ‘undocumented’ literally suggests a lack of any documentation that could certify the person’s identity. Almost all of the irregular migrants I met, however, did possess a passport or other ID, although many of them were reluctant to use it for any or at least any official purpose. As I will show, the difficulty for local authorities and welfare institutions is often precisely a lack of documentary evidence – of a person’s identity, age, income, family relationship, or address – but not necessarily their irregular immigration status as such. In everyday practice, as well as analytically, it therefor makes sense to differentiate between genuinely ‘undocumented’ and ‘irregular’ migrants. Not all undocumented migrants are irregular, and relatively few irregular migrants are completely undocumented.

I completely agree, on the other hand, that there are very good reasons for academics to speak of migrants as being ‘illegalised’ instead of ‘irregular’. For example, Harald Bauder (2014, p. 229) argues that “‘irregular’ still implies that migrants somehow are not ‘regular’” and that it describes “the outcome of the process of illegalization and thereby conceal[s] the process itself” (see also Squire, 2011). My study, however, focuses on the effects of irregularity (not the process of its production) and, more specifically, the different ways in which it functions and is thereby re-negotiated within various social and institutional settings. What interests me is the precise sense in which a lack of immigration status renders someone ‘not regular’ from the perspective of, for example, the healthcare system; and thus, what exactly distinguishes the person that has been assigned this status from a ‘regular’ patient (or resident, student, welfare recipient, etc.).

The remainder of this section looks at migrant irregularity from various perspectives: first as a theoretical concept and device of both sovereign and governmental power (Sect. 2.2.1), then as a condition that states try to ‘manage’ both directly – through measures of deportation and regularisation (Sect. 2.2.2) – and indirectly – by compelling various actors and institutions to identify and exclude irregular migrants from services they provide to other local residents (Sect. 2.2.3).

2.2.1 Migrant Irregularity as a Gesture of State Sovereignty and a Device of Governmentality

Critical scholars working in the fields of migration and border studies have often questioned the strict dichotomy between ‘legal’ and ‘illegal’ migratory status. Some have suggested alternative concepts capable of describing a certain continuum of in-between statuses (Ruhs & Anderson, 2010; Kubal, 2013; Triandafyllidou & Bartolini, 2020); others emphasised the diversity of potential paths into and out of irregularity (Black et al., 2006; Cvajner & Sciortino, 2010a; Düvell, 2011; Vickstrom, 2019). A closer look at this growing body of literature allows distinguishing further dimensions of complexity that go well beyond the notion of mere diversity. Some scholars emphasise the fluidity of migrant status: not only do individuals repeatedly move between ‘legality’ and ‘illegality’ (Calavita, 2003; Vickstrom, 2012), but also the underlying legal categories tend to change over time (Couper & Santamaria, 1984; Düvell, 2006). Others highlighted a certain stratification or hierarchy that exists even within irregularity (Cvajner & Sciortino, 2010a; Paoletti, 2010; Chauvin & Garcés-Mascareñas, 2012), as well as the chance of migrants becoming more or less ‘illegal’ through incorporation (Chauvin & Garcés-Mascareñas, 2014; Schweitzer, 2017).

Also the potential simultaneity of regularity and irregularity has been shown; that is, the possibility of irregular migrants being incorporated into some areas of society but at the same time excluded from others (Castles, 1995; McNevin, 2006; Ruhs & Anderson, 2010; Mezzadra, 2011; Van Meeteren, 2014; Chauvin & Garcés-Mascareñas, 2020). This can partly be explained by the fact that immigration status is immediately relevant only in some social and institutional contexts or spheres of everyday life but rather irrelevant in others (Cvajner & Sciortino, 2010b). Another important part of the explanation is that also “the law itself […] excludes and includes at the same time”, as Sebastian Chauvin and Blanca Garcés-Mascareñas (2012, 2020, p. 36) argue. De Genova (2013) therefore even speaks of ‘inclusion through exclusion’, while Mezzadra and Neilson (2013, p. 159) employ the notion of ‘differential inclusion’ to describe how some migrants’ inclusion “can be subject to varying degrees of subordination, rule, discrimination, and segmentation”. They thus attribute a certain function and intentionality to migrants’ ascribed irregularity, which thereby appears as a tool to perpetuate their subordinate position within local and global labour markets.

Additional complexity comes with the fact that in many everyday encounters, status irregularity intersects with other dimensions of a person’s identity, especially their race/ethnicity, class and gender. Leslie McCall (2005, p. 1771) defines intersectionality as “the relationships among multiple dimensions and modalities of social relations and subject formations”. As an analytical perspective, it means more than adding different forms of relative disadvantage (or privilege) onto each other. An intersectional perspective allows careful tracing of when and how the various dimensions mutually reinforce or neutralise each other (Weber, 2001). This can bring to light the combined effect of very different kinds of inequality and forms of discrimination. It also shows that someone’s exclusion or inclusion is seldom a question of immigration status alone, but becomes intensified or attuned by other characteristics, like being a woman or having darker skin. While much of the literature on migration-related intersectionality has focussed on labour market outcomes (like wage-inequality) (e.g. Browne & Misra, 2003), it has also been shown to shape migrants’ positionality in relation to immigration enforcement. For example, the relatively ‘hidden’ nature of domestic work tends to decrease the risk of detection for those – usually women – who perform this work, though often in exchange for a higher risk of exploitation. Also here, as Irene Browne and Joya Misra (2003, p. 505) highlighted, “it is the intersection of race/ethnicity, gender, class, citizenship status, and other factors that help explain the extent of the exploitation that these workers face”.

Kara Cebulko’s (2018) comparative study of different groups of irregular residents of Massachusetts, in turn, shows how the intersection of irregularity with either racial or social class privilege benefits those irregular migrants who are lighter-skinned and/or more middle-class. In practice, this “privilege without papers” can yield “tangible educational opportunities” and reduce suspicion during encounters with state officials (ibid., see also Romero, 2008), while it does not provide actual protection against deportation. This strand of literature thus also highlights the extent to which place and context matter for how intersectionality plays out in real life. It has been shown, for example, that “the same economic environment creates advantage for some groups of women and disadvantage for other groups of women relative to similarly situated men” (McCall, 2005, p. 1790; Valdez & Golash-Boza, 2020). Also this important fact – “that different contexts reveal different configurations of inequality” (McCall, 2005, p. 1791) – is relevant for my own comparative analysis of different spheres of service provision.Footnote 4

Through these various and crosscutting processes of inclusion and exclusion, irregular residents become enmeshed in a range of social and power relations, which can trigger formal or informal bordering practices as well as their contestation. Their being part of a local community, sports club or parents’ association, for example, can be a source of empowerment, while working in the informal economy or even the use of public transport might increase their risk of detection and deportation (Van Meeteren, 2014; Schweitzer, 2017). Given this complexity, Mezzadra and Neilson (2013, p. 168) argue that “neither sovereign nor governmental conceptions of power are adequate to account for current border politics and struggles”. In their book Border as Method, they show that

[b]orders are becoming increasingly governmentalized or entangled with governmental practices that are bound to the sovereign power of nation-states and also flexibly linked to market technologies and other systems of measurement and control (2013, p. 176).

Arguably, this is particularly true for internal borders, such as those regulating the access to most public services and institutions of the welfare state. Although their original function is not immigration control, they are becoming crucial sites for the management of (irregular) migration. This is possible because immigration legislation not only renders the entry or presence of certain migrants ‘illegal’, but thereby also prescribes the range of actions that others can or must (not) take towards them without potentially breaking the same law themselves. Sciortino (2004, p. 37) therefore argues that the “significance of the irregular status is highly correlated to the scope of states’ controls over the interactions and exchanges taking place on their territories.” At a more general level, this also reflects Foucault’s (2002b, p. 123) suggestion to understand the state itself as ‘consisting’ in “the codification of a whole number of power relations that render its functioning possible”. Seen from this perspective, migrant ir/regularity operates as a code that is attributed to a person by the immigration system and that manifests itself only in the (lack of a legal) immigration status. Arguably, ir/regularity is the most fundamental code of the immigration control logic; but even so, it is neither readable nor meaningful within most other subsystems or power relations. Only through specific laws, regulations and the corresponding documentation and identification systems (Torpey, 1998, 2000; Bigo, 2011) can the meaning of migrant irregularity be transferred to, or imposed upon, other spheres of social life and interaction. This is a crucial precondition for an effective internalisation of immigration control.

The person who hires an irregular migrant – to return to the example of before – can only be aware of doing so after checking the worker’s passport or residence card. If s/he then decides to go ahead it is probably either because s/he wants to help the other, or because s/he knows the other will accept a lower wage. If s/he refrains from hiring after discovering the other’s irregularity, then probably because of the risk of being checked and punished by the same authority that might also initiate a procedure to deport the worker. Neither of the two outcomes can be fully explained by the internal logic of the labour market (i.e., the decisions or organisational processes that normally assign a particular person to a certain job), nor the fact that the worker has no legal immigration status. Instead, whatever the outcome of this situation, it follows from the particular way in which the government in question enforces its immigration regulations upon the labour market. Generally speaking, whenever a particular logic is transferred to another sphere, its specific codes have to be ‘translated’, whereby their meaning can change, or new meanings be added. Here, the worker’s potential exploitability and the employer’s risk of having to pay a certain fine (or face a prison sentence) are the new meanings that migrant irregularity acquires when transferred from the sphere of immigration control to that of employment.

This example also shows that the meaning(s) that migrant irregularity has or acquires through translation can either be in line or contradiction with the interests and institutional logics that otherwise dominate the respective sphere or subsystem. In the case of employment, this relationship is rather straightforward: it might be lucrative but is undoubtedly against the law and thus entails a concrete risk to employ an irregular migrant. As I will show, the situation often becomes more ambiguous in the area of public service provision, where the logic of immigration control confronts powerful normative entitlements combined with intrinsic functional logics and particularly strong professional ethics. Together, they often demand at least a certain level of inclusion irrespective of immigration status. Before focussing on this issue in more detail, however, I provide an overview of the concrete policies through which states generally try to ‘solve the problem’ of irregular migration.

2.2.2 Managing Irregular Migration Through Deportation and Regularisation

In principle, the policy options available to a state facing sizeable (although typically uncertain) numbers of irregular migrants already living within its borders, are rather limited: On one hand, governments can (and quite often do) tacitly accept the unlawful presence of some of these foreigners. This, however, limits the extent of control they effectively and symbolically exercise over the territory and population. Precisely in order to ‘stay in control’, on the other hand, states can either legalise irregular migrants’ presence in the country, or physically remove them from both their territory and jurisdiction. Potential policy measures to ‘eliminate’ or at least reduce irregularity can thus be thought of as a continuum that ranges from regularisation, i.e. offering possibilities of ex post legalisation of immigration status,Footnote 5 to deportation, which can broadly be defined as the expulsion of a person from state territory by threatened or actual use of force (Anderson et al., 2011). While the extension of certain rights to migrants in irregular situations lies at the inclusionary end of this spectrum, policies of so-called ‘voluntary’ or ‘assisted’ return as well as those aiming to ‘discourage’ irregular stay are closer to the opposite extreme.

Both regularisation and deportation are part and parcel of migration management and serve pragmatic as well as symbolic functions. Both have been described as constitutive elements of citizenship (De Genova, 2002, 2010; Walters, 2002) and nation-building (McDonald, 1969), and thus provide evidence of the persistence of state sovereignty (Gibney & Hansen, 2003; Castles & Miller, 2009). Particularly the practice of deportation plays a key role in reinforcing the legal and normative boundaries of membership and belonging to a national community (De Genova, 2010; Anderson et al., 2011). Although regularisation at least questions these boundaries by offering formal possibilities to transcend the strict dichotomy between ‘legal’ and ‘illegal’ residence status, it always only does this for certain kinds of irregular subjects, who are framed as relatively more deserving or less unwanted than others. In policy discourses, deportation is often justified as a simple necessity for maintaining the effectiveness and credibility of the immigration system (Fekete, 2005; cf. Anderson et al., 2011), while regularisation has been criticised for undermining the legal framework and is frequently seen as a consequence (or even instance) of policy failure (Levinson, 2005; cf. Finotelli & Arango, 2011). Many governments justify their reluctance to grant an ‘amnesty’ with the fear that it might attract further irregular immigration and thus could have a so-called ‘magnet effect’ (OECD Secretariat, 2000).

In spite of these drawbacks, offering opportunities for regularisation to persons who have either entered a country unlawfully, overstayed their visa, or for other reasons find themselves in irregular situations has become a widespread practice within and beyond Europe (Apap et al., 2000; OECD Secretariat, 2000; Levinson, 2005; Finotelli & Arango, 2011). Between 1973 and 2008, more than 4.3 million people were ‘regularised’ within the EUFootnote 6 through a total of 68 national programmes (Kraler, 2009). Such regularisation exercises can either take the form of a permanent procedure, i.e. an on-going process open to an infinite number of claims, or that of one-off procedures, which are carried out within a fixed timeframe and often target a specific category and therefore finite number of people (Apap et al., 2000). While the former are part of the regular policy framework for the management of migration, the latter are often based on extraordinary, or ad hoc legislation, as both Kate Brick (2011) and Albert Kraler (2009) noted.

Regularisation can be justified in various ways: Joanna Apap et al. (2000) argued that such policies are put in place either for reasons of fait accompli, whereby a right of residence is derived from the recognition that a person has de facto (although ‘illegally’) been present since a specific date; or for reasons of protection against certain risks that a particular person would be subjected to if not granted legal status. The way (and extent to which) these objectives are effectively translated into policy outcomes depends on the set of criteria that have to be met by immigrants in order to become eligible for regularisation.Footnote 7 From a critical perspective, Jean McDonald (1969, p. 71) therefore argued that by “[d]istinguishing the criminal from the good, the diseased from the healthy, the lazy from the hard-working, the newly arrived from the loyal, […] the regularization process is a nation-building practice”, which by itself contributes to the reproduction of migrant ‘illegality’ instead of reducing it. By choosing the underlying criteria and setting the thresholds the government can regulate both the scale and scope of any regularisation exercise. Drawing these distinctions, however, often requires assessments by social workers, doctors and other street-level bureaucrats who are in direct contact with potential beneficiaries. The effectiveness of regularisation thus often depends on the involvement and agency of the people who are at the centre of my study.

Policies of deportation represent the opposite, explicitly exclusionary side of the spectrum of available measures to reduce the number of unlawful residents. Traditionally seen as “the state’s ultimate and most naked form of immigration control” (Gibney & Hansen, 2003, p. 1), deportation has nowadays, as De Genova (2010, p. 34) argued, “achieved an unprecedented prominence […and] seems to have become a virtually global regime”. With the notable exception of foreign nationals convicted for committing a crime in the host country, the groups targeted by this regime very much resemble those who may also qualify for regularisation, including visa-overstayers, clandestine entrants, irregular workers and rejected asylum seekers.

For a broad range of reasons, however, only a relatively small fraction of all individuals who are theoretically eligible for deportation is actually deported, a fact that Mathew Gibney (2008) described as the ‘deportation gap’. On one hand, there are several practical constraints which render deportation a difficult, expensive and time-consuming measure: most importantly, it requires appropriate documentation linking the deportee to a particular ‘home’ state, as well as that state’s cooperation or at least willingness to recognise and readmit the person to its territory. By absconding or obscuring their identity or origin, individuals threatened by deportation can thus actively delay or even prevent their expulsion. On the other hand, forceful removal of certain individuals is often constrained by governments’ obligations under human rights treaties, and as a coercive state practice it has increasingly come under scrutiny and critique by NGOs, migrant advocacy groups and human rights activists (Fekete, 2005; Ruedin et al., 2018). One of the main challenges that states confront, arises from individual immigrants’ social integration into the host society, which over time can “form a moral basis for remaining”, independent of formal entitlements (Gibney, 2008, p. 150; Paoletti, 2010; Schweitzer, 2017). Social relations such as those established within the neighbourhood, school or church community, for example, often trigger considerable resistance against deportations, which in turn tends to render them unpopular with local politicians (Ellermann, 2006; Anderson et al., 2011).

For De Genova (2002, 2010) deportation also fulfils a clearly disciplinary function, whereby it is not so much the act of deportation itself that is decisive, but rather immigrants’ constant ‘deportability’, i.e. the sheer possibility (and uncertain likelihood) of being deported. This specific condition can again be seen as a continuum that ranges from facing immediate expulsion to being under very little threat of actually being deported ever. While it also extends to ‘legal’ immigrants – thereby radically reinforcing the distinction between ‘native’ citizens and ‘aliens’ (Paoletti, 2010; Anderson et al., 2011) – the actual risk of being deported is most relevant to those lacking a legal residence status or other right to remain. For them in particular, “the possibility of removal […] casts a long, dark shadow over their daily lives, threatening at any moment to take away from them the little they have gained by residence in the host country” (Gibney, 2011, p. 43).

At the other extreme there are individuals who for whatever reason and although facing a formal deportation order cannot be deported in practice and are thus effectively ‘non-deportable’, as Emanuela Paoletti (2010) argued. What makes this observation particularly relevant for my study, is that for her, “[t]he complex net of rights and duties that link the state and the non-deportable opens up a more fluid conceptualisation of membership” (Paoletti, 2010, p. 13). According to her analytical framework, the intersection of irregular migrants’ ‘relative desirability’ (within certain social spheres) with the state’s limited capacity to enforce their deportation leads to various forms of quasi-membership. What she does not explicitly take into account, however, is what could then be called irregular migrants’ regularisability; that is, their actual prospects of fulfilling all the requirements set by the state in question for ex-post legalisation. The latter often reflect the same notions of desirability and deservingness that can also render irregular migrants less deportable, like having close family or other social ties or being seen as contributing to the host community.

Although not everyone who is ‘non-deportable’ can be regularised, or vice versa, it is always between these two poles that migrants in irregular situations must negotiate and construct their fragile position and claims for incorporation and membership in the host country. For Garcés-Mascareñas (2010) it is therefore precisely the quite often simultaneous possibility of being either regularised or deported, which defines the condition of irregularity. Seen from this perspective, regularisation and deportation are more than two functionally opposed policy approaches through which states can reduce the number of irregular migrants living within their territory. They are also the carrots and sticks through which irregular residents can be disciplined even without being in direct contact with ‘the state’ or its immigration regime. Particularly for those living ‘under the radar’, taking the necessary steps towards a possible regularisation of their status might in fact increase their deportability (by becoming known to the authorities, for example); whereas successfully evading deportation long enough (and without breaking any other law) will usually better their chances to eventually qualify for regularisation (Schweitzer, 2017).

In various ways, regularisation and deportation thus play a crucial role for the micro-management of irregular migration. While it lies within the competence of the state to establish the corresponding legal frameworks, these will have significant consequences for other actors and their interactions with migrants in irregular situations. In relation to public services, irregular migrants’ real and perceived deportability and prospects for regularisation will have an impact on the claims they might be able and willing to make. Inversely, the imperatives to provide them with at least certain services and thus to accommodate some of these claims in spite of their irregularity will be more pressing where they are unlikely to be deported or regularised any time soon. At the same time, lower levels of government as well as public institutions can become involved in the implementation of these policies. This can be by attesting the fulfilment of certain requirements for regularisation, such as continuous residence, school attendance or other instances of ‘integration’; or by helping the immigration authorities to identify potential deportees. Both kinds of involvement indicate an increasing internalisation of immigration control, whereby national governments transfer part of their own responsibility to various non-state actors and local institutions, including those providing public services. As I will discuss in the following subsection, this gives rise to various degrees of internal exclusion, but also localised forms of inclusion towards irregular residents.

2.2.3 Managing Irregular Migration Through Internal Exclusion and Inclusion

Many Western governments increasingly address the issue of irregular migration by restricting not only the access of unlawful immigrants to the territory of the state but also that of unlawful residents to employment, housing, healthcare and other services (Bommes & Geddes, 2000; Guiraudon & Lahav, 2000; van der Leun, 2003; Lahav & Guiraudon, 2006; Broeders & Engbersen, 2007; Squire, 2011). Facing a growing permeability of its external borders, it is argued, the state “raises a protective wall of legal and documentary requirements around the key institutions of the welfare state and ‘patrols’ it with advanced identification and control systems” (Broeders & Engbersen, 2007, p. 1595). This shifts some of the burden of immigration enforcement to a wide range of actors beyond the level of the nation-state and hitherto detached from its immigration authorities. They often include the local police and employers, and sometimes also banks, landlords, welfare officers or other public officials, as well as private citizens. James Walsh (2014, p. 242) describes this development as a form of ‘deputization’, which he generally defines as “the activation and empowerment of certain individuals to participate in preventing and controlling legal transgressions”. For the purpose of this study, I define deputisation more narrowly, as involving a formal obligation to carry out certain immigration control duties. Building on Marrow’s (2011) earlier conceptualisation of ‘bureaucratic and civil cross-deputization’, Walsh distinguishes ‘deputization’ from ‘responsibilization’ – whereby third party participation is not obliged but encouraged – as well as ‘autonomization’ – which is unsolicited and happens spontaneously, possibly even against the will of the authority. These various modalities of activation are thus characterised by different degrees of autonomy and voluntariness on the part of implementing actors and will help me to map the positions of different welfare workers in relation to immigration control.

Whether participation in it is obliged, encouraged or autonomous, internal immigration control always implies agency: In order for someone’s (irregular) immigration status to become a barrier when trying to access a particular service or engaging in a certain activity or interaction, someone else has to exercise a specific kind of control. More and more people thereby become engaged in what John Torpey (1998) described as ‘techniques of identification’ based on documents like the passport, through which states codify not only the identity but also the national belonging of their subjects. In the context of public service provision, it is often the receptionist or other front-line staff who are obliged or encouraged (and sometimes themselves decide) to determine service users’ immigration status together with their identity and/or concrete needs. Other welfare workers are more or less explicitly prevented or discouraged from enquiring or even considering immigration status as part of their dealings with service users, which I refer to as shielding.

In practice, deputisation, responsibilisation, autonomisation and shielding are not only relevant in relation to the task of finding out the immigration status of service users, but also the sharing of this information with the relevant state authority. Such information exchange can be a matter of formal obligation, encouragement or choice – or it can be explicitly prevented through what is often called a firewall (Carens, 2013; FRA, 2013; OHCHR, 2014). The latter can be understood as any mechanism, rule or practice “through which clear divisions between migrant policing and provisions of basic rights may be established and maintained” (Hermansson et al., 2020, p. 3). One important function of firewalls is to prevent individuals or organisations from sharing information with immigration authorities, which effectively hinders immigration enforcement. Regarding the nature of these mechanisms, Dennis Broeders and Godfried Engbersen (2007, p. 1595) noted that “[w]hether or not governments connect and combine different bodies of information will increasingly become a matter of legal constraints, as the technological constraints are quickly losing their relevance”.

The internalisation and dispersal of control functions has in some cases “incorporated new actors whose own interests coincided with those of national control agencies”, as Guiraudon and Lahav (2000, p. 177) argued. But there are also instances and sites where quite the opposite is true, in that the logic of control that underlies state efforts to reduce irregular migration conflicts with the own interests or professional duties of the new ‘deputies’. As a result, policies of internal control not only encounter resistance from local residents, civil society and activist groups, but also from professionals, civil servants and local government officials who (sometimes) “put their professional ethics above state policies” (Ellermann, 2006; Van Der Leun, 2006; Broeders & Engbersen, 2007, p. 1606). Two important grounds for criticism against internalised control are that it violates irregular migrants’ human rights and has negative effects on the communities in which they live (PICUM, 2010; Carens, 2013). The former is because such policies often involve bordering practices that (can) lead to effective exclusion of irregular residents from fundamental services like primary education or emergency healthcare. The latter reflects the fact that in spite of their irregular status, they are embedded within various social structures in both public and private domains, such as the neighbourhood in which they live or work, a church community, sports club, parent association, or their own family network. Internalised immigration control puts a disproportional burden on several ‘key social transactions’ (Cvajner & Sciortino, 2010b) and has been shown to push irregular migrants further underground, thereby increasing their reliance on informal and sometimes criminal networks or activities (Broeders & Engbersen, 2007; Crawley et al., 2011; Refugee Council, 2012). For Paoletti (2010, p. 19), “[t]he stripping from such individuals of basic rights and access to essential services can be in itself considered not only a human rights infringement but a deliberate act of exclusion from society”. While this is certainly true, the crucial question is whether the opposite is too: Does the granting of such rights or access to these services constitute a ‘deliberate act of inclusion’?

Chauvin and Garcés-Mascareñas (2020, p. 45) recently argued that irregular migrants’ partial inclusion is rather the result of certain “structural concerns [related to] public education, public health, public order, road safety, economic and urban planning, and so on […]” which they see as “acting at a deeper level than perhaps more superficial or “ideological” justifications for inclusion such as human rights”. Hence, states grant access to unlawful residents not to include them but because they “have a greater stake in regulating the actual population than in tracing boundaries between members and non-members” (ibid., p. 43). While Chauvin and Garcés-Mascareñas thereby point at the “inclusive tendencies of governmentality” (p. 44), the governmental nature of internal control policies themselves also becomes apparent in other – more exclusionary – ways: Firstly, in that their aim is to persuade unlawful residents that they themselves actually want to leave (and thereby avoid exclusion and marginalisation); and secondly, in that the state exercises its power to exclude or include irregular migrants not directly but through their interactions with other members of the population, particularly ‘street-level bureaucrats’ (Lipsky, 1980) whose position within public institutions often involves some kind of gate-keeping function. Together with the rendering of certain individuals or groups as more or less deportable or worthy of regularisation, these forms of indirect control can be seen as part of what Foucault (2002a, p. 328) called “the political management of society”, an endeavour that simultaneously involves multiple cross-cutting bordering practices to be employed by different actors within various organisational fields.

As a mode of analysis, in turn, governmentality allows to discover these “multiple sites of governing beyond the traditional boundaries of the state apparatus […b]y highlighting how government is ubiquitous in all social relationships” (McKee, 2009, p. 469). The concrete sites where irregular residents’ access to rights and services is being negotiated – like classrooms, welfare offices or reception desks – thereby also represent what Vicki Squire (2011, p. 6) described as ‘borderzones’: “dispersed, multi-dimensional and contested sites of political struggle”. For John Bowen and his colleagues (2013, p. 3), it is in these “varied and relatively autonomous social contexts that boundaries are created or reaffirmed in ways that have the sanction of the state behind them”. Those who implement the rules and thus effectively decide whether or not, and if yes then how, to respond to the claims of formally irregular subjects need to carefully weigh the meaning of this irregularity against, for example, the imperatives that come with their profession or a particular human rights norm. The next section will theorise these negotiation processes in more detail.

2.3 Public Sector Organisations and Street-Level Bureaucrats as Local Mediators of Competing Functional Imperatives and Institutional Logics

One influential strand of literature trying to explain the discrepancy between officially declared government objectives and the actual outcomes of the policies they underpin focuses on the intermediary role of (liberal) institutions (Joppke, 1998; Guiraudon, 2003). For Boswell (2007, p. 83) these neo-institutional approaches are based on two crucial assumptions: That institutions “have sufficient independence from the political system and rival administrative agencies” and that “the actors within these institutions operate according to interests and norms that are at variance with those predominating politics or rival agencies”. At the same time, as Ingram and Schneider (2005, p. 19) remind us, it is precisely “[t]hrough institutions, [that] the social constructions of target groups become semipermanent dispositions that are rarely questioned”. All three assumptions are highly relevant for understanding the complex role that public welfare systems as well as individual actors working within them routinely play when tasked with implementing state policies against irregular migration. In order to explore this role in more detail and be able to draw systematic comparisons across different local contexts and organisational fields, I draw on theoretical concepts and empirical insights from organisation studies.

I thereby start from Nils Brunsson’s (1993, p. 489) interpretation of the possible relationships “between the ideas of constituencies and leaders on the one hand and organizational, and societal actions on the other”: Whereas most understandings of rational decision-making assume that ideas always precede and control action, he argues that this does not necessarily have to be the case; particularly where it would lead to unresolvable conflicts at the level of policy implementation. Instead, certain necessary actions can either determine ideas or be systematically inconsistent with them. Both, I will argue, is likely to be the case where irregular migrants are to be granted some form of access to public services in spite of their unlawful presence. Brunsson’s (1989, 1993) theory thus provides a good starting point for understanding how (and why) organisations respond to contradictory external demands and pressures by accepting and internalising certain inconsistencies between what is officially declared (‘talk’), what is put into law (‘decisions’) and what is effectively done (‘actions’). While it is relatively easy for politicians to declare that foreigners without permission to stay should be unable to benefit from the provision of publicly funded services, the idea of fully excluding them – even if popular among the public – would create significant conflicts if it was to completely control organisational action within, for example, the healthcare system.

It is for such instances that Brunsson (1993) proposes two alternative theoretical relationships between ideas and actions, which he calls ‘justification’ and ‘hypocrisy’. The former means that “planned or accomplished actions are defended in order to convince people that they are the right ones” (Brunsson, 1993, p. 500). If successful, it thus adjusts the constituency’s ideas to actions, thereby restoring consistency at the expense of control (of ideas over action). For example, people may be convinced that the necessity to provide healthcare even in certain non-emergency cases can prevail over the need to limit unwanted immigration or to encourage the voluntary departure of unlawful residents. Where decision-makers find it impossible to openly justify the formal inclusion of irregular migrants, however, they have to resort to ‘hypocrisy’; that is, accepting inconsistencies between what is said, decided, and effectively done:

Actions that are difficult to justify can be compensated for by talk in the opposite direction. Decisions, too, can be part of hypocrisy; they can be contrary to actions, compensating for action rather than controlling or justifying it. Through hypocrisy, the ideas of the constituency are isolated from action (Brunsson, 1993, p. 501).

What according to Brunsson (1989, p. 38) theoretically links ‘talk’ and ‘action’ are ‘decisions’, which “are fundamental to organisations in which politics play an important part”. When it comes to the provision of public services in general and its extension to irregular migrants in particular, somebody has to decide under which circumstances to offer, deny, or require payment for any particular service. These are inherently political decisions and should thus ideally be taken by democratically legitimated decision makers, who then enact more or less explicit laws and regulations. As Mark Hall and Jacob Perrin (2015, p. 132) argued for the area of healthcare, however, “drawing administrable lines that define the limits of a shared humanitarian ethic can prove difficult”. For example, the legal framework for the provision of public healthcare has to leave enough room for individual doctors to fulfil their professional duties, such as those demanded by the Hippocratic oath. In everyday practice, such decisions therefore often also depend on a case-by-case assessment by the individual welfare workers that either administer or provide a service to the population.

These ‘street-level bureaucrats’ are not just implementing the law but effectively become policy-makers themselves, as famously argued by Lipsky (1980, 1987). According to him, it is their particular position within certain organisations – characterised by “relatively high degrees of discretion and relative autonomy from organizational authority” – that “regularly permits them to make policy with respect to significant aspects of their interactions with citizens” (Lipsky, 1987, p. 121). These micro-level decisions can have significant impacts on individual lives and futures (of pupils, patients, benefit claimants, and so on) and are at the same time difficult to control by state or other authorities. According to Robert Goodin’s (1986, p. 223) influential analysis, “some forms of discretion are logically ineliminable from any system of rules”. The individual discretion exercised by street-level bureaucrats in their everyday work (as teachers, doctors or social workers, for example) is necessary because the issues they deal with tend to be “too complicated to reduce to programmatic formats” and “often require responses to the human dimensions of situations” (Lipsky, 1987, p. 122). Importantly, discretion is precisely what allows them to deal with the various irregularities that more or less routinely arise in their daily encounters with service users and often demand customised solutions. Whether perceived as a source of (professional) freedom and autonomy or a practical requirement for the effective (administrative) processing of cases, discretion forms part of these workers’ professional identity. As such, it tends to be defended against limitation by a government or supervisor. In situations where it is perceived as a burden, however, discretion can also be strategically denied in order to limit one’s own responsibility, as Lipsky’s (1980, p. 149) ground-breaking study has shown:

Workers seek to deny that they have influence, are free to make decisions, or offer service alternatives. Strict adherence to rules, and refusals to make exceptions when exceptions might be made, provide workers with defenses against the possibility that they might be able to act more as clients would wish.

In order to operationalise this multifaceted concept, a basic distinction can be drawn between ‘formal’ and ‘informal’ discretion, as suggested by Goodin (1986) and later Jordan et al. (2003, p. 214), who describe the former as practices that “are foreseen or at least allowed by the law, administrative provisions or internal service rules because of the incompleteness or flexible nature of policy design”; and the latter as those that “are developed through daily routines and may run against the formal, legal provisions”. This distinction becomes blurred, however, where the available resources (usually in terms of time and money) are limited and/or the formal rules for their utilisation so vague, complex, or even contradictory that “they can only be enforced or invoked selectively” (Lipsky, 1987, p. 121/2). Such instances of ‘selective enforcement’ can fall within or beyond the legal boundaries of legitimate discretionary power attached to a particular role; but they are often simply unavoidable given the practical constraints of the working environment. At least conceptually, they thus have to be distinguished from ‘deliberate non-compliance’ with certain rules and regulations by street-level bureaucrats, whether as individuals or collectively. The latter tends to be the case where they either do not share the underlying aims or preferences held by superiors or the government, or perceive the rules themselves as contrary to their professional or organisational role (Lipsky, 1987). A particularly strong professional status, such as that of a doctor, and the lack or inefficiency of sanctioning mechanisms makes non-compliance more likely. Together, these concepts describe circumstances in which street-level bureaucrats exercise some form of political agency, whether by contesting or circumventing the implementation of a particular policy or by neglecting or re-interpreting certain aspects of it. This allows them to deal with particular situations as they see appropriate from their perspective within an organisation.

In this sense, the issue of delegating immigration control can also be conceptualised as a ‘principal-agent-problem’, in that it raises the question of how and to what extent the government (as the ‘principle’) is able to enforce and monitor compliance with its rules by the ‘agents’ who are supposed to implement them (Torpey, 1998; Lahav & Guiraudon, 2006). The ability of those who are governed to still choose from a variety of possible actions is also what according to Foucault (2002a, p. 340) differentiates ‘relationships of power’ from ‘relationships of violence’:

A power relationship, on the other hand, can only be articulated on the basis of two elements that are indispensable if it is really to be a power relationship: that “the other” (the one over whom power is exercised) is recognised and maintained to the very end as a subject who acts; and that, faced with a relationship of power, a whole field of responses, reactions, results, and possible inventions may open up.

Both of these elements characterise the ambivalent relationship of street-level bureaucrats to the government (who ‘employs’ them) and the population (who they help to ‘control’): As bureaucrats they have to adhere to a set of official rules, follow formal procedures and apply established criteria, all of which circumscribe their possible actions towards their clients; as professionals they are “expected to exercise discretionary judgement in their field [of expertise]” and to be able to deal with a broad range of individual cases and human circumstances (Lipsky, 1987, p. 121). As I will show, the balance between both aspects of their job depends on their position within the organisation as well as that of the organisation vis-à-vis ‘the state’, but also reflects whether their specific role mainly involves administrative or professionalised tasks (van der Leun, 2003, 2006).

The various ‘roles’ within an organisation can broadly be defined as “conceptions of appropriate goals and activities for particular individuals or specified social positions” (Scott, 2001, p. 55). In modern bureaucracies these organisational roles are separated from the person that performs them, which “has resulted in a capacity to constitute agency and identity in more segmented and piecemeal ways, according to the demands of distinct institutional realms” (Webb, 2006, p. 34). The precise way in which individual members of an organisation fulfil their ascribed role has also been shown to depend on the ‘culture’ of – or within – that particular organisation (Sinclair, 1993; Davies et al., 2000; Schein, 2016). According to Amanda Sinclair (1993, p. 64), organisational culture basically “consists of what people believe about how things work in their organisations” and can explain why people sometimes “behave in ways that are not necessarily consistent with individual or pre-existing norms, but apparently induced by organisational membership”. Rather than characterising a whole organisational field (such as healthcare or social work), such a culture is thus proper to only one organisation (like a specific hospital or school). It has also been shown, however, that “different occupational or professional groups” can also develop their own ‘sub-cultures’, which are thus often “associated with different levels of power and influence within the organisation” (Davies et al., 2000, p. 113).

Another important concept and theoretical approach that gained considerable traction in the field of organisation studies is that of ‘(multiple) institutional logics’, within which individual actors and their actions are also embedded (Meyer & Rowan, 1977; Scott, 2001; Reay & Hinings, 2009; Besharov & Smith, 2013; Lindberg, 2014). They have been described as providing “a coherent set of organizing principles for a particular realm of social life” (Besharov & Smith, 2013, p. 366) and defined as “the belief systems and related practices that predominate in an organizational field” (Scott, 2001, p. 139), such as healthcare or social work. They are similar to what Bowen et al. (2013, p. 3) describe as “repertoires of ‘practical schemas’ for action”, and as such “are not reducible to a national model or ideology” but proper to certain institutional settings. While in principle organisational action within any such field is organised by only one institutional logic that is dominant at any particular time, several other logics constantly tend to coexist and compete with, and sometimes replace, the dominant one as the guiding principle – a process that also helps to explain institutional change (Scott et al., 2000; Lindberg, 2014).

Marya Besharov and Wendy Smith (2013, p. 365) argued that the concrete “implications of logic multiplicity depend on how logics are instantiated within organizations” and, more precisely, on what they call the ‘compatibility’ of a competing logic with the dominant one as well as its ‘centrality’ to the functioning of the organisation. What is crucial to my analysis is that organisations can actively reduce the risk of competing logics generating internal conflicts through structural adjustments that either make compliance with a new set of (conflicting) rules more likely, or non-compliance less visible. According to Besharov and Smith (2013, p. 376), this can be achieved by “[a]ltering the degree of logic compatibility or centrality – for example, by developing a cadre of organizational members who are less strongly attached to particular logics or by buffering members from the influence of those logics” (emphasis added). In contrast to this, Trish Reay and Christopher Hinings (2009, p. 645) posit that “actors guided by different logics my manage the rivalry by forming collaborations that maintain independence but support the accomplishment of mutual goals” (emphasis added).

On one hand, both of these accounts recognise the idea that in order to have an actual effect on organisational practice, institutional logics have to be ‘enacted’ or ‘performed into being’ by individual actors working within the organisation (Lindberg, 2014). On the other hand, they reflect one of the central premises of neo-institutionalism, which posits that organisations constantly strive for legitimacy and – in order to be seen as legitimate by their environment – need to effectively fulfil their ascribed function for society (Meyer & Rowan, 1977; Scott, 2001). Some structural elements are thereby incorporated because of their resonance with certain ‘institutionalised myths’ that reflect what their environment sees as proper functioning and successful performance, even if in practice they do not help or even hinder the efficient realisation of the organisation’s own specific goals (Meyer & Rowan, 1977). This often requires their formal structure to be ‘decoupled’ from organisational action, for example by delegating central activities to professionals: “decoupling enables organizations to maintain standardized, legitimating, formal structures while their activities vary in response to practical considerations” (Meyer & Rowan, 1977, p. 357).

My own empirical analysis supports the argument that from the perspective of public service provision, migrant ir/regularity represents such an ‘institutionalised myth’.Footnote 8 Based on this myth, it is often formally decided that access to public services must be contingent on legal residence status in order to not undermine the sovereignty of the state, the efficiency of its immigration regime or the sustainability of the welfare system. Almost unavoidably, some members of the organisations providing these services will thereby become responsible for exercising some form of immigration control, and thus to ‘enact’ a new institutional logic within these organisations. While probably seen as legitimate or even necessary by a majority of the population (and thus the organisations’ ‘regular’ clients), this may for various reasons contradict service providers’ individual interests, professional ethics or the particular institutional logic or organisational culture that dominates their field or place of work. Through their routine face-to-face interactions with their clients, and given the discretionary nature of their jobs, these micro-level actors often “develop private conceptions of the agency’s objectives” (Lipsky, 1980, p. 144).

At the macro-level, a certain ‘hypocrisy’ in what politicians say and decide not only increases the scope for individual discretion but thereby also makes these inconsistencies less visible to the public: “If decisions are ambiguous it is easier to interpret them as consistent with ideas, both when the decision is made and when the action is completed” (Brunsson, 1993, p. 499). The underlying (political) conflicts are thereby not solved but delegated to the implementing agency, where they have to be managed through “the actions of micro-level actors […] developing localized structures and systems that [enable] day-to-day work”, as Reay and Hinings (2009, p. 630) have shown. Only in cases where “the rivalry between competing logics is resolved through collaboration at micro levels, macro-level actors will develop field-level structures to support the coexistence of multiple logics” (Reay & Hinings, 2009, p. 647). As my empirical data and analysis will show, such reconciliation can have inclusionary as well as exclusionary effects on irregular migrants’ access to public services. In order to allow a systematic examination of these processes, the final section of this chapter incorporates the concepts and arguments established so far into a simple analytical framework.

2.4 A Framework for Systematic Analysis of the Micro-management of Irregular Migration

Immigration status is sometimes described as a ‘master status’ (Gonzales, 2015); that is, a status that overshadows all other aspects of a person’s identity. This would mean that independent from the social or institutional context and of whether a person is sick or healthy, old or young, rich or poor, a criminal or a ‘good boy’, s/he will always be defined by her immigration status and be treated accordingly. This might be true for direct encounters with the immigration system as well as other situations in which (irregular) migrants directly face ‘the state’. A much more nuanced picture emerges, however, when we look more closely and from a comparative perspective at their various encounters within other spheres or subsystems of society.Footnote 9 According to Luhmann (1995) these subsystems have to a large degree become ‘self-referential’ as a consequence of (and requirement for) their functional differentiation and specialisation. This allows them to “tolerate indifference toward everything except very specific features of their respective environments” (Luhmann, 1982b, p. 237). For example, the educational system accepts pupils based on their age (in compliance with mandatory school attendance rules), doctors treat patients according to how serious their illness (as demanded by the Hippocratic oath), and social services assess cases according to the urgency of social needs or the degree of destitution.

The framework I present here will help to understand what exactly happens – both at the level of organisational fields and that of individual workers assuming different roles within these – where the logic of immigration control interferes with otherwise dominant institutional logics. This brings me back to Park’s (2013) essential question of what we (as ordinary citizens) “should do when we encounter an ‘unlawful’ person”. I will look for answers by embedding the question in more specific social contexts and by adopting the perspective of street-level bureaucrats, who are not only citizens but also ‘citizen-managers’. For Park, it is primarily an issue of whether or not we should report the ‘unlawful person’ we have encountered to the relevant state authorities (‘Should I tell?’). Given that migrant irregularity is a largely invisible marker, however, the question we will face before that is whether and how to actually find out, and thus even come to know, the immigration status of that person (‘Should I know?’).

These two questions constitute the basis and simple way to operationalise my two-dimensional framework, which situates individual actors according to whether or not they might, should, or even have to, detect and/or report migrant irregularity in their everyday dealings with other people. Figure 2.1 illustrates this framework: The horizontal dimension of the diagram reflects whether or not individual actors should know potential service users’ immigration status in order to take it into account in their interactions with them. The vertical dimension relates to whether or not welfare workers should tell, i.e. share any such knowledge with immigration authorities. While both questions can simply be answered with yes or no, these answers can also be refined along an ordinal scale that ranges from deputisation (‘Yes, I have to know/tell’), via responsibilisation (‘Yes, I should know/tell’) and autonomisation (‘No, but I can know/tell if I want to’) to shielding (‘No, I must not know/tell’). What marks the difference between autonomisation and shielding is thus the lack or existence of an effective firewall.

Fig. 2.1
figure 1

Potential positions of individual actors or organisational roles in relation to migrant irregularity and its control

Full size image

The combination of the two dimensions results in a field of possibilities that can be divided into four sectors (A–D), each of which has four sub-sectors: Sector ‘A’ represents the positions of actors who are obliged or at least encouraged to both know and tell. Its four sub-sectors reflect whether both, only one, or none of the two tasks is a matter of (legal) obligation and thus an instance of ‘deputisation’. All actors placed in the opposite sector ‘D’, on the contrary, are not obliged nor explicitly encouraged to either know or tell. Here, the sub-sectors indicate whether or not they are thereby effectively shielded by, for example, an explicit ‘don’t ask’ and/or ‘don’t tell’ policy. Sector ‘C’ encompasses roles and positions of people who are obliged or at least encouraged to systematically check immigration status but who are not expected or even explicitly forbidden to share any such information with immigration authorities. Arguably the least obvious positions are those in sector ‘B’, where actors are not obliged nor encouraged to check, but either legally required or encouraged to report any irregularity they nonetheless encounter or suspect.Footnote 10

In Foucauldian terms, already the knowledge of someone’s irregularity is likely to increase the disciplinary power that street-level bureaucrats routinely exercise over their clients. Such knowledge therefore modifies existing power relations, like that between doctor and patient, teacher and pupil (or parent), social worker and benefit claimant, administrator and applicant, and so on. Since it rests on the sheer possibility of being reported, this power operates even in the absence of any formal obligation, explicit encouragement, or moral expectation to report. Its concrete force can thus depend on the real or perceived likelihood that being reported would actually lead to detention or deportation, and thus on the individual as well as contextual circumstances that render a person more or less deportable. In addition, more power can be exercised over migrants who would potentially be deported to a country where they fear for their life or livelihood. At the same time, only by receiving information on somebody who is deportable is ‘the state’ enabled to exert its sovereign power to exclude, which in addition to other constraints is thus always contingent on having this kind of knowledge. This also means that by making use of the varying degrees of discretionary power attached to their roles, street-level bureaucrats can sometimes actively contest and resist the power of ‘the state’. In instances where irregular migrants should (or could) be excluded, offering a service and thus not excluding them can be seen as a form of resistance, which for Foucault (2002a) is always endemic to power relations.

Importantly, the two basic questions underlying my framework can not only be answered for individual actors or particular roles within organisations but can also be transposed to the level of organisations (such as schools or hospitals) as well as organisational fields (like the healthcare or educational system). As Fig. 2.2 illustrates, these can be placed in essentially the same diagram, according to (i) whether or not access formally depends on immigration status, and (ii) whether or not a structural firewall separates them from immigration authorities. The difference to the individual level is that while organisations can be subject to either deputisation or shielding, this is not the case for responsibilisation and autonomisation, which happens at the level of individuals.

Fig. 2.2
figure 2

Potential institutional arrangements for the inclusion or exclusion of irregular residents and likely outcomes for the latter

Full size image

Here again, the combination of both questions results in four sectors, each of which can be linked to a certain outcome for irregular migrants trying to access any particular service: They are excluded where access hinges on legal residence (‘A’ and ‘C’), but only where there is no firewall in place will even the attempt to access trigger immigration enforcement (‘A’). Where access is formally granted irrespective of immigration status (‘B’ and ‘D’), the lack of a firewall implies a tangible risk that still acts as a deterrent and thus leads to informal exclusion (‘B’), while the existence of a firewall permits the formal inclusion of irregular migrants, which I conceptualise as micro-regularisation (‘D’).

The various positions within the framework also have important implications for the overall effectiveness of internal immigration control as well as the “routine interactions among the institutional personnel and its ‘publics’ through which constraints, core beliefs, and role assignment are constantly negotiated, rearranged, and reinvented” (Bowen et al., 2013, p. 13/4). Apart from mapping the various positions that individual actors, organisations or systems that deal with irregular migrants are assigned through formal and informal rules and regulations, this framework also helps to register and compare the underlying motivations and tensions between these. On one hand, I am interested in how individuals or organisations are being incentivised or pushed to detect and/or report migrant irregularity. By looking for individual interests and institutional pressures or logics that tend to converge with the logic of (internal) immigration control, my analysis highlights the various forms of governmentality through which the government encourages compliance with its rule. On the other hand, the framework allows identifying different instances and forms of resistance (by individual actors, professional groups or organisations) against having to know or tell. These will be related to institutionally embedded interests or logics that are at odds with immigration control. Such resistance can be performed at various organisational levels, through either formal or informal discretion but also deliberate non-compliance with explicit rules.

By facilitating a systematic and comparative analysis of these issues, my framework contributes to a better understanding of how immigration control works within society and what that means for some of the core institutions of the welfare state. For Luhmann (1982a, p. 237), not only ‘system boundaries’ but also territorial borders fulfil a crucial role for the increasing differentiation of modern societies because they too function as a “means of production of relations” (cit. in Rigo, 2011, p. 207). Foucault’s analysis of the ‘microphysics of power’, on the other hand, reflects his interest in “showing that power ‘comes from below’, that is, that global and hierarchical structures of domination within a society depend on and operate through more local, low-level, ‘capillary’ circuits of power relationships” (Gordon, 2002, p. xxiv/v). Based on the work of Foucault, Hannah Jones (2013) described the way in which street-level bureaucrats are implicated in these power relations as ‘the conduct of conduct of conduct’, whereby the government acts on the actions of others who themselves act towards others. Seen from this perspective, states do not directly regulate the quantity of migrant irregularity as such, nor the various effects it has on (irregular) migrants’ rights, opportunities and ability to make claims. Yet, by defining and constraining the actions that street-level bureaucrats as well as other citizens may legitimately take towards them, the government provides the framework for, and thereby exercises some control over what I call the micro-management of irregular migration. The following chapter provides a brief but necessary overview of the research design and methodology I employed to collect and analyse the empirical data.