Keywords

Introduction

Migration policy is a field characterised by huge implementation problems and there is a vast literature on the Gap hypothesis, referring to the systematic divergence in policy. In other words, there is a gap between what governments aim to do and what they actually achieve. In general, implementation failures may have many causes and, putting it briefly, implementation of policy can be more successful if the implementing organisation has enough capacity for its task, knows what it should do, for example, clear and explicit decisions, and it genuinely wants to complete the task.

This field of policy and decision implementation addressed here is something traditionally approached by public administration researchers. It encompasses different traditions, approaches (top-down, bottom-up), theories and schools of thought. In broad terms, implementation research is about why intentions and decisions among governing bodies are not always put into practice and do not always achieve the intended goal (Sannerstedt, 2001, p. 28).

Implementing policy when the targeted or affected individuals are less than willing to adapt and submit to the legislation adds one further dimension to the challenges. This leads us to the fundamentals of return policies in Sweden. In essence, an asylum seeker’s primary goal is to stay in the host country and, ultimately, be granted a permanent residence permit. However, according to the statutes and rules set up in legislation and custom, some seekers are not entitled to either asylum or other grounds for staying in Sweden. Thus, their applications are rejected, and after a legally binding (non-appealable) decision, they are deported from Sweden. It is expected that the rejected asylum seekers are available and cooperative in relation to the migration authorities. However, if they do not return voluntarily, there is the threat of being forcibly denied entry or deported from the country by the police. Not every mean can be used to enforce the policy. Beyond being efficient, the Swedish policymakers and the agencies in charge also clearly state that humanity and legality are two other goals or, perhaps, values to be upheld. Despite efforts by different actors in Sweden, only a minority submit and return voluntarily: roughly 44 percent of all return cases result in voluntary return (Malm Lindberg, 2020).

Thus, states in general—not just Sweden—are facing an uphill task, using several measures attempting to track, identify, and, ultimately, remove rejected asylum seekers and other irregular migrants from its territory. In that struggle, the states have attempted to implement different measures for control, monitoring, and identification as well as economic incentives and different kind of information, but the ultimate threat is to use coercive measures and forcibly remove the irregular migrants (Noll, 1999; Broeders & Engbersen, 2007; Scalettaris & Gubert, 2018, p. 2). Return or deportation policies are often considered a secondary instrument of immigration control, where liberal democratic states increasingly make an effort to make ‘unwanted’ migrants leave their territories through return policy instruments aiming either for forced or voluntary return (e.g. Bucken-Knapp & Zelano, 2023, this volume). Thereby, it complements measures designed to control the entry of immigrants, including visa procedures, border patrols and identity checks at ports of entry (Ellermann, 2005). This ‘deportation turn’ is part of a broader tendency among states to make it harder to immigrate and to control migration (de Haas et al., 2016).

In this chapter, I contribute to the discussion of the so-called Gap hypothesis on the disparity between policy on paper and its implementation through the bureaucracy. Furthermore, grounded in a basic framework for implementation established by Lundquist (1987), I also examine the role of street-level bureaucracy in the implementation of return policy in Sweden and, albeit briefly, touch upon how the Covid-19 pandemic may impact implementation.

Method and Data

The empirical foundation—and the voices of the informants—is based on semi-structured interviews with roughly 40 bureaucrats from different levels, ranging from top- and mid-level to, primarily, front-line or street-level bureaucrats, for instance persons in the field. The bureaucrats are employed by Migrationsverket (the Swedish Migration Agency), Polismyndigheten (the Police), and Justitiedepartementet (the Ministry of Justice), as well as other public agencies and select civil society organisations. Data collection took place in the pre-pandemic years 2018–2019 within the Return and Reintegration project at Delegationen för migrationsstudier (Delmi—the Migration Studies Delegation). The interviews, which lasted for 40–120 minutes, were recorded and subsequently transcribed. The lion’s share of the interviews was conducted individually and face-to-face. They were semi-structured with questions on topics known by the informants, but there was also room for additional questions. Some respondents were mainly asked specific questions about ‘hard facts’ on, for example, the return policy or enforcement procedures in different situations. Other respondents were mainly asked about their judgements and assessments of different issues and what they struggled with in their work; thereby these individuals often gave anecdotes and examples. These interviews were manually coded, based on keywords, and then classified into different themes based on the implementation challenges that exist. When possible, a triangulation of data from diverse interview sources was made, including the substantiation of claims using written resources (Malm Lindberg, 2020).

Earlier Research and Theory

Implementation and Street-Level Bureaucracy

Which factors are traditionally associated with success or failure in implementing policy in general? In the aftermath of the early implementation studies in the 1970s, ‘checklists’ with some prerequisites that should be met in order to achieve success were often drawn up (Löfgren, 2012). These include Hogwood and Gunn (1984), who gave a long list of recommendations for successful implementation: to have no insurmountable external constraints that affect the implementer’s job; to allocate adequate time and sufficient resources; to have the required combination of resources at the right time; to base the political decisions on an accurate causal understanding of goals and means; that dependency relationships are minimal; to specify the correct sequence of tasks; to ensure perfect compliance and communication between those involved; etc. It is safe to say that all these prerequisites will probably never be present simultaneously; hence perfect implementation almost never happens. However, such checklists can still have a clarifying function when there are decent chances of accomplishing the goal.

Van Meter and Van Horn (1975) use organisational theory to show the complex and dynamic relationships that exist between many different, mutually independent, variables. They point out that implementation is successful when the organisation has sufficient capacity, knows what it should do, has clear and explicit decisions, and wants to complete the task. Roughly the same aspects are also elegantly expressed by Lundquist (1987, pp. 76–78) and the concepts used later.

  • Does the implementer understand what is to be implemented?

  • Does the implementer have the ability to implement?

  • Does the implementer have the will to implement?

I assume here that the affected authorities and their staff can be characterised as street-level bureaucracies. This expression, coined by Michael Lipsky (1980), basically means that employees in lower-level posts (grassroots level), who have direct contact with the clients, find themselves in a position between clients and authority. They are essentially responsible for translating governmental policies into concrete action. By interpreting regulations and political decisions, they can thereby take over much of the implementation of the decisions that are made. In close interaction with clients, whether pupils, patients, asylum seekers or irregular migrants, the front-line bureaucrats establish their own routines and working methods to manage their jobs. They can choose to invest more formal and informal resources in their clients and provide them with access to more or fewer resources. It is essential that street-level bureaucrats, at least to some degree, have the possibilities to formulate (their own) interpretations of how the assignment should be carried out, thereby sometimes bending and sometimes also breaking the formal rules (Portillo, 2012; Baviskar & Winter, 2017; Lavee, 2021). Having this bottom-up approach and connecting to the street-level bureaucracy literature, in line with Lipsky (1980), leads in the direction of a deeper investigation of the organisational context of implementation.

This group of front-line bureaucrats must manage resource shortages and difficult prioritisations, which often causes their actions to be questioned (Lipsky, 1980; Löfgren, 2012, p. 9). Earlier Swedish research characterises both Polismyndigheten (Hydén & Lundberg, 2004; Hansson, 2017) and Migrationsverket (Rosén, 2010) as street-level bureaucracies, with the characteristics that such agencies and front-line bureaucrats have. When the term came up in some of the interviews the respondents often felt familiar with it. I choose to proceed based on Lundquist’s (1987) three conditions (understanding, ability, willingness) for how the bureaucracy can successfully implement policy or decisions.

The Gap Hypothesis and Implementation Problems in Migration and Return Policies

Migration policy is often deemed an area characterised by huge implementation problems. Failing to meet migration policy goals is more of a rule than an exception, true across time and space. The literature on the gap hypothesis, formulated early on by Cornelius et al. (1994), points to the gap between what governments aim to do and what they actually achieve. The list of political failures is long and examples can be drawn from numerous Western countries. Thus, the interesting question is not if there are discrepancies between goals and outcomes, but rather how this phenomenon can be measured, understood and explained (Cornelius & Tsuda, 2004, p. 4).

Despite a large literature referring to the gap hypothesis, the concept remains surprisingly vague, referring to different types of gaps within the policy process (Lahav & Guiraudon, 2006; Boswell, 2007; Czaika & de Haas, 2013). There are several reasons for the difficulties associated with migration policy (see also Jerneck, 2023; Lerpold et al., 2023; Bucken-Knapp & Zelano, 2023, all this volume) and the fact that goals are rarely achieved. In addition to the factors associated with the migration process itself, globalisation and the disparity between the affluent North and the impoverished South give rise to more or less unwanted migration flows (see Massey, 2023, this volume). Castles (2004) points out that several domestic factors in the receiving country can also affect the outcome. These include conflicts of interest between different groups, the political desire to control migration, the actions of civil society, and, finally, the significance of the migrants’ rights that are recognised by liberal democracies (Hollifield, 2008). The failure to meet migration policy goals may also have something to do with the fact that the means used can lead to goal attainment in theory, but in practice have diametrically different consequences. One example is stricter border controls that, in many cases, do not lead to lower immigration but instead increase the significance of human smugglers and, furthermore, lead to irregular migrants staying in the country when the possibility of return is limited (Cornelius & Tsuda, 2004).

The argument over the lack of effectiveness in migration policy is mostly because international migration can be seen as mainly driven by structural factors, like labour market imbalances, inequalities in wealth and political conflicts in origin countries; all factors on which migration policies have little or no influence. Rather than affecting overall volumes of inflows, immigration restrictions primarily change the ways in which people migrate, such as through an increased use of family migration or irregular means of entry (Czaika & de Haas, 2013). Moreover, there is also a considerable gap between tough immigration discourses by politicians and actual migration policies, which are generally much more nuanced and varied. Tough rhetoric may give the misguided impression that immigration policies have become more restrictive.

Thus, there is a need to further dissect policy practices by acknowledging the considerable gap between policies on paper and their practical interpretation and implementation. The extent to which written policies are implemented varies widely and depends on factors ranging from financial and human resources, the weighing of different and potentially competing policy priorities, to the discretion of civil servants and other state agents (Czaika & de Haas, 2013).

There are four distinct levels in migration public policy: discourses, actual migration policies on paper, policy implementation and the actual migration, the latter being the policy outcome. This fourfold distinction allows for the identification of three ‘immigration policy gaps’: (a) the discursive gap, or the discrepancy between public discourses and policies on paper; (b) the implementation gap, or the disparity between policies on paper and their implementation; and (c) the efficacy gap, or the extent to which implemented policies can affect migration. The discursive gap, between discourse and practice, is common in public policy, not necessarily related to a policy failure, and it will not be covered here. The efficacy gap is also not addressed here, for that matter.

The implementation gap, which is covered here, relates to the fact that some rules and regulations are not—or only partially—implemented because of practical, planning or budgetary constraints. The non-implementation could also be a consequence of corruption, ignorance or subversion. Politicians, civil servants and private companies (e.g. airlines implementing carrier sanctions, asylum case workers, border agents, public institutions or private institutions processing work visa requests) often have considerable discretion and agency in the way they implement policy on the ground (Ellermann, 2006; Wunderlich, 2010).

This implementation gap seems to be particularly significant if there is a large degree of discretion and assessment involved in policy implementation. This leaves considerable scope for subjective interpretation and political or public pressure, for instance in refugee status determination and work permit applications (cf. Ellermann, 2006). Having discretion increases the willingness to implement policy in general. Empirical studies also confirm this, linking discretion via client meaningfulness to a larger willingness to implement policy. The positive effect that discretion has on the bureaucrat’s perception of client meaningfulness can be seen as a condition for the second effect: more willingness to implement the policy. When street-level bureaucrats perceive that their work is meaningful for their clients, this strongly influences their willingness to implement it (Meyers & Vorsanger, 2003; Tummers & Bekkers, 2014).

Thus far, I have discussed implementation of migration policy in general. If we move closer to the enforcement of rejected asylum applications (also called deportation, removal or expulsion), several normative, empirical and theoretical issues surface. To restore and retain the integrity of the immigration system, deportations are necessary. Sometimes deportation policy targets what can be called less attractive groups, such as ‘bogus’ asylum seekers, illegal immigrants and criminal foreigners (Arnold, 1990).

However, the actual implementation of a deportation policy affects other individuals as well, including families with children or socially and economically integrated immigrants. When confronted with these cases, public opinion tends to change and point in another direction. The need to use coercion is not an easy thing to handle for liberal democracies and there is always the risk or possibility that there is grass-root mobilisation against some deportation cases from civil society actors (Ellermann, 2005, 2006). This mobilisation, regardless if we agree or not, is a substantial threat to the bureaucratic capacity in this field.

Understanding the extent to which states have the capacity to implement the enforcement of rejected asylum applications within the existing institutional structures is interesting. One issue is the complexity, meaning that enforcement capacities require an effective bureaucracy and interpersonal relations that enable the collaboration between the different state and non-state organisations—operating at the international, national and local levels—involved in return (Leerkes & Van Houte, 2020). To a large extent, governments are developing policy to apprehend and deport unauthorised migrants. The legal and administrative framework of Western Europe generally allows for strict interior policing of unauthorised migrants (Leerkes et al., 2012).

Implementation of Return Policies: The Case of Sweden

Sweden can, in political and administrative terms, broadly be characterised as a non-corrupt, liberal democracy with a centralised state that generally has high capacity in implementing policies in different fields. This well-functioning bureaucracy has long roots in Sweden and was already present in the pre-democratic epoch. This feature is often seen as a precondition for efficient policy making.

When it comes to enforcement of rejected asylum seekers, both main implementing agencies—Migrationsverket and Polismyndigheten—are independent and have an autonomous role in relation to the government. The present system with two agencies in charge, where Migrationsverket handles the ‘voluntary return’ and Polismyndigheten all remaining cases, developed in the late 1990s.

In numbers, the return cases increased significantly in Sweden between 1999–2018. From about 5000 cases per year around the turn of the century to 20,000–25,000 cases in the 2010s. This increase is partly behind why the issue of return was put high on the agenda around 2010, becoming a priority for letters of appropriation. Despite having a reputation and a tradition of being generous in terms of asylum and refugee reception, migration policy was overhauled in a severely restrictive direction and the border controls were increased in the aftermath of the large refugee influx in 2014–2015.

The Understanding

The question of understanding how to implement a decision is about understanding the real meaning of the phenomenon. In implementation literature, this depends on, among other things, decisions, goals and guidelines being communicated to relevant enforcing authorities and other bodies, as well as that there is no stark discrepancy between what is said centrally and what is carried out in the field. Previous studies of enforcement work in Sweden indicate that there is evidence of communication deficiencies, partly explained by the phenomenon of ‘the whispering game’. It essentially means that communication across each link of the policy chain with an agency and the exercise of authority is distorted or misunderstood—either downwards or upwards (Borrelli, 2018).

One example, from Borrelli’s research on how street-level bureaucrats handle their tasks regarding enforcement of return decisions, is from Gränspolisen (the Swedish Border Police), where officers note, with great deal of irony, that their prime minister publicly announced an increase in deportations to 80,000 per year, without taking into account its feasibility. The street-level bureaucrats felt excluded from decision making, thus affecting their acceptance of the policies. Potentially this can strengthen distortion and wilful dismissal as well as ignorance of the information communicated. Decisions that are passed down are often not understood since they come from a detached group of superiors, without any practical knowledge or street-level experience (Borrelli, 2018, p. 810).

Communication deficiencies are one of the more common obstacles noted in the implementation literature; its presence is regularly highlighted when listening to the experiences of the informants. Several representatives for the agencies provided examples of when information, fully or partially, stays within an agency, unit or division when it ought to have been communicated onwards: the more links in the chain, the greater the risk of transfer losses. A top-level senior officer at Migrationsverket, Sverker Spaak, points out that his agency, through its return liaison officers, usually has good knowledge of what is happening in individual countries, like Afghanistan (Spaak, 2019). However, this knowledge at the higher echelons rarely reaches front-line bureaucrats who directly communicate with individual asylum seekers.

Similarly, there are several examples of where the agencies are not allowed to communicate freely because information must not be disclosed due to privacy or confidentiality rules:

[t]he privacy rules need to be reviewed so that information can be shared about the returns between the different agencies and actors [in a way] that the procedures can be further improved […] For example, in the REVA project, where there were many experiences that may not have been followed up and implemented properly. (Eva Qvarnström, Nationella Transportenheten [NTE—Prison and Probation Service’s Transport Service] 2018; REVA stands for Legal Certainty and Effective Enforcement, a project carried out by the Swedish Police together with other agencies in order to boost the effectiveness of enforcement of deportations)

From the informant statements, it is also evident that there are many potentially conflicting objectives for return policies that front-line and mid-level bureaucrats face, both in the enforcement process itself and in relation to other migration policy goals. One such conflict is between the goal of effectiveness and the other two often-mentioned goals of humanity and legality. In the daily work of the front-line bureaucrats, certain situations create the need to determine how to weigh the different goals against each other. What sort of means should be used in the field in order to successfully enforce a decision, without breaking any law or practice intended to protect values such as humanity and legality? Methods and approaches are tested time and again, as recounted by many informants at both Migrationsverket and Polismyndigheten. Another source of tension exists between the agencies’ right to use coercive measures to enforce decisions versus the right for individuals to not be subjected to restrictions of their freedoms (Malm Lindberg, 2020).

The front-line bureaucrats might also suffer from conflicting messages arising from different policy goals, namely integration versus return. How to make priorities when the focus is shifting? One informant at the municipality level states that:

[r]eturning has not been an alternative that has been talked about at all. When we started with this, both children and adults were saying, “what, we’re not going home, are we?” There was like no … you work, work and work for integration but you haven’t even looked at reintegration in the home country … but in fact, a rather large percentage of both children and adults have their applications denied. And our society has not been prepared to deal with this at all. (Elisabeth Lindholm, Strömsund municipality, 2019)

There are other kinds of conflicting objectives as well. In the present system there are several obstacles to enforcement that are often mentioned by the street-level bureaucrats—especially at Migrationsverket and Polismyndigheten—that conflict with their daily work with the cases. It is possible to identify three different themes, which are also extensively elucidated by our informants, that influence implementation efforts: the possibilities to apply for a change of track, the possibilities to argue impediments to enforcement (verkställighetshinder, VUT) and the possibility to get a regularisation decision.

The first illustration of potentially conflicting objectives can be taken from the possibility to switch tracks in the asylum system. The possibility of obtaining a labour market work permit after a rejected asylum application has come to be called a change of track and one consequence of the opportunity to change track is that it puts a strain on the system. A senior officer at Migrationsverket points out that there is also a risk of confusing operations by making the boundary between being asylum seekers and job seekers more diffuse. By applying for asylum, one gets the opportunity to work and, thereby, a chance to stay in the country despite not having grounds for asylum. The broadened opportunities for a change of track mean that a ‘no’ in the asylum process does not have to be a ‘no’ to stay in the country, which then means that those who have been denied asylum have no incentive to prepare for return (anonymous interviewee, Migrationsverket). A similar argument is presented by another senior officer at Justitiedepartementet:

[c]oming here and applying for asylum can just be a way to legalise your stay and be here for a short time and do something else. This can be working or … our inability to handle these asylum applications very quickly makes it tempting to come to Sweden because you can legally be here for a long time. (Mikaela Eriksson, Justitiedepartementet, 2019)

The opportunity to claim and have an impediment to enforcement (VUT) assessed occurs after a refusal-of-entry or expulsion decision has gained legal force. It refers to new circumstances that have not been assessed before in the asylum process. The question of how to deal with impediments to enforcement is important to many of the informants at the agencies that were interviewed. One informant, a regional process manager at the Gränspolisen (2019) also emphasised the problems—with an ironic touch—associated with an already legally binding case being taken up again:

[i]t’s great that there are so many instances and that it will be examined and so on, but you can be, like, applying for impediments endlessly. (Malin Köhler, Polismyndigheten, 2019)

At Migrationsverket, a team leader, also points out that the asylum seeker has no incentive to prepare for a return and that, thereby, the process becomes less predictable and thus, from her perspective, less legally certain:

then I actually don’t think that we are meeting the legal requirements of both the Return Directive and the Aliens Act. This whole thing that a process must be able to be predictable […] That’s even difficult for ourselves [to know]. “Wait, a closed enforcement case, but it’s up as a secondary school case. Is she enforceable, or is she not?” How is the applicant supposed to understand all of this herself then? (Jessie Ahrdenberg, Migrationsverket, 2019)

Her latter statement points to the fact that her job being a front-line bureaucrat is severely constricted by legislation—including new laws like the High School Law as described by Bucken-Knapp and Zelano (2023)—that gives extra possibilities for rejected asylum seekers to reapply and start the process anew.

In summary, bureaucrats on all levels face conflicting objectives for the return policies, whether in relation to the enforcement in itself or in relation to other migration policy goals affecting implementation. The different objectives also seem to send out a signal through the system, not only to bureaucrats but also to the rejected asylum seekers and their supporters that a no in the asylum decision is not always a no to stay. Consistency in the system is often claimed to be a necessary condition for successful implementation, but terms such as ‘consistent’ are not used by our informants when they speak about the present system.

The Ability

When policies are to be implemented in a certain field, the ability to do so must be present. To have the ability to implement, an effective division of responsibilities and duties between the actors involved is required. Assets like knowledge, competence and capacity must be at the public administration’s disposal, at the right time and at the right place. The most important relationship within the implementation chain is between Migrationsverket and Polismyndigheten. Here there is an interface, a sort of shared boundary, where cases pass mainly in one direction and where information is lost, partly because of the differences in the case management systems.

These differences also mean that statistics are quite difficult to interpret and draw conclusive implications from, even by professionals. This was highlighted in the final report of the REVA project, something we will return to, which points to the large improvement potential that lays in creating common metrics and statistics across agency boundaries (Migrationsverket, 2014, p. 52). This feature is noted by the Ministry of Justice, but with little success.

We stand here and stamp our feet a bit and we are a little frustrated ourselves because we don’t come any further, our ministers expect the statistics they get to be complete, but we always have to add that there are deficiencies in the statistics from the police, year after year […]. We don’t really see that we have moved forward. (Mikaela Eriksson, Justitiedepartementet, 2019)

When it comes to ability in the sense of working methods and routines, Lipsky’s (1980) theory of front bureaucrats emphasises that not all of them act the same. Rules are interpreted a little differently in the field and are stretched from time to time—which is broadly expressed by the informants. This results in a lack of uniformity, but it also points to the discretion that, in turn, is needed in the profession. Since both agencies have the nature of front-line bureaucracies, this is a result we should expect. This applies especially to Polismyndigheten, which before 2015 was divided into different agencies at a regional level, a factor that still seems to have an impact.

If capacity is about having the necessary physical resources at hand, competence is about having the skills or know-how to complete the tasks. At both agencies, core competencies are either lacking or sometimes outright missing. Polismyndigheten often uses the term border police competency, which refers both to knowledge of laws concerning aliens as well as to methods and approaches that are used during so-called in-country checks. Several informants emphasise that there are far too few police officers who have this special competency and that the education at the police training academies is not enough to complete the tasks.

Moreover, not only are the training initiatives too small, but additionally the level of ambition in the courses has also decreased. Officers who work on the streets are given only basic training that, “allows police officers to ask questions during, for example, in-country checks, but not always to interpret the answers” (RPS, 2014, p. 6). Thus, in order to be able to carry out their tasks, police officers must build their own networks of support outside their own circles. Besides the lack of training, there is also the issue of experience. Outside of the three metropolitan districts, the various non-metropolitan police districts have so few cases that the competency and know-how level of the staff is not easily maintained. The result is that individual police officers refrain from completing (justified) checks of aliens because they are inexperienced and afraid to make mistakes (RPS, 2006, p. 4; Ann-Marie Orler, Polismyndigheten, 2019; Göran Millbert, Justitiedepartementet, 2019).

However, the complexity that this work entails already requires greater competence among the personnel, including those at the Migrationsverket. Not least, in order to feel confident in their roles and develop as decision-makers,

[y]ou have to see that this job, the complexity we have talked a lot about, but we have to admit that it is difficult and requires training. Time is needed to reflect and take in what you have learned, new practices, and to discuss. (Niclas Axelsson, Migrationsverket, 2019)

Without this knowledge and skills, the ability to perform operations, such as in-country checks, is limited. Competency deficiencies can also be an important reason why Migrationsverket has difficulty solving the problem of uniform and reliable statistics. This makes it very difficult to accurately categorise and measure return, which then makes it difficult to evaluate the return policy.

The lack of resources to handle enforcement decisions can, according to the informants, be attributed to a combination of factors. One aspect worth mentioning is the distribution of resources over time:

[w]hen … a lot of people are coming in during a short period of time, the process can be shaken because it is difficult to allocate resources correctly. (Milot Dragusha, Justitiedepartementet, 2019)

This, in turn, stems from the large and often unexpected fluctuations of cases. The picture drawn by several of the informants is that enforcement operations are difficult to plan, thus affecting the sustainability of their work.

Long-term planning and coordination are also affected by the control that different units inside return operations and enforcement have over resource allocation. Those in charge of return operations do not control the flow of resources for the activities linked to return. To channel more resources, it is essential to convince, for example, regional managers at Migrationsverket around Sweden to allocate these (Malm Lindberg, 2020).

In conclusion, the relationship between the two implementing agencies is of utmost importance since there is an interface where cases pass from the Migrationsverket to Polismyndigheten and sometimes back again. Information is often lost due to differences in the case management systems or rules on sharing information as well as the inherent fact that there are two types of agencies involved. However, the ability challenge also touches upon the ways that resources are handled. Thus, personnel, premisses and equipment are not the real issue, rather it is the competence.

The Willingness

One feature that research investigates in connection to the work efforts and motivation of street-level bureaucrats is their willingness to implement policy. Although this willingness, or alternatively resistance, inside the bureaucracy to implement public policies is the third and last of our categories, it is not the least important.

Tummers et al. (2012) construct a framework, based on empirical findings, for analysing the willingness to implement policies and examines three factors that influence it: (a) the policy content and discretion; (b) the organisational context; and (c) the personality characteristics of the public professionals. The first factor (a) primarily relates to the discretion available during the implementation and the perceived meaningfulness of a policy. The second factor (b) relates to the influence of professionals, the subjective norms of managers and the subjective norms of colleagues towards the policy. The last factor (c) relates to the rebelliousness or rule compliance amongst implementors. While it is not possible to systematically investigate these factors in the interviews that were performed, it is still a good framework to proceed from.

An issue we touch upon is the resistance to implementation among street-level bureaucrats. We know from previous research that personnel at Migrationsverket can have a certain degree of resistance to implementation if the decisions from the agency do not align with what the street-level bureaucrats feel is right (Qvist, 2008). The motivation of those who implement the decisions can also be influenced by aspects like the nature of the case. There are signs, among informants at both of the two most important agencies, that resistance to a specific part of the job in this group changes the time spent and efforts made. Within the enforcement of return cases, some of the hardest cases seem to be those involving children. Some questionnaire responses from the staff in Jönköping County to one of Rikspolisstyrelsens (RPS—National Police Board) (RPS, 2007a, pp. 22–23) earlier investigations illustrate this fact:

[f]or one thing, it’s really hard on the staff to have to frequently enforce decisions on families with children, especially when the children are used as ‘weapons’ to get to stay in the country […]. Eventually, the strong pressure, from the public, municipal authorities, church organisations, and even from the press, does a lot to make our work more difficult.

In this case, it seems to be the organisational context regarding the role of professionals and managers inside the organisation, but also actors outside of it, that de-motivates them.

In order to implement policy, those who perform the actual enforcement actions, both organisations and individuals, must be fully motivated to carry out their tasks. Research in this field, including on Swedish conditions (Hansson, 2017), points out that the enforcement work is emotionally charged and difficult to manage, not least for the front-line bureaucrats who, in the worst case, suffer a deterioration of their psychological well-being. Here, it seems to be the policy content in itself that is hard to motivate from. Another aspect that seems to influence willingness is whether (return) activities are noticed and to what extent these are prioritised within the organisation or the agency in question.

Talking about prioritisations: all organisations with different functions divided into departments or units tend to, sooner or later, develop internal hierarchies where some parts are superior to others. This superior–inferior order expresses itself in dimensions like status differences, where the units receive different levels of attention from the management. So how does it look like in the area of return within, primarily, Migrationsverket and Polismyndigheten? Both internal and external investigations and reports, as well as informants’ statements, paint a picture that return work is not prioritised. The RPS’s (2007a) extensive investigation into more effective and legally certain enforcement mentions right in the second paragraph that these operations, “live a somewhat obscure and vulnerable existence at many police agencies”, and it emerged that enforcement of return decisions “is considered by many police officers as an odd business” (RPS, 2007b, p. 1).

The informants who were interviewed testified that the status of return and enforcement issues is not exactly a top priority at either Polismyndigheten or Migrationsverket. To a direct question posed to some team leaders in three Migrationsverket regions, the response was that return activities were not prioritised. It is considered that,

[t]he agency talks about the importance of return, but the experience is that this is not reflected in the conditions given to carry out qualitative and effective return work. (Anonymous at Migrationsverket 2019)

Sverker Spaak expresses that it is very rare that anyone at agency management level talks to the press about return issues. In the organisation’s internal hierarchy, these issues, along with detention centres, are at the bottom of the list, which also manifests in staff flows, attention and influence over the agency’s management. During the time as process manager, the senior officer worked closely with Polismyndigheten, often collaborating with them. However, this was met with some internal resistance at the agency. It was seen as a risk that the agency was ‘too exposed’ in these matters—and if this had to be the case, it should preferably not be together with Polismyndigheten.

A desk officer from Gränspolisen asserts that, “border policing has been an odd bird [and] a bit of a satellite in policing operations, where there has been focus on so much else” (Åsa Petersson, 2019). This phenomenon, being somewhat outside the core of the policing duties, is also vividly described by the head of Gränspolisen:

it’s about the border police and policing activities in general going through a sort of paradigm shift, where the border police have long been considered—and consider themselves—to be the extended torpedo arm/muscle of the Migration Agency. The Migration Agency with force. So now we’re very actively trying to bring border policing operations much closer to the Police’s core mission. (Patrik Engström, Gränspolisen, 2019)

The above statements do not really answer the question of why return activities appear to be given lower status and lower priority in the work of both Migrationsverket and Polismyndigheten. Thus, one reason for the low priority could lie in the fact that employees more often want to deal with issues other than return and detention. Naturally, the causal relationship can also be the reverse: that the staff notices and feels that return is not really a priority inside the organisation and, thus, they seek other areas of activity within both agencies and that those who are interested in these issues are deterred by its low status.

As Ellermann (2005, p. 1234) mentions, political support from the government is vital, in her case from different regions in Germany. Our case also shows the fundamental importance of having the politicians’ support. The aforementioned REVA project (2009–2014) did not, in and of itself, give Polismyndigheten greater authority in doing in-country checks, but was merely a process overview of enforcement work. The ambition was that in-country-checks were supposed to be an integral part of regular police work or police operations (Malm Lindberg, 2020). What happened in 2012 and 2013 is that it received media attention and triggered a debate. Criticism came from many actors, including politicians and the Polisförbundet (the Police Union) Chairperson, among others. Several informants from Polismyndigheten also confirm that the criticism affected their actions:

[t]here are several reasons behind the reduction of in-country checks. Partly, I would say that the so-called REVA debate was absolutely a reason that many policemen no longer dared. (Patrik Engström, Gränspolisen, 2019)

Another informant emphasised that this debate, portraying police enforcement as inhumane acts of state coercion, in his mind, had even deterred candidates at Polishögskolan (PHS—Police Academy) from completing internships in units conducting enforcement work (Jerk Wiberg, Gränspolisen, 2019).

To summarise, it is evident that those who execute the enforcement work, both organisations and personnel, must be motivated to carry out also unpleasant tasks that are emotionally challenging. Another aspect that seems to influence willingness is whether return and enforcement activities are prioritised within the organisation. The ‘odd bird’ metaphor points to the fact that such issues might be prioritised in rhetoric at the top-level, but it seldom trickles down to the lower levels. It is also clear that agencies are not immune to opinions from the outside and, in the case of return and enforcement, a partial explanation for its low status may be the sentiments it arouses.

Effects of the Covid-19 Pandemic

As mentioned earlier, the interviews that formed the empirical foundation for this chapter took place in a pre-pandemic era. But in early 2020 the outbreak of the Covid-19 curbed the ability to implement rejected asylum seekers and other irregular migrants. The main reason seems to have been from both legal and physical reasons, such as the travel restrictions in place and the lack of available flights. Here we note that the policies of the countries of origin played as an important role, but also restrictions in Sweden such as reduced detention capacity to meet sanitary measures (EMN, 2021). More generally, and not only related to return policies, it also seems that the current toolbox of international migration governance has few or ill-adapted instruments for dealing with the impacts of a pandemic of the nature and reach of Covid-19 (Newland, 2020, p. 2).

The implementation literature has dealt with the challenges posed by crises like the Covid-19 pandemic. Such crises are characterised by unexpected and continuously changing circumstances. Government responses to such events depend even more than normally on a constant bidirectional information flow between decision-makers and implementers at the front line. The downward flow of comprehensive and clear information is necessary, but also the upwards information flow providing feedback about how formal policy decisions are met and affect the street-level (Gofen et al., 2021, pp. 484–485).

Albeit more speculative, with regards to the Covid-19 pandemic it is possible that the challenges imposed by a turbulent environment give an unusually long ‘window of opportunity’ to change policy design—if policymakers can adapt. The lack of regularity and previous experience disrupts the status quo and surely provides challenges in the short run, but possibly also opportunities in the long run.

Conclusions

This chapter starts with a reference to the Gap hypothesis, first and foremost the implementation gap of policies. Essentially it means that some rules and regulations are not, or rather only partly, implemented because of different kinds of constraints. The bureaucratic capacity is limited, and this is mainly because the prerequisites for a successful implementation are seldom available. The Covid-19 pandemics can be considered a crisis when circumstances change unexpectedly and continuously which gives, at least in the short run, even larger challenges.

We should keep the following in mind: To succeed in their intent, implementers inside the migration bureaucracy must understand how to implement policy, while also having the ability and the willingness to do so. Further, they should also have a favourable environment to work in. As noted earlier, the irregular migrants or rejected asylum seekers belong to the group that is supposed to return ‘voluntarily’; certainly, these individuals (and families) have a different goal to that of the personnel designated to enforce these decisions. It is a fundamental logic of the system that is worth repeating.

Another fundamental aspect has to do with the constitutional liberal democratic state that gives the right to individual assessments and humanitarian considerations; these constitute a certain inherent restriction on enforcement and return tasks. Yet another aspect is the fact that working with return and enforcement is not a priority at either agency and not really in line with their core tasks or mission. That return work is sometimes seen as ‘low status’ at these agencies can also be a factor that explains why both Migrationsverket and Polismyndigheten have a hard time meeting the need for competence.

Moreover, charged with the mandate of implementation, the front-line bureaucrats, along with those at the upper level in this study, consider public attitudes on enforcement to be, more or less, biased against deportations. The phenomenon is easy to criticise and almost no matter how bureaucrats act, there is a risk of ending up sitting on the accused bench.

Lastly, it would be unfair to the struggling return or deportation bureaucrats to portray them as outright shirking or resisting implementation or outright failing in their mission. In many senses, they, as respondents, show a strong commitment to fulfil their duties. Much of the material in the empirical parts of this chapter have a certain bias towards problems and challenges faced by the deportation bureaucrats. In many ways, it is typical of implementation research to paint a gloomier picture than the underlying facts deserve.