Responsive professionalism in post-NPM reforms: the case of the Swedish police

ABSTRACT

This article investigates responsive professionalism in local collaboration across organizational boundaries, introduced by public reforms as a means to address the downsides of new public management (NPM) and to foster professionalism ‘closer to the citizens’. The case in point is the Swedish police reform, and the empirical material comprises 39 semi-structured interviews with police officers. Drawing on the work of Noordegraaf (2020. “Protective or Connective Professionalism? How Connected Professionals Can (Still) Act as Autonomous and Authoritative Experts.” Journal of Professions and Organization 7: 205–223. doi:10.1093/jpo/joaa011) and Anteby et al. (2016. “Three Lenses on Occupations and Professions in Organizations: Becoming, Doing, and Relating.” Academy of Management Annals 10 (1): 183–244), indicating that occupations today earn recognition for their professionalism within wider networks of other occupations, this study examines police officers’ occupational awareness of how to maintain responsive professionalism during local collaboration. Findings illustrate how police officers conceive of themselves as moderators, orchestrators, and mediators guided by occupational awareness of different organizational strains and of demands for responsive professionalism. The Swedish police reform was found to promote officers’ independent engagement in formalized or ‘abstract’ modes of professionalism that, nevertheless, continue to be responsive to local demands.

KEYWORDS:

• Responsive professionalism
• public reforms
• police
• occupational awareness
• post-NPM
• abstract policing

Introduction

New public management (NPM) reforms, informed by hierarchical and market-oriented principles, have transformed public-sector organizations into actors commanding their own resources and pursuing their own goals (Hood 1991; Christensen and Lægreid 2007). Rules emanating from governmental centres have accordingly become fewer and less specific, but studies show that there are disadvantages to this. In particular, the increasingly fragmented policy coordination urges each organization to reinforce a management-by-objectives schema that easily becomes an end in itself, marginalizing occupational groups and their autonomy to make informed judgements in dialogue with local actors (Brunsson and Sahlin-Andersson 2000; Evetts 2011). Public authorities’ local legitimacy may thereby also erode, as occupational opportunities to enact professionalism based on expertise and ability to translate knowledge, codes, and standards to the needs and features of the case at hand are shrinking (Noordegraaf 2020).

This paper addresses a recent set of reforms that, over the past decade, has set out to address these downsides (Adler, Kwon, and Heckscher 2008). By giving occupations opportunities to engage independently in collaboration across organizational boundaries with other local actors providing public services (Kirkpatrick and Noordegraaf 2015; Rolandsson 2017), these reforms are intended to restore public professionals’ opportunities to manage their daily practice in accordance with their expertise (Bringselius 2019), making public services more sensitive to local needs (Adler, Kwon, and Heckscher 2008). However, studies suggest that public professionals also have to encounter negotiations with increasing numbers of collaborators (Noordegraaf, Van der Steen, and Van Twist 2014), exposing them to continuous new demands for what can be described as responsive professionalism (cf. Anteby, Chan, and DiBenigno 2016) and for an ability to address the increasing ‘coordinative complexity’ (cf. Bohle Carbonell et al. 2014).

Research has investigated how occupational groups address such coordinative complexity in collaboration across occupational boundaries, in teams within organizations and in temporary inter-organizational projects (Bohle Carbonell et al. 2014; Tyskbo 2020). In addressing measures that dissolve organizational constraints, these studies illustrate how occupations strive to maintain their jurisdictions (Abbott 2015; Kellogg 2019) or strengthen trust in their capacity to provide public services (e.g. Gleeson and Knights 2006). The analysed occupations often seem to engage in various kinds of boundary work, allowing them to maintain recognition of their unique expertise (Bos-de Vos, Lieftink, and Lauche 2019). However, these studies have rarely addressed the impact of post-NPM reforms intended to redress the identified downsides of NPM. More precisely, we lack knowledge of what occupations exposed to these post-NPM reforms do in practice to maintain their status as legitimate actors when collaborating across organizational boundaries as part of their daily practice, beyond specific teams or temporary projects.

This article addresses this gap by analysing how the Swedish police reform has conditioned police officers’ awareness of how to integrate such coordinative complexity in local policing. Unlike previous studies of boundary work, occupational strategies, or mechanisms for maintaining jurisdiction and authority in defined teams or temporary projects, this study examines how police officers make sense of legitimate or ‘good work’, continuously enacted relative to various collaborators across organizational boundaries (Noordegraaf 2020). The focus is on police officers awareness’ of how to accommodate expectations regarding their capacity to maintain professionalism in collaboration with other local actors of importance to their provision of local services (Björk 2019). The research question is: how do the police officers relate to demands for professionalism in cross-organizational collaboration involving collaborators guided by a variety of aims and expectations?

As an occupation, policing does not display professionalism that is justified by time spent in higher education (Fyfe 2013). The police also do not have the type of universal code of ethics that guides many other professions (Neyroud 2008). In line with previous studies, this study nevertheless recognizes that police professionalism draws on institutionalized expertise and practices that provide broad guidelines for confronting the risks and uncertainties found in modern societies (Evetts 2003; Fyfe 2013). This study investigates how the police conceptualize their practical understanding of such uncertainties through collaborative arrangements introduced by the Swedish Police reform, which explicitly claims to ameliorate the downsides of NPM. This reform has centralized executive power (21 police authorities have been combined to form one national authority consisting of seven regions) while introducing new procedures for local collaboration with municipalities and for citizen dialogues (Björk 2019; BRÅ 2018). The present findings describe how, by introducing collaboration across organizational boundaries, the reform promotes occupational awareness of both collaborative tensions and organizational routines and resources. We show how a new form of responsive professionalism has emerged, requiring that police officers throughout the organization independently engage in what can be described as formalized or ‘abstract policing’.

Before looking closer at these findings, the following two sections describe how this article contributes to previous research and to the theoretical foundations for analysing police officers’ awareness of how to integrate the reform. The subsequent sections then describe the methodological framework for data collection and analysis, followed by the findings, discussion, and conclusion.

Previous research on public reforms and professionalism

Ample research describes how NPM reforms foster organizational professionalism informed by hierarchical and market-oriented principles involving formalized decision-making and management by objectives (cf. Hanlon 1998; Evetts 2003). These studies identify mentioned downsides of NPM, and repeatedly conclude that organizational professionalism marginalizes modes of occupational professionalism, drawing on the enactment of autonomous expertise informed by substantial assessments and collegial principles (Freidson 2001; Evetts 2011). By replacing independent and informed judgement with habitual patterns of appropriately carrying out organizational functions and goals, professional actors sometimes emerge as functionaries recalling Simon’s (1997) classic description of the administrative man, indicating a decline in professionalism. Other studies addressing attempts to manage strains between formalized governance and occupational expectations of autonomous expertise (Hanlon 1998; Fournier 1999) have described how public reforms forge new forms of hybrid professionalism, blurring the distinction between occupational and organizational professionalism (Kirkpatrick and Noordegraaf 2015; Kirkpatrick 2016). In most cases, however, a strained relationship between occupations and management continues to condition occupational expertise, authority, and ability to autonomously translate knowledge and standards to the needs and features of the case at hand (Styhre 2014; Noordegraaf 2020; Swinkels and van Meijl 2020).

References to this strained relationship recur in research into how post-NPM reforms have altered governance and policy coordination (Christensen and Lægreid 2007; Holmberg 2019). While these studies do not explicitly investigate professionalism, they indicate that occupations may restore professional autonomy locally and gain the ability to move public service provision closer to citizens by engaging in cross-organizational collaboration (Minas, Wright, and van Berkel 2012; Bringselius 2019; SOU 2019, 43). Accordingly, the Swedish police reform explicitly sought to restore trust in independent police officers’ ability to build a ‘closer’ professional relationship with citizens and local communities (BRÅ 2018; Statskontoret 2018). Research into professionalism in cross-occupational collaboration, as enacted in formal teams or temporary projects, implies that such collaborative measures could restore professionalism based on occupational autonomy and competence (Adler, Kwon, and Heckscher 2008; Tyskbo 2020). Involved occupations must address ‘coordinative complexity’ (cf. Gleeson and Knights 2006; Bohle Carbonell et al. 2014; Kellogg 2019) or spend time on boundary work to maintain recognition as exclusive experts (Bos-de Vos, Lieftink, and Lauche 2019). Because these occupations are required to adjust their practices to constant negotiations with others in their teams or projects, they may foster emergent forms of professionalism, allowing them to (re)gain independence relative to management (Noordegraaf, Van der Steen, and Van Twist 2014; Bechky and Chung 2018).

This article underscores the need for empirical investigations of public reforms that introduce different forms of collaboration beyond specified teams or temporary projects, as measures intended to ameliorate the downsides of NPM. The article analyses inter-organizational collaborative measures aimed at strengthening occupational expertise in daily practices that continue to depend on regular organizational routines and boundaries. Professionalism in this case is not bound by narrow or tightly organized units (e.g. teams or projects) introducing collaboration that temporarily dissolves managerial distinctions between organizations. Furthermore, the analysis assumes that the involvement of increasing numbers of external collaborators in post-NPM reforms ought to make it difficult to restore professionalism based on conventional forms of occupational autonomy (Noordegraaf, Van der Steen, and Van Twist 2014). The study suggests that, rather than bringing back exclusive policing, solely accountable to the law and to collegial norms, police officers must engage in constant negotiations that shape how their authority and competence are perceived (Adler, Kwon, and Heckscher 2008; Bohle Carbonell et al. 2014).

In addition, the analysis recognizes that international research on police reform has continued to identify a tension between demands for externally oriented occupations with certain types of expertise and the management of these occupations. Reforms involving evidence-based police work (Terpstra and Fyfe 2014), local partnerships, problem-oriented community policing, etc. (Rolandsson 2017) expose conflicts between exclusive forms of police professionalism, based on the notion of an autonomous occupation of ‘crime fighters’ (Fyfe 2013), and professionalism guided by organizational concerns for accountability and legitimacy (Stone and Travis 2011; Larsson and Lundgren Sørli 2018). Furthermore, recent studies comparing, for instance, post-NPM reforms in the Dutch and Scottish police (Terpstra and Fyfe 2014), have identified an emerging tension between demands for loyalty to local communities and organizational directives shaping local policing. Facing both accusations of conducting outdated policing that is too loyal to local collaborators (Terpstra and Salet 2019) and demands for managerial instructions and alignment with strategic targets, police officers prioritize what can be described as ‘abstract’ or formalized modes of local policing (Terpstra, Fyfe, and Salet 2019), the latter implying the persistence of some kind of organizational professionalism.

Theory

To analyse how police officers relate to collaborators’ expectations of police professionalism, the analysis introduces occupational awareness as an analytical concept, referring to how occupations understand themselves as able to navigate demands for professionalism in practice (Rolandsson 2020). In line with Boltanski and Thevénot’s (2006) description of agents’ ‘critical capacities’, the focus is on their practical understanding of how to manage collaborative relationships across organizational boundaries. By engaging with awareness of professionalism in responsive acts, the analysis differs from previous field-level studies of professionalism as a matter of socialization, and from studies of how specific occupations maintain professional boundaries or specific jurisdictions (Freidson 2001). Instead, the analysis draws on what Anteby, Chan, and DiBenigno (2016) called a relating lens on professionalism, guiding how occupations conceptualize their engagement in responsive acts across organizational boundaries. In line with Noordegraaf (2020), expertise, autonomy, and authority also constitute key dimensions shaping occupational awareness of how to answer demands for professionalism in these responsive acts. That is, the article examines the occupational response to collaborators’: (1) reliance on state-of-the-art knowledge, standards, and skills in the police; (2) expectations regarding the ability to independently translate knowledge and standards to the needs and features of the case at hand; and (3) recognition of police authority while complying with and updating standards and codes.

According to Noordegraaf (2020), contemporary occupations encounter various difficulties in safeguarding recognition of these key dimensions. Technological and organizational pressures force these occupations to stop relying on institutionalized forms of what he called protected professionalism, and to engage in new forms of connected professionalism dependent on a wider environment of other actors. Noordegraaf’s description of these challenges resonates with post-NPM reforms, which introduce new collaborative measures that promote coordinative complexities (Bohle Carbonell et al. 2014) and force professionals to rely on appearances and responsive acts, rather than on association. In line with scholars who have frequently noted different degrees of continuity in such changes (Adams et al. 2020a, 2020b; Noordegraaf 2020), the following analysis nevertheless recognizes that occupations such as the police have always been obliged to somehow connect to a wider environment of other actors. This study thus sets out to treat all claims about new and dominant forms of responsive police professionalism with caution, instead aiming to deepen our understanding of how demands for responsive professionalism emanating from post-NPM reforms become meaningful relative to varied responsive acts and practices. Drawing on what Anteby, Chan, and DiBenigno (2016) called three filters underscoring collaboration, coproduction, and mediation as key dimensions of the relating lens on professionalism, this analysis scrutinizes context-bound variations in police officers’ awareness of how a wider environment of actors recognizes their expertise, autonomy, and authority.

The three filters constituting a relating lens on professionalism

By applying these filters, this analysis examines police officers’ awareness of how to accommodate expectations of responsive professionalism in settings involving both organizing mechanisms that enable stabilization of cross-organizational collaboration and engagements with wider, more complex webs of stakeholders and collaborators (Bechky 2003).

The first filter, relating as collaborating, lets us scrutinize the meaning of various mechanisms enabling collaboration across organizational boundaries. This filter views routines, regulations, or rules as mechanisms allowing occupational members to come together to specify shared responsibilities for targets and tasks (Feldman and Pentland 2003). Boundary objects, such as agreements, formal directives, or jointly stated goals, may also emerge as mechanisms facilitating the translation of meanings and negotiation of status across occupational or organizational boundaries (Bechky 2003; Levina and Vaast 2005), enabling occupational groups to collaborate despite their differing professional expertise and beliefs (cf. Rolandsson 2017). Anteby, Chan, and DiBenigno (2016) also cited studies illustrating how common spaces or zones can function as mechanisms that facilitate collaborative communication and identification. Previous studies have frequently emphasized the importance of occupations’ ability to identify such occupational mechanisms to overcome inherently tense cross-occupational relations that could otherwise trigger distrust in their professional capabilities (DeBenigno and Kellog 2014; Bos-de Vos, Lieftink, and Lauche 2019).

The second filter, relating as coproducing, directs attention towards how occupations align and interact with multiple other parties to coproduce recognition of their professionalism in ways allowing them to manifest or extend their societal influence. In this filter, we include an entire field of stakeholders with which the occupations are interdependent. We may identify the recognition of professionalism in local arrangements as less restrained by formalized tasks and occupational knowledge, letting the involved parties complement one another while functioning as a network of expertise (Eyal 2013; Huising 2015). Anteby, Chan, and DiBenigno (2016) illustrated this filter by referring to mental-health professionals who must rely on their ability to involve patients’ parents in diagnosing conditions such as autism. We may also think of police officers who extend their social influence while engaging in crime prevention with collaborative networks, including municipalities, local football clubs, churches, and mosques (Donnelly 2013). Briefly, this filter lets us examine the recognition of occupational influence in broader networks of actors (Huising 2015).

The third and last filter, relating as brokering, may eventually enable us to examine the recognition of professionalism in intermediary practices. This filter concerns brokering that fills critical gaps in multifaceted networks, enabling complex webs of collaboration to persist by connecting, buffering, and mediating among the involved people and tasks. The involved intermediaries link actors across official boundaries – that is, instead of focusing on one organization or occupational group and its advancement, they connect actors in attempts to facilitate, for example, reforms or innovation processes. Relating as brokering could then also emerge in response to the increasingly complex divisions of labour within and across organizations. Some studies have described, for instance, how reforms that rely on cross-occupational collaboration require brokers who can either connect or buffer high-status occupational groups from increasingly complex interaction with others (Kellogg 2014). Mediating actors may also have to distance themselves from local conditions and downplay their own interests and demarcated responsibilities if they are to continue serving as links in complex webs of actors (Maguire, Hardy, and Lawrence 2004; Canales 2011).

Relating lens: analysing the performance of responsive professionalism in practice

Summing up, these three filters direct the analysis towards variations in occupational awareness concerning how to enact responsive professionalism in interaction with collaborating actors – i.e. they enable us to identify and analyse the various meanings of cross-organizational collaboration that emerge (cf. DeVault and McCoy 2012). The collaborating filter directs attention towards how occupational actors relate to what can be considered narrow ties, forging routines, and organizational mechanisms that enable collaborating occupations to specify tasks, goals, and responsibilities (Adler, Kwon, and Heckscher 2008). The coproducing filter allows us to examine how these actors deal with the influences emerging when occupations engage with wider networks of stakeholders (Powell 1990). The brokerage filter may eventually enable us to scrutinize how occupational actors facilitate mediation in complex webs of collaboration that support different types of dissemination or knowledge sharing (Canales 2011).

By constituting guidelines for various responsive acts, the filters also help us understand how police officers assess their own ability to integrate expectations concerning expertise, autonomy, and authority (Anteby, Chan, and DiBenigno 2016). Complementing previous studies, for instance, by Noordegraaf (2020), emphasizing the impact of connections with the wider and more complex environments of others, we examine police officers as occupational subjects who actively bring meaning to the enactment of responsive professionalism in cross-organizational collaborations. In other words, the analysis deepens our knowledge of how the outlined filters shape police officers’ understanding of themselves as occupational subjects able to accommodate collaborators’: (1) daily reliance on their state-of-the-art knowledge and skills; (2) expectations regarding their ability to independently translate knowledge and standards to the needs and features of the case at hand; and (3) recognition of their authority or way of complying with and updating standards and codes.

Method

The case of the Swedish police

By centralizing executive power and policy coordination within the police, the Swedish police reform analysed here is intended to delimit managerial fragmentation. Twenty-one former Swedish police authorities have been combined into one national police authority containing seven regions. Within each of these regions, former police authorities also constitute several police districts, comprising increasingly larger local police districts that have replaced multiple smaller local districts. In this way, the reform has introduced organizational conformity across the country (cf. Christensen and Lægreid 2007). By introducing cross-organizational collaboration, however, the reform also set out to strengthen local police officers’ autonomy and provide them with opportunities to influence and adapt their practices to be ‘closer to the citizens’, in accordance with their own expertise and local conditions (Larsson and Lundgren Sørli 2018; Statskontoret 2018). This involves a set of coordinative measures, such as promises to citizens, routines, and ideas about information sharing as well as new positions introduced to enhance police sensitivity to external actors and their expectations (BRÅ 2018; Bringselius 2019). In combination with the above-mentioned centralization, these measures provide the analysis with local police officers who have been exposed to recognized post-NPM conditions that encourage them to reflect on how to accomplish responsive professionalism shaped by ‘coordinative complexity’ (cf. Bohle Carbonell et al. 2014).

The empirical material

The main empirical material for this analysis draws on interviews with 39 police officers. Regarding the contextual aspects (e.g. demography and geography) that condition how respondents react to the reform, they work in police districts in different parts of Sweden. These include inner-city police districts, police districts struggling with high criminality, police districts covering socially stable medium-sized cities (50,000–100,000 inhabitants), and rural police districts. Moreover, still other respondents work in police districts encompassing both medium-sized cities and rural areas. By interviewing police officers in charge of municipal collaboration, officers in charge of specific neighbourhoods, inspectors, first-line police managers, and local police managers, the empirical material includes respondents in various positions expected to influence the organization of local police work. Some of these are new positions introduced by the reform and intended to harness police engagement with local communities (see Table 1). In addition, other interviewees include investigators and police managers higher in the hierarchy, in charge of budget and policy coordination on a more central regional level.

Table 1. Empirical material, including semi-structured interviews and documents.

Pilot interviews	Interviews with municipal staff (n = 6)	Interviews with police officers (n = 39)	Documentation of local police measures
- 1 regional police manager - 1 police manager in charge of local police districts - 8 community police officers	- 6 security coordinators working for municipalities in a big city, midsized town, and/or rural district	- 4 regional police managers - 5 police managers in charge of local districts - 13 community police officers - 2 inspectors in charge of local teams - 9 local inspectors - 1 operative coordinator - 3 officials involved in investigation and strategy	- public promises by the police to citizens in local districts - operational and communication plans - agreements about municipal collaboration - local crime and safety statistics - evaluations of local crime and crime prevention measures

To widen the scope of concern and verify statements about cross-organizational collaboration by the police, the study included interviews with six civil servants representing local municipalities and collaborating with local police. To contextualize the interviews, the researcher also conducted observations at two conferences in which police officers discussed local policing in rural areas or areas exposed to high criminality. To deepen our understanding of the local contexts shaped by the reform, the analysis also examined documents, i.e. operational or communication plans as well as goals and instructions for local police practices (involving the above-mentioned promises and agreements with the municipalities). Documents emanating from the central level (e.g. public reports reflecting broader political discussions, national policy documents, national instructions for local police work, and evaluations) were also examined (Brå 2018; Statskontoret 2018; SOU 2019).

Conducting and analysing the interviews

The data collection started with pilot interviews with 10 respondents (see Table 1) representing different parts of the police authority. These respondents provided useful descriptions of cross-organizational collaboration that enabled the identification of valid questions and the design of a reliable interview guide. The interviews were semi-structured, addressing themes and questions regarding broader issues of public reform and professionalism in collaboration. The questions encouraged the police to consider how they handled their daily practice as representatives of a public authority. Applying a semi-structured procedure allowed the researcher to follow up comments with questions encouraging the respondent to elaborate on how the police reform has conditioned local police work. The interviews lasted one to two hours (Gubrium et al. 2012).

Drawing on a logbook used in connection with each interview, the researcher began identifying possible themes during the data collection. The analysis intensified after completing and transcribing all the interviews. The first phase involved open coding, allowing the researcher to segment the text and identify concrete collaborative activities recurring in the transcripts. By coding these activities and relating these codes to one another, the analysis then entered a phase of axial coding (Miles, Huberman, and Saldana 2014), identifying broader themes describing how the respondents conceptualized their own agency and conditions for inter-organizational collaboration. During this axial coding phase, the analysis identified how the police made sense of relational arrangements that tied them to their local collaborators. Informed by the theoretical foundation and the relating lens, the analysis also identified how the previously mentioned filters applied to the different activities and means that facilitate police collaboration with other stakeholders in the local community. Some codes that emerged in the process were: local presence, networking, collaborative arrangement, personal ties, representative authority, broker resources, and normative judgements.

The analysis eventually identified tensions in each filter that guided how the respondents made sense of the collaborative arrangements. These tensions proved to be important, enabling deeper analysis of how the police officers conceptualized their own agency and conditions for responsive professionalism in collaborative local police work. By rereading the transcripts, the researcher continuously tested and recoded the statements throughout the analysis (Miles, Huberman, and Saldana 2014). To ensure trustworthiness, during the early analytical stages the researcher took part in a seminar with police staff, who provided input on preliminary interpretations. At the final stage, when the analysis began identifying different types of police agency that underlie responsive professionalism, the researcher discussed preliminary interpretations with a group of police officers.

Results

This section presents the empirical findings. The text starts by presenting how the relating as collaborating filter underlies the police officers’ occupational awareness, before considering how they apply the relating as coproducing filter and the relating as brokering filter. The police officers emerge as: (1) moderators who must be aware of how to temper expectations of constituting a capable and present local authority, (2) orchestrators who must be aware of how to coordinate collaborators, continuing to recognize them as independent and influential police officers, and, finally, (3) mediators who must be aware of how to gain and provide knowledge that enables them to be recognized as competent and locally informed.

Relating as collaboration: moderating local negotiations and expectations

Beginning with how the police officers apply the collaborating filter, we see that it evokes their awareness of measures or mechanisms that offer formalized aims and routines for collaborating across organizational boundaries, primarily with local municipalities. The respondents referred to collaboration agreements and various formalized meetings that facilitate negotiations and communication between police and the municipal management of, for example, social services, housing, drug prevention, and local schools. Drawing on these agreements, local action plans, and crime statistics, the interviewees also described how they formulated promises articulating police goals for the local community. Local police discuss these promises with the municipalities so as to adjust them each or every other year, and they also hold citizen conferences with the municipalities, allowing them to justify themselves as knowledgeable officers who are sensitive to local citizens’ experiences and expectations regarding local policing. The new position of community officer, introduced by the reform, emerged as an important factor in coordinating these mechanisms. A police manager organizing the community officers in one region emphasized their strategic importance:

I appreciated the idea immediately – that is, community officers who work with local collaboration, who engage with the collaboration agreements and formulate our promises to the local communities, and who do so guided by this long-term perspective, which I think is the crime prevention perspective. This is great, because we have not had this before, there has been no such specific function.

By sustaining regular negotiations about responsibilities and goals, these mechanisms tie local police and municipalities to local policy coordination, supporting the perception of the police as both a strategic and locally present authority. The mechanisms constitute formal means crucial to the collaborating filter guiding the police officers, particularly when they engage with municipal concerns about the negative consequences of the centralized police authority for the local police presence and officers’ ability to maintain social order. However, the interviewees also described how – if they are to be recognized as police officers who can independently manage local collaboration – they struggle to maintain municipal interest in collaboration: they must deal with unengaged municipalities and constant disagreements. They also must prevent municipalities from being disappointed by avoiding promising too much or being too explicit about providing local police services – that is, they must be aware of how to moderate municipal expectations.

So, we’ve learnt a lot from the procedure leading to these so-called promises to the local citizens. We have gained a certain understanding of the different perspectives in each other’s organizations. We realize that we must have a certain understanding of the somewhat slower procedures in the municipality. It takes some time, but when they come to a decision, they stick to it – something we previously struggled with in the police. We have sometimes decided to leave an initiative, simply stating that we don’t do it anymore because we’ve found something else to prioritize. In that way, these promises have the advantage of at least providing some continuity.

The quoted officers described how they continued collaborating even when upset with their municipal partners (municipal interviewees also confirmed that they can become irritated with the police). Bluntly, the mechanisms underlying the collaborating filter make it possible to accommodate and endure tensions in collaborations between the police and municipalities. By constituting a framework for negotiations with municipal representatives, the various mechanisms emerge as boundary objects allowing the officers to address differences and endure the collaborative policy coordination that underlies official recognition of the police as a locally present authority. However, the police officers were aware that they still must accommodate expectations that may be excessive relative to their means and aims. That is, to maintain recognition as local authorities, they must be moderators who actively determine how to both regulate expectations and uphold their own agenda relative to the municipalities’ agenda. One police officer, in charge of local community policing strategies in a central big-city police district, explained:

… but we prioritize networking that provides us with information that enables us to take police measures and do what is actually our job. To do so, this group is currently also increasing demands on schools, stating that that if we are to collaborate with you, you have to sign this agreement. And this agreement comprises a drug policy that permits us to actively use the means at our disposal, for instance, to enter the school with our dogs in order to search lockers, etc. In this way, we are getting somewhere, but of course not everyone wants to be part of it.

How this officer described agreements as a means for both parties to reach consensus and push for action (e.g. supporting effective drug prevention) recurred among respondents from other police districts as well, particularly when they referred to how, in specific situations, they addressed concerns about lack of resources. There were circumstances, however, that appeared to be too difficult to manage, implying both that collaborative mechanisms do not always fit local conditions and that it is sometimes difficult to provide the kind of cross-organizational collaboration recognized by responsive professionalism. For instance, the police could end up spending too much time or too many resources managing agreements, citizen promises, and citizen dialogues. They also encountered municipalities with excessive expectations concerning police officers’ local engagement: some municipalities had become disappointed and stopped negotiations concerning, for instance, the above-mentioned promises. Such tension could challenge recognition of the police officers as local authorities worth collaborating with. In some cases, this type of challenge also caused frustration. One police inspector explained:

If you ask me, these promises are bullshit! They are simply a few nice words in a text, and you wonder if there is anything wrong with the other agreements we’ve had with the municipalities before. What about the different plans we used to have with them previously? And then again, it means that they use staff who could have worked on our core tasks and call them officers in charge of local collaboration: people who used to be part of our core are being moved out. It’s the same story when it comes to the other new position, assigned specific districts, where we objected because we claimed that we all do that type of local policing. By reallocating these people, the ones who are left have to work even more.

The quoted inspector worked in a rural district comprising several municipalities and a widely distributed population that expects the police to maintain a local presence. To her, it was difficult to spend all the time needed to temper the collaborating partners’ expectations: they ended up moderating formal strategies and expectations rather than conducting police work ‘closer to the citizens’. Other officers, working in a rural setting, also noted that the vast distances characterizing the new local police districts meant that they must be pragmatic and try to do a little bit of everything, while complaining that these negotiations were too time consuming, making it difficult to deal with ordinary people’s complaints about their lack of local presence. This concern was explicitly articulated by rural police districts, but some officers in city centres and in districts characterized by high rates of criminality also described having to struggle independently to find enough time for routines and negotiations concerning these collaborative mechanisms. Being absorbed in forging formal connections with the local community, they described continuing to be perceived as local officials from far away. This was confirmed by a police authority evaluation describing how the reform fostered indirect rather than direct relations with citizens. Subsequently, concerns about police officers’ ability to act independently as moderators, tempering expectations of local police abilities and authority, emerged in various local settings.

Relating as co-producing: orchestrating a sense of presence and influence

The relating as coproducing filter emerged when the police officers described how they arranged initiatives with wider groups of collaborators who can support their local influence. In this case, they did not primarily engage in negotiations, but in activities whereby they coproduced local policing by coordinating wider networks of actors, involving a set of local governmental partners, property owners, neighbourhood watch groups, and other volunteer groups. More precisely, they described acting as orchestrators of wider networks of collaborators, strengthening recognition of the police’s local capability by fostering a sense of community in certain neighbourhoods or districts. One interviewed officer described how this is done by, for instance, engaging civilians in a group of police volunteers who wear vests decorated with the police emblem and who serve more or less as police proxies who can talk with citizens in person:

For instance, we’ve used our special group of volunteers to knock on doors in areas known for high criminality. Even if we don’t do the talking, we can still learn about people’s opinions and concerns. This is interesting, because we had this rather bad crime a couple of years ago, when we asked them to be involved in knocking on people’s doors. This was by the time the local police district had expanded, and the volunteers thought it was wrong: ‘We have other things to do, after all this is a rather serious crime’. However, when they started doing it, they realized it was great. They got to meet people who otherwise never meet the police; they got invited in for coffee and received tips. Since then, our volunteers have been out there knocking on people’s doors on a more regular basis.

The volunteers referred to in this quotation coproduced local policing with officers who were investigating a serious crime in a neighbourhood marked by social problems. When walking around talking to citizens, they became the police’s extended connection with local citizens, bringing a sense of safety and confirming the local police presence. Dressed in yellow vests with the Swedish police emblem, they represented the police in face-to-face interactions with citizens; they also enabled the police to take advantage of the fact that wider groups of locals found it easier to talk informally with these volunteers. By facilitating person-to-person connections in the neighbourhood, the volunteers emerged as proxies who could ensure a sense of local police authority. In many other cases, the police also orchestrated these networks by engaging in direct personal connections with key actors in the local community. One police officer in charge of community policing in an urban city centre described her work:

I work with the head and the security boss for the central shopping mall, the security boss for the central station, and business associations in different city areas, and I simply don’t have the means needed to work with everyone. It is simply impossible to do it all, because then you also try to connect with, for instance, elderly care. After all, you can’t try to meet each pensioner individually, or sit down and talk with him or her during the day. You have to identify bigger arrangements. I’m so happy if I can find an occasion that allows me to talk to a hundred pensioners at the same time, saying that, for example, on Monday next week, we’re going to have a citizen dialogue about safety.

The respondents described orchestrating person-based connections with key actors representing private or public housing associations, congregations, schools, volunteers, shop owners, municipalities, etc., and providing access to broader groups of citizens, which could strengthen recognition of both their expertise and their capacity to adapt their skills to the local needs expressed by a range of resident groups. At the same time, they had to be aware that such orchestrating involves time-consuming meetings and networking, which encourages them to be increasingly strategic. Several inspectors also complained that such engagement left less time to meet ordinary people on the streets, and that they still struggled with the fact that citizens no longer expect police to be directly involved with the local community: citizens tend to be surprised and ‘ask if something has happened’ whenever the police turn up. In addition, officers’ increasingly strategic engagement with key actors requires that they spend more time critically considering whether these actors have expectations regarding coproduction with the police that may or may not be appropriate. One police officer in charge of community policing strategies in a district known for high criminality explained how he had to consider legal as well as normative implications:

Yes, we think about the law and regulative constraints – this comes automatically. If I have a group of parents or volunteers maintaining safety in a specific neighbourhood, there will be questions about resources, and you might have that tenant who promised to provide a place for them, but who suddenly withdraws. The chair of this group might then phone me, saying ‘After all, we have been supporting you in the police by walking around the neighbourhood, making people feel safe. Couldn’t you just phone up the tenant and make sure he’ll provide us with this room anyway?’ This group is truly good, but I simply can’t engage in this type of networking. What type of collaboration would that foster? However, there are situations in which we have to collaborate with people whom the intelligence service has told us to avoid.

Although these networks provide opportunities to orchestrate a sense of police influence, the preceding quotation shows that the police officers must be aware of public expectations that may be linked to normative concerns. Once again, their ability to regulate expectations emerges as important to their being locally recognized as independent and capable authorities, but in this case they must be aware of how to exercise judgement when managing direct relationships with different key actors. Some police officers also admitted that certain relationships were crucial to their opportunities to remain influential, and that these relationships could become too complex to manage. In particular, in districts known for high crime rates, officers had to be aware of police formalities and remind themselves of regulations and the importance of maintaining integrity relative to their collaborators. They appreciated their connections with local key actors, but they might, for instance, be tempted to ‘get things done’ and end up collaborating with doubtful actors. This could undermine community recognition of them as a local authority capable of maintaining law and order, by making other local groups suspicious. Police officers in rural districts appeared less concerned about challenges involving doubtful personal connections, simply because they were struggling with geographical distance and lack of organizational resources, making it difficult for them to orchestrate local networks.

Relating as brokering: maintaining a sense of local familiarity

The police officers eventually applied the relating as brokering filter when describing measures or situations in which they proved themselves able to act as relatively independent mediators, sharing information at regular meetings with local neighbourhood watch or crime prevention groups arranged by the municipalities with local schools, mosques, churches, shop owners, football clubs, etc. They often identified key individuals in their partner organizations who facilitated communication between them and different local interests. Some interviewees spoke of using closed digital platforms covering specific city districts, where they acquired information about local circumstances via instant communication with local shop owners, and other social media also emerged as means for conveying and receiving information. Briefly, they proved themselves familiar with local communities, or even specific geographical locations, by engaging in both formalized communication with collaborators and conventional police work based on informal networks and personal trust. One respondent working as a community police officer described being dependent on less formalized connections:

The formalized meetings support the official description and understanding of local circumstances, eventually communicated to all units, but we may then lack information. Thus, we also have an unofficial description of local circumstances, based on our ability to establish trust through personal connections. We gain such information due to that trust, and it is crucial to our intelligence unit, which is expected to forward it to other parts of the organization.

The quoted officer emphasized gaining personal trust in informal networks, which can be time consuming and challenging and has always been important to local police. Combined with formal procedures and meetings introduced by the reform, such informal networks serve an instrumental purpose, smoothing communication and enabling officers to acquire a more comprehensive understanding of local circumstances. However, formal meetings and functions explicitly introduced by the reform are also important in that they provide organized and stable measures that support the perception of the police as locally informed authorities familiar with circumstances in specific districts. For instance, community police officers emerged as having specifically important functions, in that they are explicitly responsible for the formal exchange of information between the police organization and organized stakeholders (e.g. social services, housing associations, and schools). One inspector explained:

The community police officers may, for instance, attend a meeting arranged by the municipality, presenting certain problems as urgent – ‘We’ve got increasing numbers of illegal settlements of addicts at certain spots in our neighbourhood. What should we do?’ Our community police officers receive this information and bring it back to us, asking if we can have a look in the parks, take pictures, simply check what’s going on. Or it might be the housing company that informs us about neighbourhoods where locals are having problems with addicts. The housing companies talk to our community police officers, who talk to our boss, eventually telling those of us who work on the streets to go to this spot and solve the problem.

Specific police officers, local police managers, and inspectors are also involved in information sharing in even more specific neighbourhoods, conveying information and managing official contacts with, for instance, neighbourhood watch groups, local housing associations, and shop owners. They act as interfaces between the police and ordinary citizens in specific areas in their police districts. Compared with the above-mentioned community officers, we can detect a division of expertise associated with the reform, intended to facilitate various types of brokering in different local settings. Because they foster concerns about the time and resources police spend on mediation, these measures nevertheless emerge as ambiguous. In particular, the interviewees noted that they had to engage in increasing numbers of meetings where they simply ended up talking to one another without any pay-off in the form of action. One police officer in charge of a specific neighbourhood explained:

This type of platform can be good if social services, the school, and the police exchange experiences openly, making it possible to identify teens and kids who have problems early on. If you don’t manage to establish open discussion, however, it’s meaningless. You might as well close down if you have someone always referring to confidentiality and that they can’t say anything. Thus, it’s important that you can establish the type of climate that allows you to be open about specific individuals. That is when you can do something, when the social services, the police, and the schools all contribute.

This quotation reflects the importance of the police maintaining recognition as capable local authorities. The interviewees repeatedly described having to consider whether there was anything to gain from these meetings, i.e. they had to avoid being perceived as police who only engage in constant talk. This could be done, for instance, by identifying individuals in partner organizations who were committed to joint tasks and were inclined to downplay organizational constraints when helping update the police. The police also mobilized opinions in networks, to establish priorities and persuade collaborators to share more valuable information. Once again, they could easily get absorbed in strategies for identifying key actors, and they had to be aware that persuasion demands time and resources that could previously have been spent on police work embedded in the local community. In addition, having to follow regulations and police priorities whenever they begin establishing connections with specific individuals made local information sharing time consuming. For officers involved in and establishing arrangements for information sharing in rural districts, geographic distance emerged as explicitly troublesome, in some cases almost impossible to surmount, even if the officers spent time devising applicable strategies. One interviewee representing a rural municipality confirmed that she increasingly engaged in strategic information sharing. She indicated that she only talked to officers higher up in the police hierarchy, discussing more general strategies and agreements, and only heard rumours that police officers sometimes spent time in the local police office.

Discussion

Here we answer the research question by looking more closely at how the three filters discussed above constitute the police officers’ awareness of themselves as capable of addressing expectations regarding cross-organizational collaboration guided by the suggested three dimensions of professionalism (Noordegraaf 2020). That is, the article continues by considering how the three filters inform the police officers’ responses to collaborators’: (1) reliance on state-of-the-art knowledge and skills in the police; (2) expectations regarding the ability to independently translate knowledge and standards to the needs and features of the case at hand; and (3) recognition of the police as an authority complying with and updating standards and codes.

We can start by concluding that the previous section illustrated how the relating as collaboration filter made the officers into what can be described as moderators. That is, the filter urged them to raise their awareness of how to maintain and temper excessive expectations regarding their role as locally present authorities in regular negotiations with their municipal collaborators (cf. Anteby, Chan, and DiBenigno 2016). In other words, the emphasis is on their ability to manage expectations regarding how to independently translate knowledge and standards to the needs and features of the case at hand. To agree on common ground for collaboration, they also had to consider how to respond to expectations regarding state-of-the-art knowledge in the police, and how to keep being recognized as a local police authority complying with standards and codes (Noordegraaf 2020). However, the relating as collaboration filter primarily made the police officers into moderators who translate their knowledge and standards to the needs encountered in local negotiations and agreements.

Concerning the relating as coproducing filter (Anteby, Chan, and DiBenigno 2016), we can also conclude that this filter made the police officers into orchestrators. It was still important to translate substantial and procedural judgements to local needs, and address normative concerns that could undermine the recognition of the police as a local authority. However, as orchestrators, the officers primarily directed their attention to coordinating networks confirming the local influence of the police’s state-of-the-art knowledge, expertise, and skills. Ultimately, the findings identify how the relating as brokering filter made the police officers into mediators engaged in accessing networks that facilitate communication (Anteby, Chan, and DiBenigno 2016). This last filter comprised the police officers’ awareness of local connections by which they could both inform themselves of the local situation and distribute information supporting local communities’ familiarity with the police. That is, the relating as brokering filter primarily directed their attention towards how to act as a locally informed authority able to follow and update standards and codes. Table 2 sums up these findings, displaying in italicized text the emphasized dimensions of professionalism in each filter (Noordegraaf 2020).

Table 2. Police occupational awareness of how to enact responsive professionalism.

The three key dimensions of professionalism	Relating as collaboration filter fostering awareness of	Relating as coproducing filter fostering awareness of	Relating as brokering filter fostering awareness of
Responding to collaborators’ reliance on state-of-the-art knowledge and skills in the police	by reaching agreements that foster common ground and formal expectations of police competence	by acting as orchestrators coordinating networks that maintain recognition of police influence and local presence	by maintaining police familiarity with the local community
Addressing expectations of the ability to translate knowledge and standards to the needs and features of the case at hand	by acting as moderators tempering negotiations and external expectations.	by responding to demands for substantial and procedural judgements in networks of collaborators	by mobilizing efficient mediation and dissemination
Responding to the recognition of the police as an authority complying with and updating standards and codes	by managing policy coordination and formal expectations of police involvement	by managing affinities for and normative constraints on local police involvement	by acting as mediators in networks distributing and providing access to useful information

The findings presented in the table indicate that the Swedish police reform has not restored older forms of police professionalism by engaging in cross-organizational collaboration. The police officers have not responded to anticipation in their environment by restoring their autonomous or exclusive position as local police representatives (Fyfe 2013; Björk 2019). The way in which the interviewees depicted themselves as moderators, orchestrators, and mediators instead indicates that their occupational awareness has been shaped by a tense combination of demands for both responsiveness and particular organizational conditions (cf. Rolandsson 2017). They emerge as knowledgeable actors who have had to relieve strains between their local engagement with collaborators and demands for organizational control. To compensate locally for the new and increasingly centralized governance of the police organization (cf. Minas, Wright, and van Berkel 2012), the police officers have attempted both to temper local collaborators’ expectations that shape recognition of their competence and authority, and to determine how to align themselves with their own organization (Kirkpatrick and Noordegraaf 2015).

We also note that how they reacted to the tension between their local engagement and organizational control unfolded differently according to each filter. In other words, the police officers’ occupational awareness addressed various challenges to being recognized as knowledgeable, independent, and locally present authorities. For instance, relative to the relating as collaboration filter, the police officers described having to temper the strain between feeling pressed to seek common ground for joint actions with local municipalities, and municipal collaborators’ expectations concerning local police capacity. Relative to the relating as coproducing filter, the police officers described not only trying to gain recognition as local and influential actors, but also having to address legal and normative concerns to avoid local policing being perceived as biased or too loyal to local collaborators. Guided by relating as brokering, the police officers eventually stressed the importance of turning information into action: they had to prove themselves capable of gaining trust as a locally informed authority, without spending too much time and resources on mediation.

The emergence of responsive professionalism in abstract policing

Swedish police reform has fostered various ways to tackle the tension between engagement in local collaboration across organizational boundaries and organizational control. Contrary to previous findings, increasing the number of collaborators, which fosters tension or ‘coordinative complexity’ (cf. Gleeson and Knights 2006; Bohle Carbonell et al. 2014), does not necessarily encourage police officers to restore a more exclusive professionalism independent of managerial constraints. Instead, their professionalism emerges as increasingly shaped by a set of distributed demands for managerial awareness. By acting as moderators, orchestrators, and mediators across organizational boundaries, police officers make managerial and wider organizational concerns into strategic considerations that form their own occupational awareness.

Strategies for aligning cross-organizational collaboration and police organizational capabilities have become important to police officers. In line with, for instance, Terpstra, Fyfe, and Salet (2019), who examined increasingly formal community engagement among local police officers exposed to similar police reforms in the Netherlands and Scotland, the present findings suggest that Swedish police officers also make their local engagement into more abstract considerations. By identifying how the police officers conceptualize their role as independent agents engaged in formal collaboration strategies, this study illustrates how they can make this type of ‘abstract policing’ into their own occupational awareness. Like their Dutch and Scottish colleagues, the Swedish police can then deflect accusations of outdated or biased local policing when complying with the reform’s explicit goal of moving closer to citizens. Still, the way police officers have enacted rather formal or constrained engagement with local communities (Terpstra and Salet 2019) also raises concerns that the so-called closer connections with the local communities primarily involve what a national evaluation of the reform described as an emphasis on ‘indirect relations’ mediated by police collaborators. Rather than being more locally embedded, the police have simply improved their familiarity with local settings via cross-organizational collaboration.

Finally, we note that this emerging demand for an indirect or ‘abstract’ mode of policing (Terpstra, Fyfe, and Salet 2019) does not mean that the police officers conceive of themselves as administrative actors, simply reducing their initiatives to habitual patterns of doing whatever is appropriate in order to realize organizational goals (Simon 1997). The present findings do not necessarily indicate the decline of professionalism, but rather illustrate how the reform has forged awareness of what we call responsive professionalism among wider groups of occupational representatives. The filters constituting the relating lens suggested by Anteby, Chan, and DiBenigno (2016) unfold in a way that coincides with Noordegraaf’s (2020) abovementioned work on connective professionalism. In addition, this means that this study has not presented just another case of hybrid professionalism enacted by professionals who have become managers (Kirkpatrick 2016). By empirically identifying how police officers throughout the police organization conceive of themselves as aware of both substantial and managerial prerequisites for professionalism in collaborative connections, the present findings instead provide concrete examples of how the recognition of expertise, autonomy, and authority is earned based on appearances in wider networks of other actors (Noordegraaf 2020). By describing how the police officers conceive of themselves as moderators, orchestrators, or mediators, and considering the tensions within a police reform that set out to ameliorate the downsides of NPM, this analysis provides a nuanced picture of what connective professionalism can mean in practice. After all, the findings imply that managerial concerns formerly associated with NPM have had a persistent impact, underscoring once again that we should treat any claims about a paradigm shift with caution (Adams et al. 2020a, 2020b).

Conclusion: fostering abstract policing in connections with the local community

By introducing inter-organizational collaboration, the Swedish police reform was intended to improve police officers’ opportunities to provide services ‘closer to citizens’ based on their skill and judgement (Bringselius 2019; SOU 2019, 43). In this way, the reform set out to ameliorate the downsides of NPM (van Berkel, Van der Aa, and Van Gestel 2010; Christensen and Lægreid 2011) caused by formalized decision making and management by objectives that erode police professionalism based on occupational expertise and ability to translate knowledge, codes, and standards to the needs and features of the case at hand (Evetts 2011). Accordingly, the present findings confirm the emergence of what Noordegraaf (2020) described as a new form of connected professionalism, dependent on occupational appearances and responsive acts in the wider environment of other actors.

This type of responsive professionalism urges police officers to be knowledgeable and gain ability to resolve coordinative complexities (Bohle Carbonell et al. 2014). However, this study concludes that police officers’ enactment of responsive professionalism through collaboration across organizational boundaries still demands awareness of managerial strategies and routines. Rather than restoring what previous research has described as occupational professionalism that is autonomous from management, or promoting occupational involvement in local citizens’ daily life, the results show that the reform has directed officers’ attention towards a set of collaborative tensions and what Terpstra, Fyfe, and Salet (2019) have described as ‘abstract policing’, which demands their engagement in organizational strategies for collaboration.

The engagement required by abstract policing implies that there are further questions for future research to investigate, for example, how police officers navigate and internalize such strategies (cf. Rose 1999). This study, however, concludes that police officers’ awareness of how to apply professionalism in responsive acts across organizational boundaries directs their efforts towards rather abstract concerns of organizational procedures and strategies. The study thereby complements Noordegraaf’s (2020) conceptualization of connected professionalism by showing how the emerging form of professionalism responds both to new demands for a more receptive occupational approach and to managerial concerns previously linked to forms of so-called organizational professionalism (Evetts 2011). This form of professionalism do not necessarily move closer to citizens’ daily lives. In addition, we note that previous studies, describing how post-NPM reforms in other parts of public sector introduces collaboration marked by coordinative complexity (Christensen and Lægreid 2011; Bohle Carbonell et al. 2014), imply that other occupations also may face demands for the type of connected professionalism that emerge from this study. The fact that other occupations encounter similar forms of cross-organizational collaboration, for instance in healthcare (cf. Bringselius 2019), suggests that contemporary reforms in public sector provide additional examples of abstract professionalism which have been shaped both by demands for a more receptive occupational approach and by distributed managerial concerns.

Finally, by identifying how a persistent managerial logic continues to shape the enactment of responsive professionalism (Adams et al. 2020a, 2020b), the present findings also indicate that such changes entail different degrees of path dependency, raising questions as to whether these reforms have really enabled organizations such as the Swedish police to move beyond NPM. As reforms claiming to move beyond NPM become more common in different parts of public sector, future research must deepen our knowledge of how and whether such increasingly abstract responsive professionalism can reconcile tense combinations of cross-organizational collaboration and managerial concerns.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by FORTE [grant number 2016-00037].