Limits of Activation? Street-Level Responses to the 2015 Refugee Challenge in German Job Centers

Introduction

The arrival of about two and a half million refugees from Syria and other humanitarian crisis regions in 2015 and 2016 constituted a massive challenge for European governments and administrations. In the absence of a comprehensive European approach, migratory pressures led to a variety of political and administrative responses (Trauner, Citation2016). For Germany, the situation was particularly demanding: with more than 1.1 million refugees applying for asylum, the country turned into one of the top five countries in the world hosting refugees (Bundesamt für Migration und Flüchtlinge, Citation2017; United Nations High Commissioner for Refugees, Citation2021). This massive influx posed a challenge not only to those agencies responsible for asylum procedures and humanitarian aid (Dittmer & Lorenz, Citation2021). Managing reception and integration processes—a task that would continue for years to come—also involved finding solutions to a range of interconnected problems, including housing, social assistance, and access to the labor market. These responses together with policies already in place affected the refugees differently given that German law differentiates between asylum seekers and legally recognized refugees, two groups whose entitlements and social rights differ greatly. Asylum seekers and those denied asylum are “kept outside the usual framework of welfare regulation” (Liedtke, Citation2022, p. 479) and only enjoy limited labor market access (Crage, Citation2016). Recognized refugees granted protection, who are the focus of this article, enjoy full labor market access. As long as they are unemployed, they are entitled to basic security according to Germany’s Social Code II. Thus, the job centers, which are responsible for providing social assistance, were assigned a key role in the process of integrating this group of refugees (Brücker et al., Citation2020; Knapp et al., Citation2017).

In the job centers, the massive and abrupt influx of refugees produced important consequences for the daily work of street-level bureaucrats (SLBs). For several reasons, one could expect the increase in client numbers to disrupt their work routines and thereby the way activation policies are produced on the ground (Hupe & Hill, Citation2016; Brodkin, Citation2011; Lipsky, Citation1980). First, the situation required rapid organizational adjustment which, in turn, likely modified the SLBs’ framework of action. Second, given the large and unfamiliar group of clients, SLBs were confronted with uncertainties and new demands in their daily work. Third, given the refugees’ claim to protection and their social integration, which was required politically (emphasized by former chancellor Merkel’s claim, ‘We can do it’), the situation might have created a high sense of moral purpose and politicized the role of SLBs.

Much of the SLB literature has explained the coping behavior of SLBs in situations of particular stress and in the field of working with refugees with individual-level factors (James & Julian, Citation2021; Belabas & Gerrits Citation2017b; Gofen et al., Citation2019). The question of how the organizational context affects street-level practice has, to date, not been fully elaborated (Gofen et al., Citation2019; Hupe & Buffat, Citation2014).

Based on qualitative interviews, this study takes a two-level approach to explore how (1) local job center organizations and (2) placement officers have responded to the abrupt change in client structure and respective demands, and how the two levels are intertwined. With this question, we also examine how and whether “public service gaps” (Hupe & Buffat, Citation2014), i.e. situations in which demands exceed resources—not only in terms of numbers but also in terms of substantial expertise in dealing with the new client group—have been successfully prevented.

The aim of this article is twofold: We first try to advance recent SLB research that focuses on the question of how formal and informal working environments and relationships shape street-level practice (Peake & Forsyth, Citation2021; Raaphorst & Loyens, Citation2020; Siciliano, Citation2017). By examining how specialized refugee teams motivate empathetic approaches toward clients at the street level, thereby moving the limits of activation faced in the standard operation procedure, our case study explores the mechanisms through which specific organizational features can shape SLB perceptions and action patterns. Second, we aim to foster a dialog between migration studies and SLB research. We argue that an organizational lens can advance migration research, which, for a long time now, has neglected the question of how state organizations shape migrants’ trajectories (Castles et al., Citation2014; Lang et al., Citation2021). Taking the inherent logics and limits of activation work as a point of departure, we thus take up the claim to de-exceptionalize migrants and migration, and embed migration research in broader theories of social structure (Castles, Citation2010).

The next section briefly reviews the literature on professional frontline work with refugees and coping at the street level, to then present and explicate a heuristic of the inherent logics and limits of activation work. Following the description of methods and data, we describe two organizational responses to the refugee challenge in Germany and present a theory-based and empirically grounded typology of SLB action orientations. The last section summarizes the findings and contributions of this article and draws out implications for the in/exclusion of refugees as mediated by state bureaucracy.

Logics and limits of activation work: An analytical framework

As a rational way of coping with high workload, scant resources, and the need to fulfill institutional criteria, SLBs commonly move away from clients (Lipsky, Citation1980; Tummers et al., Citation2015). In situations of high workload, the work with migrants can be experienced as especially difficult (Schütze & Johansson, Citation2020) leading to the sidelining of clients that pose problems and higher sanctioning rates (Schram et al. Citation2009).

Yet, recent research on professional frontline work with refugees has identified a different action pattern: when refugee clients are perceived as extremely burdened and deserving of support, and when frontline workers such as teachers, integration counselors, or placement officers are critical of relevant state policies or programs, they bend the rules in their clients’ favor. This pattern is presented as an individual decision of frontline workers “beyond the formal task description” (Belabas & Gerrits, Citation2017b, p. 144), which takes place in the shadows despite “neoliberal managerial constraints” (James & Julian, Citation2021, p. 534). The logic of helping clients through rule adjustment or deviation also resonates with patterns mentioned in the broader literature on street-level bureaucracy, such as the “moral carer” (Zacka, Citation2017), or the “citizen-agent” (Maynard-Moody & Musheno, Citation2022). If at all, these patterns are explained with individual-level attributes such as SLBs’ moral disposition, values, or world views (Zacka, Citation2017), or as resulting from interaction with clients and specific perceptions of them (Maynard-Moody & Musheno, Citation2022).

As we know, organizations can powerfully shape street-level practice (Rice, Citation2012; Scott, Citation1997). Yet, as Sandfort has rightly argued, this influence does not only works through “organizational charts, formal procedures, or written rules that staff passively enact” but also through “the collective, daily experience shared by front-line workers” (Sandfort, Citation2000, p. 731).

Relatedly, recent SLB scholarship has focused on the relevance of formal and informal networks and social relationships for street-level practice (Siciliano, Citation2017; Zacka, Citation2017). It has been shown that and how working environments and social dynamics between officials, colleagues and citizens can increase the autonomy of SLBs but also hamper it (Peake & Forsyth, Citation2021; Raaphorst & Loyens, Citation2020). In line with this literature, this article conceives of SLBs as nested within organizational contexts (Gofen et al., Citation2019).

Zooming in on our case, the task of integrating refugees in the German workfare system, we argue that we first need to understand the general logics and limits of activation. Figure 1, the activation rhombus, summarizes our understanding of the activation process.

Figure 1. The activation rhombus.

Display full size

We understand activation as a policy to create market-oriented and marketable subjects by promoting employability and the corresponding motivation (Wright, Citation2016; see Hansen, Citation2019, regarding the moral foundation of activation policies). Activation policies are implemented by organizations and the rules, success criteria, management practices, and operational procedures that they have in place. Organizations themselves are complex entities with various intermediate layers of authority. In decentralized systems in particular, the local organizations—job centers, for instance—are often semi-autonomous and act as mediating or buffering instances between the macro level and the street level. They frame and guide the activation efforts of SLBs through formal and informal action prescriptions (see Figure 1, axis 1-2, solid arrow). As individual cases differ greatly, SLBs are entitled to exercise case-related discretion to integrate their clients into the workforce—be it through support, persuasion, or legal coercion. Thus, placement officers act in a system marked by bounded autonomy and institutional control (Brodkin, Citation2011; Evans, Citation2013). Central to activation work is the activatability of welfare claimants. Based on the standardized assessment of a client’s job readiness and potential barriers to labor market integration, SLBs construct working strategies, including the assignment of support measures and potentially achievable demands regarding participation (see Figure 1, axis 1-3-2, solid arrow). That said, activation is always oriented toward desired outcomes. Its success is commonly judged according to the prevalent criteria of the implementing organization, that is, measurable indicators of performance and outcomes (see Figure 1, axis 1-4, solid arrow).

Given that rules and criteria are always ambiguous and “never precise enough to cover the complexities” of every possible decision-making situation (Mahoney & Thelen, Citation2010), SLBs inevitably have to apply their own interpretation in casework. In doing so, they also respond to various accountability regimes (Hupe & Hill, Citation2007) and the related work pressures (Hupe & van der Krogt, Citation2013) emanating from the regulated public-administrative sphere, professional standards as well as societal and client expectations. For this reason and to turn the conceptual thinking into an empirical research agenda, we translate the external view of the activation process—as illustrated by the solid arrows in the activation rhombus—into a structural model of action patterns—visualized by the dotted arrows. Zooming in on the internal views of SLBs, we propose to break them down analytically into three distinct but interrelated dimensions of activation efforts (see Table 1). In doing so, we refine existing comprehensive frameworks to explain street-level practice (e.g. Loyens & Maesschalck, Citation2010; Rice, Citation2012) in the case of activation work, and expand them through a focus on outcomes.

1. The constitutive dimension of SLBs’ activation efforts: In their daily practice of applying abstract rules to concrete cases, SLBs need to interpret the (sometimes uncertain or conflicting) expectations and requirements of the organization and their superiors and define their own role and professional mandate in this context (see Figure 1, axis 2-1, dotted arrow). This also includes and entails a definition of their own goals on the way toward the labor market integration of clients, their perception and prioritization of problems, and the corresponding general classification schemes in regard to clients.

2. The substantive dimension of activation efforts: Deriving from the constitutive orientation is the action level, that is, SLBs’ working strategies and routines. These encompass counseling concepts and the use of discretionary powers, guided by different degrees of self-authorization in regard to superiors and by the assessment of clients based on personal encounters (see Figure 1, axis 2-3, dotted arrow). The nature of fit between a client’s performance and the strategies and categories applied informs moral judgments in everyday practice.

3. The outcome-related dimension of activation efforts: SLBs might assess the effectiveness of their own work in line with performance or outcome metrics defined by their organization. Yet the way SLBs perceive success or failure can also follow different criteria that emanate from their professional identity. In addition, SLBs might reflect upon their role and work, consider possible unintended consequences, and factor these into their approach (see Figure 1, axis 2-4, dotted arrow).

Based on the activation rhombus, we now ask what could render activation work vulnerable. What are the external and internal factors that constitute limits to successful activation? One external factor is undoubtedly the regional labor market situation, as low demand for labor likely reduces the opportunity to place welfare claimants in work (Johnson et al., Citation2021). In our case, the labor market situation was generally favorable (Konle-Seidel, Citation2017). However, some structural factors, including the legally imposed residence obligation for refugees or poor transport infrastructure in rural areas, restricted labor market mobility and (may have) hampered activation work (Bonin et al., Citation2021).

Limitations also arise when external and internal factors collide in unforeseen ways. This is the case when a sudden increase in the number of welfare recipients meets insufficient internal resources (quantity). Resulting capacity constraints might jeopardize the institutionalized promise to provide time- and labor-intensive individual support. A further limit to activation can arise when the characteristics (qualities) of the cases change, inter alia due to an increasing diversification in case types and their specific difficulties. This might reduce the suitability of existing tools and resources, ranging from substantive support measures to placement officers’ communicative and case-analytical skills.

In 2015/16, Germany was certainly not completely unprepared. The German Residence Act of 2005 and the Integration Course Ordinance, together with the non-legally binding National Integration Plan of 2007, set important parameters for German integration policies, including mandatory language courses for migrant welfare recipients with little German skills, and measures to support the intercultural opening of state agencies (Bundesministerium für Arbeit und Soziales, Citation2011; Mushaben Citation2017).

Nevertheless, the sudden and unprecedented rise in the number of refugee clients clearly represented a challenge to German job centers in terms of quantity and quality. It strained the capacities of an already burdened system and challenged operational routines. Placement officers were suddenly confronted with a large, widely unfamiliar, and heterogenous group of clients. Most of the refugees did not speak German upon arrival and were unfamiliar with the German culture and the procedural logics of German state agencies. Many were burdened by experiences of war and flight. Our data suggests that even in big cities with a high share of migrant welfare recipients where one could expect job center staff to be used to working with migrants, the new group of clients was a real challenge (“We realized that we were clueless”, team leader, sampling point C, Bavaria).

Relevant questions in this context are to what extent the classification schemes of SLBs get confused or reinforced, how new ones emerge, and how these are nested in and influenced by organizational adjustments. In addition, as we know, an increased workload can lead SLBs to perceive work with migrants as difficult (Schütze & Johansson, Citation2020). In this case, would such perceptions prompt SLBs to rely on stereotypes, question the legitimacy of refugees as welfare recipients, and sideline or punish those who appear uncooperative (Ratzmann & Sahraoui, Citation2021; Schram et al., Citation2009)? In other words, do tough working conditions lead to an erosion of SLBs’ moral sensibilities, as Zacka (Citation2017) suggests? Or does the greater scope of discretion granted through formal directives or tacit permission generate new forms of practice that work to support the new client group (Brodkin, Citation2021)?

Methods

To explore how local job centers and SLBs coped with the new challenge, this article uses a qualitative case study design (Yin, Citation2017). Following the criteria of maximum variation sampling (Patton, Citation2002), we selected 18 (out of 402) job centers at four regional sampling points that varied in terms of socio-geographic location, labor market situation, and share of refugees among the unemployed (see Table 2).

The resulting sample proved to generate a variety of organizational and individual responses suitable for an exploratory study. The article builds on semi-structured qualitative interviews with 45 placement officers, team leaders, and management staff. The research design received prospective approval by the Review Board of the Institute for Employment Research. All interviewees have provided written informed consent to the data being analyzed and published.

In the interviews with frontline officers, we encouraged reflections on practices and routines in working with refugees. The interviews with superiors focused on organizational responses to the refugee challenge, strategic alignment, and task-based recruitment strategies. All interviews were conducted in 2018 and 2019, hence, at a time when SLBs had already gained a wealth of experience in working with refugees. They lasted from 30 to 170 min with an average of 60 min, and were all audio-recorded and fully transcribed.

The transcripts were coded thematically and analyzed comparatively. Following the advice for theory building from cases (Eisenhardt, Citation1989), we constructed the dimensions of action patterns presented in the structural model (see Table 1) through an iterative process. By going back and forth between data and concepts we identified two maximally contrastive ideal types, that is, model pictures representing action patterns “within a specific interaction context” (Lindbekk, Citation1992, p. 290). In Weberian sense, the types are heuristic in nature and serve to conceive of activation efforts as a meaningful practice (Swedberg, Citation2018). Conceived as stylized descriptions of action patterns in specific institutional contexts, the typology is not intended to explain the emergence of the types systematically.

“People are coming, and they are coming now. What do we do?”
Different approaches in the activation work with refugees

Discussion and conclusion

The aim of this article was to explore the responses of German job centers to the refugee challenge at the organizational and individual levels. We have identified two types of social action: the formalists, with their bureaucratic understanding of activation work, and the mentors, who adapt to their refugee clients with an approach based on empathy and trust. While the types mirror the state-agent and citizen-agent narrative (Maynard-Moody & Musheno, Citation2022), our findings goes further. They indicate that the two action types are not random but shaped by the respective organizational setting: either the standard procedure or special teams.

The special unit approach proved to be an appropriate response to the refugee challenge, as it can encourage SLBs to search for new ways to support their refugee clients. We find a pattern of authorized rule bending according to which problem-sensitive SLBs feel directly or indirectly authorized to interpret rules flexibly and work for the client. Self-authorization is fostered by problem-sensitive superiors and new organizational structures at the local level that in many ways signal the specificity of activation work with refugees. Authorized rule bending is an expression of professional agency based on in-depth case knowledge and exchange in teams, which is promoted more strongly in specialized units than in the standard procedure.

By showing how local organizational structures that facilitate interactions with refugee clients are positively associated with empathetic approaches toward clients and by highlighting the role of supervisors as mediators of organizational structure, this paper extends recent insights from the international literature on frontline work with refugees that commonly explain rule bending as individual resistance to managerial demands or policy guidelines (Belabas & Gerrits, Citation2017b; Eriksson & Johansson, Citation2022; James & Julian, Citation2021). Moreover, our findings advance SLB research with its growing interest in the question of how formal and informal working environments and relationships can powerfully shape street-level practice (Peake & Forsyth, Citation2021; Raaphorst & Loyens, Citation2020; Siciliano, Citation2017).

Our analytical framework, the activation rhombus, and the derived dimensions of activation efforts (see Figure 1 and Table 1) proved to be promising tools to explore the links between coping mechanisms at the organizational and individual levels. Future research should advance such reflections to increase our understanding of “nested influences, such as individuals in organizations” (Gofen et al., Citation2019, p. 345) in the complex field of social interaction between migrants and the state. Another relevant question emerging from our finding on nested agency is: What structures the structure? It is an important task for further research to understand more systematically what factors at the macro and meso levels have shaped the decision for or against organizational restructuring (e.g., governance processes, historical legacies of regional labor markets, servant leadership characteristics, etc.) in this moment of crisis and such times in the future.

Finally, unpacking the inherent ambiguities of the two configurations (specialized-mentorist and standardized-formalist) leads us to consider the wider implications of our findings for the inclusion/exclusion of refugees as mediated by state bureaucracy. The specialized-mentorist configuration works to overcome the limits of activation and to prevent the emergence of a public service gap by explicitly acknowledging the deservingness of refugees (Jilke & Tummers, Citation2018) and through a language-sensitive approach (Holzinger, Citation2020) while at the same time preparing the preconditions for refugees’ self-determined social positioning in the host society. While inclusionary in nature, this approach risks being too paternalistic at times and may sometimes underestimate the need of some refugees to earn money quickly. This could become particularly problematic in constellations where refugee clients shy away from disappointing their caring mentors (see Whitworth, Citation2016, regarding the contradictions and implications of paternalism in a neoliberal system).

In the standardized-formalist approach, the goal of rapid labor market integration requires refugees to assimilate to the demands of the system quickly. Formal bureaucratic criteria then lead SLBs to distinguish between cooperative (good) and recalcitrant (difficult) clients, and subsequently to the common action pattern of creaming and parking (Kaufman, Citation2020). Following Hupe and Buffat (Citation2014), these advisors have to do “more with the same” as they face an extra workload and new demands. They can easily become frustrated by seemingly unmotivated clients, demanding superiors with little problem sensitivity, and their own awareness of limited success. As indicated by the feedback loops in Table 1, their own demotivation can, in turn, reinforce prejudice and negative views of clients (constitutive dimension of activation efforts), lead to lower commitment (substantive dimension), and reduce activation successes even further (outcome-related dimension). While the formal activation targets and the interest of many refugees in getting into work quickly may coincide, the long-term risk of marginalization is high. We can expect opt-outs by disappointed clients, but also revolving-door effects and a group of refugees who become the new long-term unemployed because they are not adequately addressed by the activation system. In addition to the risk of biographical fates and breaks in employment trajectories, the standardized-formalist configuration thus carries the broader risk of reinforcing social exclusion if refugees feel formally labeled and see themselves as outsiders (Borrelli & Bochsler, Citation2020; Schierup et al., Citation2015). Therefore, following up on our findings by examining the manifest and latent effects of early bureaucratic contacts on migrants’ integration trajectories is a promising avenue for future research (Belabas & Gerrits, Citation2017a; Kyriakides et al., Citation2019).