Municipal Activism on Irregular Migrants: The Framing of Inclusive Approaches at the Local Level

Abstract This article explores the ways in which city policymakers in Europe frame the reasons for provision of welfare services to migrants with irregular status. In the context of restrictive national legal and policy frameworks, the article explores a tendency toward municipal activism in extending access to services, providing a typology of six policy frames used by policy makers that identify the intended beneficiaries and stated policy aims. Drawing on interviews with local policymakers and documentary sources, the article contributes to our understanding of the reasons why some European cities provide services to this particular section of the migrant population.

This article explores the ways in which local policymakers frame their reasons for inclusive approaches toward migrants with irregular status in Europe. While there is now a significant body of literature focusing on local policies relating to migrants with regular status, there is growing interest in the less visible policies of municipalities in relation to their irregular migrant population. Set in the context of largely restrictive national legal and policy frameworks governing access to welfare services for this population, the article provides evidence on the ways in which policymakers justify the provision of services that goes beyond the minimum level required by national law. We use the term municipal activism here to refer to actions that facilitate access to services for irregular migrants that are taken in spite of, and to a degree mitigating, restrictive national legal and policy frameworks. The article is informed by and contributes to the emerging literature on irregular migrants' access to welfare services and on the framing by local policymakers of their reasoning for service provision to this group of residents. Drawing on fieldwork and documentary evidence, the study contributes to our understanding of the reasons for the formal inclusion (that is, as a matter of policy) of this particular section of the migrant population.
By "migrants with irregular status" or "irregular migrants" we refer to non-EU nationals who have entered Europe without permission or who have breached their conditions of entry or stay. We use "irregular" in preference to the alternative term, used synonymously, "undocumented," as the migrants may not be without any documentation and because irregular more clearly embraces a range of differing immigration status. While data regarding migrants with irregular status is sparse and the most authoritative estimate for the European Union is now outdated (Clandestino, 2009), they are known to be concentrated in urban areas (Gebhardt, 2010). Following the recent "refugee crisis," the European Commission estimates that more than a million asylum seekers will have their applications refused; yet the return rate in the EU is well below 50% (European Commission, 2017, p. 1). Those who remain in Europe, including many who will never be removed, will swell the numbers of those already living in urban areas with precarious status and limited access to welfare support.
While immigration control is largely a national competence, welfare policies are a shared responsibility in which substate tiers, to varying degrees across Europe, have a level of autonomy in how they interpret and administer national policy (Penninx et al., 2014;European Commission, 2015). Whereas municipalities are expected by national governments to contribute to the integration of regular migrants, in part through provision of welfare services, that is not the expectation of governments for municipal policy toward migrants with irregular status. On the contrary, national legal frameworks governing access to services for those migrants are highly restrictive. Mapping of entitlements by the EU Fundamental Rights Agency (FRA, 2011) and subsequently Spencer and Hughes (2015) nevertheless reveals a variable geography across the EU with a level of inclusion permitted, notably to health care and education. While those exceptions provide for a minimal level of service provision by municipalities, such as schooling for children for which specific entitlements are granted in 10 EU states and an implicit entitlement in a further 13 (Spencer & Hughes, 2015), the policy norm of exclusion presents a challenge to municipalities which bear responsibility for the welfare of local residents and for service provision to facilitate it.
Municipalities have responded to that challenge in different ways: in some cases showing no inclination to be inclusive (Ambrosini, 2013) but elsewhere making provision that is more permissive than national policy requires. Services that are variously provided to irregular migrants by some cities include night and day shelters, food banks, health care, provision of information, legal advice and representation, preschool and school education, apprenticeships, language tuition and skills training, outreach services to street prostitutes, safe reporting of crime for victims and witnesses, subsistence support, and assistance in reconnection with the migrant's country of origin (Delvino, 2017).
In this article, we explore the reasons given by local policymakers for this provision. We look first at insights from the research literature on the relationship between immigration control and access to welfare, on the reasons for the local turn in policies relating to regular migrants, and in particular on policy frame analysis, before setting out the methodology of our study and our findings. In the analysis and conclusion, we consider what those findings add to the understanding of the inclusion of irregular migrants in municipal service provision and the questions this raises for future research.
Understanding policymaking on access to services for irregular migrants A significant theme in migration literature has viewed the selective exclusion of migrants from the welfare state by national governments as a form of immigration control. The aim has been to ensure that access to welfare does not provide an incentive for unwelcome arrivals and to remove an incentive to remain (Bommes & Geddes, 2000). A further aim, through the minimal access to services provision permitted, is to provide a means of regulating stay pending removal. Migrants are required to meet certain conditions to retain their right to remain and to access a minimal level of support. In this analysis, immigration control and selective access to welfare are not in tension but together serve the aim of effective management of migration, supplementing controls exercised at the border.
Scholars have argued, however, that national governments may provide access to welfare in ways that are not to strengthen immigration control but to serve another purpose. In this analysis, access to welfare may limit the efficacy of immigration control rather than enhance it. Access is provided, nevertheless, because of conflicting policy priorities that override that of enforcement. This is not a reference to the discretion exercised informally by street-level bureaucrats, granting access to services in ways that undermine the rules (Van der Leun, 2006). Rather, access is permitted by law or policy: formal exclusion and formal inclusion authorized at the same time (Chauvin & Garc es-Mascareñas, 2012). An example is the extension of access to health care for irregular migrants in Sweden in 2013. It followed an official report setting out reasons relating to public health and standards of administration that necessitated reform despite opposition from those responsible for immigration control (Social Affairs Department, 2011). This conflict between two areas of policy can be seen as an instance of tensions evident in other areas of migration policy wherein the pursuit of a policy objective is constrained by competing policy aims that limit the space for pursuing the original objective (Ruhs, 2013, p. 33). Just as those competing objectives can be played out in disagreements between government departments, we can see them reflected in tensions between national and substate tiers of government for which policy competencies overlap but territorial interests diverge (Hepburn and Zapata-Barrero, 2014).
Where national legal frameworks provide a minimal level of access to welfare for irregular migrants, local municipalities may do no more than facilitate the access required. Where they go further than national law requires or allows tensions in the vertical multilevel governance of this issue (Spencer, 2017) may arise. The reasons why municipalities take this approach, given the constraints and the risks for their relationships with national governments, is as yet little understood.
The "local turn" Migration scholars have noted the growing activism of municipalities in relation to the integration of regular migrants. There has been "a sharp local turn" in policymaking in which municipalities and cities in particular "are becoming increasingly entrepreneurial in developing their own integration philosophies and policies. This has led to cities having markedly different approaches to migrant integration, even within the same countries" (Scholten & Penninx, 2016, p. 91). The local turn is attributed to municipalities being faced with challenges that differ from those faced at the national level. While some argue that local policies reflect a pragmatic need to address particular socioeconomic problems (Popperlaars and Scholten, 2008;Scholten, 2013), others emphasize the way in which policymakers' sense of their city and the place of migrants within it, their "urban imaginary," leads to differing interpretations of what is "pragmatic." In distinct, city-specific accounts, their meta-narratives connect an interpretation of the city's past with its current situation (Hoekstra, 2018, p. 3).
Whether the political affiliation of elected representatives is to the center-right or center-left of the political spectrum has been found to be a factor in exclusionary or inclusionary approaches in relation to regular migrants (De Graauw & Vermeulen, 2016). Bureaucrats, however, can be more responsive to the needs of immigrants than elected representatives are, if immigrants do not have the right to vote. Their sense of professional identity and corporate or personal values (Schiller, 2016) are contributory factors in municipal responses. This has been found to be particularly the case in relation to irregular migrants for those in service-providing jobs, including police officers (Leerkes, Varsanyi, & Engbersen, 2012). There is clearly a continuum here between the exercise of discretion by bureaucrats that diverges from the official policy of the municipality and instances in which they successfully shape official policy to reflect their judgment concerning what needs to be done. In some cases, the distinct approach that municipalities take is nevertheless in line with national policies, but in others it diverges, leading in some cases to a de-coupling of the multilevel governance relationship (Caponio & Borkert, 2010;Scholten & Penninx, 2016, pp. 98-101).

Policy framing
In our study, the source of evidence on the reasons for inclusive approaches available to us was the reasons cited by policymakers in interviews and in policy documents. This is not evidence on the factors that are driving policy per se but on the framing of those reasons by policy makers. The literature on policy framing reminds us that there is no single interpretation of a situation-no one version of the facts. Problem definitions are not given but framed (Van Hulst & Yanow, 2016, p. 105). The process of framing is selective, highlighting aspects of a situation while disregarding others, creating a coherent narrative that defines and explains a problem in a particular way and from that identifying what needs to change. Framing "lays the conceptual groundwork" for future courses of action (Van Hulst & Yanow, 2016, p. 99). In policy discourse, rhetorical framing may be seen in the persuasive narratives used to win an argument, while action framing, which may differ on the same issue, more directly shapes the actual policy devised. The framing process defines, categorizes, and makes sense of the issue in ways that can differ from how others see it. Where policy actors (and indeed tiers of government) differ not only in their framing of solutions but in their problem definition, discussion of policy options alone may be insufficient to reconcile differences. Rather, negotiation that includes "frame reflection" can be needed (Sch€ on and Rein 1994;Spencer, 2017).
In an analysis of city policies facilitating access to legal status for irregular migrants, Caponio (2014, pp. 5-8) identifies four policy frames that shape their approaches based on who is the intended beneficiary, the aim of the policy, and the initiative that is promoted. Where cities adopt a security frame the beneficiaries are intended to be national residents; the aim is to reassure them that there exists respect for law in the city and that there are effective immigration controls. The context is a negative discourse on migration and the policy measures exclude irregular migrants from local society. In contrast, in the humanitarian frame, the prime beneficiaries are migrants. The framing focuses on their vulnerability, their welfare needs, and their deservingness of municipal support. A variant of the humanitarian frame is a human rights frame in which particularly vulnerable irregular migrants are identified as in need not simply of welfare services but of protection of their fundamental rights. The problem is here framed not as a welfare issue but as a deprivation of human rights and the solution lies in granting legal rights through securing a regular immigration status.
Finally, in a deserving workers frame, irregular migrants are seen as workers who are needed in the local economy but who have not been able to secure or retain their right to work. Here the migrant is not defined as vulnerable but as deserving of legal status because of his or her contribution to the labor market (chiming with a broader "civic performance" frame identified in analysis of national regularization schemes; Chauvin, Garc es-Mascareñas, & Kraler, 2013). Here the municipal responsibility is to facilitate access to the right to stay and to work. While the migrant is a beneficiary here, the ultimate beneficiary is the city's population, through a thriving labor market (Caponio, 2014, p. 16).
Notwithstanding that element of economic utility, it is striking in these frames that the beneficiaries of the proactive policies are largely identified as the migrants themselves. Only in the security frame, in which cities are seeking to exclude irregular migrants, and to an extent in the framing of "deserving workers," is the beneficiary intended to be other city residents. This analysis related, however, to one area of city policymaking-access to legal status. Our findings enable us to explore whether a wider range of policies toward irregular migrants reflects the same policy frames or defines the problem and the beneficiaries of the policy in other ways.

Method
As our sources of evidence on the factors that have led municipalities to provide access to services we rely on the reasons cited by local policymakers from three sources: a series of interviews conducted between 2012 and 2015 in western European cities 1 ; two roundtable discussions with representatives from European cities in 2014 and 2017; and policy documentation. Interviews were secured by snowballing via policymakers with whom the authors had contact and through academic, city, and civil society networks. The study, which included a mapping of national legal frameworks reported elsewhere (Spencer & Hughes, 2015), focused on 14 countries selected according to whether irregular entry or stay is a criminal offense 2 (in nine states but not in five); the relative size of the estimated irregular migrant population at the time of the most recent available estimate (varying from 0.1 percent in Finland to 1.7 percent in Greece); and for differing levels of autonomy of local tiers of government. 3 Among a total of 61 interviews with 79 policymakers, 35 were with city or (in a few cases) regional policymakers (including four politicians and one political advisor). Among officials (local civil servants), interviewees ranged from those responsible at a senior level for migrant-related policy through those managing a relevant service to, in a few cases, those delivering the service. Interviews followed a semistructured set of questions which explored the nature of local provision, the rationale for it, and the decision-making process. Interviewees were offered anonymity, including for the name of their city. Where available, related documentation was reviewed. Responses were analyzed using NVivo, a software facilitating inductive thematic analysis of qualitative data. Themes arising in responses to questions that explored reasons for a city's overall approach to service provision to irregular migrants and to provision of a particular service were coded; the range of codes expanded as new themes emerged from the data.
In this article we refer directly to evidence from 14 cities from northern to southern Europe: in Finland, Sweden, Denmark, Belgium, Germany, Ireland, the Netherlands, the UK, Italy, Spain, and Greece. The strength of drawing on evidence from this range of countries is in being able to explore the reasons given for access in cities across western Europe, from which we could identify, inductively, the differing frames that they use. The disadvantage is that, while we provide some contextual information, we could not perform a comparative analysis that might have explained why cities take contrasting approaches.
Our second source is two roundtable discussions. The first took place in October 2014 in Barcelona between representatives from 11 cities (City of Barcelona and COMPAS, 2014); 4 the second, which took place in Utrecht in June 2017, involved representatives from 10 cities. 5 Four of the cities were represented on both occasions. Overall, cities from 13 European countries were represented. The city representatives were responsible for differing aspects of municipal policy with relevance to migrants with irregular status. 6 The discussions were held on a nonattributable basis but presentations at the first event are cited.
Reliance on the views expressed by policymakers has strengths and limitations. The reasons given to explain service provision reflect the understandings of individuals closely involved in shaping or implementing the policy and were nonattributable, encouraging full disclosure. The settings provided an opportunity to explore underlying reasons and, in the roundtable, for representatives to contrast reasons that applied in their own cities. There was acknowledgment, however, that outside of these settings the reasons used to justify provision could differ depending on the audience (a gap between rhetorical and action frames) and between the reasons cited and the range of underlying factors that may have led to the policy. Nevertheless, the reasons cited by policymakers are one source of evidence that allows exploration of this issue.
Where possible we also drew on documentation, in the minority of cases where policy reasoning has been spelled out-as in a Barcelona city council policy paper (Ajuntament de Barcelona, 2017). In most cases however we found that municipalities have opted for low visibility provision (a finding reported elsewhere; Spencer, 2017) and such explicit reasoning is not documented, hence the reliance on this issue on the oral testimony of policymakers themselves.
The cities we report on are, and were at the time of interview, at different stages in their experience of migrants with irregular status. For cities in Scandinavia, the experience of having irregular migrants approaching municipal services and seeking assistance has been relatively recent. While the issues they raise for municipalities are challenging, not least in states in which access to the welfare state is controlled through a personal identification number, the number of people involved is small relative to the numbers in large cities in, for instance, Germany, Belgium, the Netherlands, Spain, and the UK. Those countries have more experience addressing this issue but their national contexts differ. The German Residence Act ( §87), for instance, requires most service providers to pass on details of service users with irregular status to the immigration authorities, limiting the extent to which municipalities can directly provide any service; whereas in Spain the law requires municipalities to allow all residents to register in the local civic register, regardless of immigration status. That provision enables Spanish municipalities to take account of irregular migrants in their planning of local services while still facing constraints on the extent to which they can provide or use public funds to pay for service provision. The Belgian legal framework allows some flexibility in interpretation of what is permitted. In the Netherlands, municipal solidarity and the political strength of the largest cities has allowed some significant challenge to restrictive national policies. In the UK, the political sensitivity of the issue has arguably limited municipal challenge to an area in which they do have a legal mandate but limited resources: provision of support for children "in need" under the Children Act 1989.

Findings: Framing inclusive measures for irregular migrants
In this section we explore the extent to which our evidence is compatible with Caponio's policy frames relating to legal advice to secure regularization-that is, her security, humanitarian, human rights, and deserving workers frames-and propose two additional frames for which our findings, relating to a wider range of services, complement that analysis.
Cities across Europe have implemented a number of measures to ensure access to services, including providing shelter, legal advice, health care, education, documentation (e.g., birth certificates), and initiatives facilitating the reporting of crime by victims and witnesses, and a generalized access to municipal services.
In relation to housing, cities operate within particularly restrictive national legislation, but nevertheless often fund NGOs to operate a shelter service for homeless people (or particularly vulnerable individuals) irrespective of immigration status, or to provide assistance in accessing private rented housing. In other cases, cities have been able to operate shelters directly, relying on provisions in domestic law (or ambiguity in the terms of legal restrictions), or on international law. In some cases they condition access to the shelter to the migrant's cooperation for their return.
Many initiatives focus on the provision of legal advice to address the challenge posed by irregular migrants' presence by resolving their immigration status, either by facilitating regularization or facilitating voluntary return. In this field, municipal initiatives range from providing legal counseling within a shelter in which migrants are hosted to setting up specific municipal information centers or funding NGOs to provide legal advice.
In relation to health care, initiatives address barriers to access including procedural obstacles (e.g., a lack of ID documents), economically nonaccessible medical treatments (in cases of irregular migrants not being included in national insurance schemes), and lack of knowledge of entitlements (through information or counseling). In some cases, municipal practices simply extend the level of access granted by national legislation. In relation to education, municipal activism aims to address practical barriers to the enrollment of children with irregular status in school (e.g., avoiding enrollment procedures requiring a residence permit), supporting the financial needs for enrolling children with irregular status, or enabling access to preschool provision.
Many other initiatives have been identified including foodbanks, language classes, outreach service for street prostitutes, and facilitating the "safe reporting of crime"(to be discussed). Other measures include provision of documentation, such as birth certificates, ID cards, or registration into municipal registers, as in the case of Spain (Delvino, 2017).
Each service can be intended to secure a series of aims and hence falls within different frames. Policymakers may also emphasize different frames for a particular audience for whom that framing has more resonance-focusing on the crime prevention rationale, for instance, when that is most likely to be well received.

Security frame
In our study we were looking at cities for which we had identified inclusive measures and would therefore not expect to find evidence of Caponio's security frame. We nevertheless found instances in which the rationale for provision of a service did indicate a security frame to the extent that the aim is to prevent crime and disorder and the intended beneficiaries are all city residents. As an Amsterdam official put it in relation to the city's approach overall, "The underlying reason why we are involved in these activities with irregular migrants is that we don't want trouble, social disorder" (Interview 2).
In each case when the prevention of crime was mentioned, however, protecting vulnerable irregular migrants from harm was also intended. Inclusion of irregular migrants rather than their exclusion was the objective. In these cases, the security framework differs from that identified by Caponio in that it fosters the inclusion not the exclusion of irregular migrants. In that sense we can talk of an inclusive variant of the security frame.
Crime prevention and detection were thus cited as reasons for provision of shelter and food; as necessary to reduce the danger of irregular migrants engaging in crime but also to protect them from the vulnerability of life on the streets. Protection of residents and of irregular migrants was likewise cited in relation to providing a means of safe-reporting of crime for victims and witnesses, and for measures to protect victims of domestic violence.
Amsterdam's police, then accountable to the city's mayor, developed a "free-in, freeout" approach to enable irregular migrants to report crime to the police without fear of arrest for immigration purposes unless suspected of other offenses. The approach was driven by concern that the police were not accessing information they needed to prevent and detect crimes in an area of the city with a significant migrant population. Prior to the policy, an officer who was engaged in the initiative said that police "intel" position had been poor (Zwart, 2017). The justification for the policy was also, however, that irregular migrants are more vulnerable as victims of crime than those who have a regular permit. The police needed to build trust "so people feel they can go to 'our' police and the police can also give information to them" (Interview 2). Framing of the rationale for this approach includes officers' perception of their appropriate role as police officers, reference being made to the requirement in policing legislation that "the task of the police is to maintain law and order and assist people who need help." A concern to protect the public at large is thus here paired, as in the earlier examples, with a concern for the victims themselves (Zwart, 2017).

Humanitarian frame
There is also evidence in our findings of Caponio's humanitarian frame-a concern for vulnerable individuals. That problem definition in turn frames a municipal responsibility to meet their welfare needs. Concern for the health and well-being of individuals is cited in relation to provision of shelter, health care, and education and for targeted protection for the most vulnerable, particularly children. Within this frame we find frequent reference to the relative deservingness of children who, as a German city official told us, "are not responsible and should not be held responsible for the actions of their parents" (Interview 10). The professional ethics of service providers is also cited; those responsible for health services argue that exclusion from health care on grounds of immigration status conflicts with their medical ethics. According to one provider, "Everyone gets treated, every medical emergency gets treated, because of the medical ethics … the questions "why you are here?" "what is your legal status? don't get asked' (Interview 6). Those responsible for education services similarly argue that exclusion would conflict with teachers' perception of their role and responsibility toward children.

Human rights frame
Policy makers from some cities explain their overall approach toward irregular migrants within a human rights frame, while citing additional reasons related to particular services. Utrecht, self-designated as a "human rights city," contextualizes its approach within that frame, 7 while others refer not to their perception of their city but to a broader national ethos. A Helsinki city official saw the exclusion of irregular migrants from welfare services as contrary to the value Finland places on equality: How to meet the needs of people who are undocumented is more than an administrative question though. It challenges the whole equality principle on which our system is based by accepting that there will be some people living here who have fewer rights than others, and that is pretty difficult to swallow. (Interview 4) For a Berlin city politician, exclusion from essential services was equally incompatible with her perception of her country's national ethos: "We cannot live in a country which keeps very highly its value of the dignity of human beings and then have people who do not even have a birth certificate and who if they are ill cannot get treatment" (Interview 5).

Competing legal duties
In some cases, the reasoning in the humanitarian and the human rights frames is supported by legal duties. The municipal duty of care in the Netherlands is cited in relation to provision of shelter by Dutch cities, while in the UK the duty to safeguard children "in need" is given as the primary reason for provision of accommodation and welfare support to destitute migrant children and their parents, who have "no recourse to public funds" (Price & Spencer, 2015). Municipalities may also, however, use a domestic legal duty, international law or the European Social Charter 8 as a justification for provision for which pragmatic concerns lie elsewhere (City of Utrecht, 2014).

Deserving workers frame
We also find evidence of Caponio's deserving workers frame. An official from the industrial city of Genoa explained his city's inclusive approach in the role that irregular migrants were playing in the city's labor market and the city's "history of working class struggles and of solidarity among workers" (City of Genoa 2014). Some cities' approach is informed by awareness from long experience that many of those living with irregular status in the city will eventually secure legal status. They conclude pragmatically that it is better to facilitate their integration from an early stage. In Barcelona, for example, "Our expectation is that most of these migrants will stay, will behave well and become full citizens, so we invest in them" (Interview 13). In that situation, there can be recognition that vacancies in the local labor market attract irregular workers when legal migration channels are not available (Interview 2; Roundtable 2017).

Socioeconomic frame
Some of the reasons cited by policymakers for providing access to services, however, do not fit within an (inclusive) security, humanitarian, human rights, or deserving workers frame. In contrast to the primary focus of those frames on migrants as the beneficiaries, our findings suggest two additional frames in which the problem definition relates to the impact of migrants' exclusion on the city and city administration, and in which the intended beneficiaries are the residents of the city as a whole. These are a socioeconomic frame and an efficiency frame.
In a socioeconomic framing, the problem is defined as the economic and social harm caused by the exclusion of irregular migrants and the way in which this undermines the municipality's capacity to achieve its economic and social policy goals. Barcelona's action plan on its approach toward irregular migrants cites the primary harm caused by the exclusion of irregular migrants as economic: the effect of irregularity on the informal labor market, notably on domestic work, the hotel industry, and construction (Ajuntament de Barcelona, 2017). Dublin's provision of shelter is in part because the image of the city was being undermined by people living on the street: "Tourist businesses are very sensitive to any negative portrayal of the city" (Interview 3). Similarly, when Barcelona's outreach service for street prostitutes was established, the rationale was the negative impact of their presence on the tourist industry and thus on the city's economy (Interview 16).
The risk of social harms is frequently cited: the ways in which exclusion can threaten public health, social cohesion, and the integration of (regular) migrants and can undermine municipal efforts to tackle homelessness, street prostitution, and domestic violence. Here the ultimate beneficiaries of measures allowing access to health care for irregular migrants are not, as in a humanitarian frame, the migrants themselves. Rather the aim is to mitigate a threat to others. As a Berlin city politician put it, "Health is the most urgent issue as their exclusion from health care other than emergency care potentially endangers others-where they have communicable diseases such as AIDS for instance" (Interview 5). The threat may not be to the public as a whole but to a subsection of it, including migrants with regular status; sharing overcrowded accommodation with those excluded from work and welfare services is seen to undermine municipal efforts to foster their integration (Interview 14). Access to vaccinations is explained in terms of public health; access to language classes, in terms of cohesion. Shelter is necessary not only to avoid homelessness but to avoid the substance abuse to which it can lead (Roundtable 2017).

Efficiency frame
Finally, policymakers cite a range of reasons relating to the efficiency of public administration, including that it is more cost-effective to include this group than to administer procedures for ensuring their exclusion. For Spanish municipalities, registration in the municipal register (the Padr on) is necessary "so that the municipality has information on the people in their area and can plan services accordingly … like the number of school places and doctors" (Interview 13). Issuing of birth certificates is a prerequisite of accurate population data. Providing a means to report crime without fear of deportation means, as we saw, that the police have access to intelligence that improves their efficiency in tackling crime. Child protection, primarily defended on legal and humanitarian grounds, is also framed as avoidance of reputational risk should a child, notwithstanding their immigration status, come to harm because the municipality failed to act (Interview 15).
Exclusion puts pressure on emergency services, impacting their capacity to deliver. In Florence the municipality provided postoperative care facilities for irregular migrants because their lack of entitlement to postemergency care meant hospital staff were prolonging their stay in emergency beds (Delvino, 2017, p. 10). In cases in which irregular migrants rely on emergency services, their staff can look to municipalities to make alternative provision, as for the street homeless in one Danish city: These migrants can go to a hospital for free emergency treatment … And one of the things that happens is that they use the hospital like a hotel, coming back every few days. The hospitals employ social nurses who ring me up and ask me to come and help deal with them (Interview 7).
In Helsinki, the pressure was from those providing day cafeterias for homeless people (Interview 4). In these examples, the framing is not humanitarian concern for the vulnerable (though that may also be present) but a concern for efficient public administration. The definition of the problem is pressure on hard-pressed services; the solution is alternative, more appropriate, provision to relieve that problem.
Caponio identified provision of legal advice on regularizing immigration status under the deserving workers frame. We found that service was framed, in part, in terms of rights protection but also as a service provided to reduce the number of people with irregular status in the city because of the overall challenge that goal poses. Legal counseling may identify grounds for regularization, lead to the individual being supported by the state-funded asylum system, lead to a temporary residence permit for victims of domestic violence, as in Athens (Interview 8), or to the municipality helping the individual to reconnect with his or her country of origin. While migrants are in part the intended beneficiaries, the aim here is also to address the challenge their exclusion poses to the city: [Legal counseling] is provided because "illegal stay is not a solution-it is a problem to be resolved" (Interview 9).
We set out this typology of framing in the table below.

Discussion
What does this evidence tell us of the way municipal policymakers frame the problem definition and solution? What does it add to the broad framing of city responses in earlier work? The reasons given by policymakers for service provision convey a narrative that reveals the problem definition from which their policy solutions derive. Caponio had identified four frames relating to the provision of a particular service, legal advice on regularization, and we saw that with the partial exception of the security and deserving workers frames the beneficiaries identified in the frames were largely the migrants themselves.
In our findings we also see evidence of a humanitarian, human rights, and deserving workers frame. We also see a security frame, but here it is a frame to explain provision of services to irregular migrants, not to their exclusion from services, and the beneficiaries are intended to be migrants and other city residents. A security frame can thus be used to justify provision and to reject it: an inclusive variant of the security frame. Within the humanitarian and human rights framing we saw that the reasoning is backed in some cases by a legal duty to protect the public or sections of it, which municipalities claim overrides restrictive national measures. The legal duty can be in national law or in an international human rights obligation of the state by which the local state is also bound. Municipalities may nevertheless cite legal obligations to support provision for which other reasoning is also at play. In the humanitarian and security frames we also saw most clearly the evidence of service providers' own perceptions of their role shaping the case for inclusion-in the police responsibility to protect, in medical ethics, and in teachers' responsibility toward children's education.
We find that many of the reasons provided for services, moreover, fall outside of those Caponio identified in relation to legal advice. We identify, first, a socioeconomic frame in which the exclusion of irregular migrants is defined as posing an economic or social threat, undermining the municipality's economic and social policy goals. Second, we see an efficiency frame in which exclusion undermines the efficiency of the city administration and its service provision. In each case it is significant that it is not irregular migrants per se who are framed as posing the threat but rather their exclusion. Moreover, while in the humanitarian and human rights frames it is the migrants who are the intended beneficiaries of the services, in the socioeconomic and efficiency frames (in particular) the intended ultimate beneficiaries are the residents of the city as a whole-that is, to meet the needs of the city not, primarily, the needs of the migrants themselves. Yet the narrative of the frames is not solely that of a pragmatic response. We see it shaped by the "urban imaginary" that policymakers have of their city and, more broadly, their sense of their national values; values deemed incompatible by these policymakers with the exclusion of a section of local residents from the services they need.

Conclusion
In this article we address a gap in the literature on municipal policies toward migrants in Europe with irregular immigration status. While some municipalities provide no additional access than restrictive national legal frameworks dictate, others go further, engaging in a municipal activism we set out to explore, drawing on fieldwork and documentary evidence. Our research identified a wide range of services provided, from health care and preschool education to legal advice, shelter, and language tuition. Here one part of the state, the local tier, is providing for a level of formal inclusion of irregular migrants, not merely the informal inclusion facilitated by sympathetic service providers.
While there is now a significant body of literature on the "local turn" in policies relating to regular migrants, we knew less about the rationales for service provision toward those who entered the country without authorization or who have breached the conditions attached to their entry or stay. Our findings address that gap. They show that, as found in relation to regular migrants, the local turn reflects not only a pragmatic need to address socioeconomic challenges that can differ from those at the national level but also the way in which policymakers make sense of their city, the place of migrants within it, and the values of the policymakers themselves.
In relation to the literature on the relationship between welfare provision and immigration control, our evidence from the local level shows that here too provision of welfare can contribute to immigration control, as when the aim is to facilitate resolution of immigration status or reconnection with country of origin or when access to shelter is conditional on compliance with return procedures. We see, however, that welfare provision to irregular migrants may also be for the purpose of serving an alternative policy objective, such as public health or child protection. What we do not know from our data is whether such provision, as the literature suggests, competes with immigration control objectives-that is, whether welfare provision attracts more people to come or remain with irregular status in that particular locale for instance; one of many questions on which further research is required.
Our findings provide new evidence on the way in which, for irregular migrants, local policymakers frame their reasoning for inclusion. We identify a typology of six frames that emerge, classified on the basis of the intended aims, beneficiaries, and stated policy aims. We find that a security framing can include migrants themselves as beneficiaries, an inclusive variant of the security frame Caponio had found in her earlier study. Most significantly we find that what we term a socioeconomic framing and an efficiency framing figure prominently in the reasoning local policymakers provide for inclusion of irregular migrants in service provision; in such framing, in contrast to a humanitarian or human rights framing, the intended beneficiaries are less the migrants themselves than the residents of the city as a whole. The typology provides the basis for future research on the reasoning at the state or substate level for the formal inclusion of irregular migrants in welfare services.
A study that draws on evidence from policymakers across a range of cities, as our study has, provides insight into the ways in which local policymakers define the challenges posed by the exclusion of irregular migrants and the reasons these policymakers have for providing services. Comparative studies are needed to deepen our understanding of the reasons why different cities frame their overall approach, and the rationale for particular services, in different ways; of whether and why particular frames are dominant at different times; of whether rhetorical framing to garner support for service provision differs from the action framing that is shared in a private interview setting; and, turning to outcomes, of whether provision of services has an impact on immigration enforcement. Within a city there are many policy actors and our findings suggest that they can differ in their framing. An in-depth study could explore why. We also need to understand why some cities with an irregular migrant population do not adopt these inclusive approaches and how they frame their rationale. There are limits to municipal activism in this area and just as we need to understand why some cities are proactive, equally we need to understand why others are not. There is a rich research agenda here to explore. Notes 1. The interviews were part of a study carried out under the auspices of an Open Society Fellowship. They were conducted by the authors and/or by Vanessa Hughes or Jonathan Price, whose contribution to the study is gratefully acknowledged. Some interviews were