Community policing and reassurance: Three studies,one narrative

Reassuring the Community in Hard Times: Similar Narratives across UK Policing

This article is based on findings from three distinct studies, all of which were conducted independently of one another in different parts of Scotland between 2009 and 2012. The discussion is divided into four parts. First there is an introduction where the background and methods underpinning the three studies are briefly discussed. The article then reviews the development of reassurance policing within the United Kingdom. Findings from the studies are then presented. While the studies all had differing objectives, common themes emerged relating to the trajectory of both reassurance and community policing in Scotland. Finally, the implications of these findings are discussed, and related to insights gained from the wider literature on reassurance policing, community policing and procedural justice.

The three studies are as follows: (1) a project which piloted a method for gathering ‘community-level intelligence’ on crime through interviews with the public run by community police officers (‘the Strathclyde Community Intelligence Project’, undertaken by Hamilton-Smith and Mackenzie); (2) a project studying police reform in Glasgow, with an emphasis on the recent introduction of an Integrated Service Delivery Model which brings community policing and response policing closer together organizationally and strategically (‘the Glasgow project’, undertaken by Davidones); and (3) a knowledge exchange (KE) project with the city of Edinburgh police, which involved qualitative and observational research with community police officers and developed a KE forum known as the Edinburgh Police Research and Practice Group (‘the Edinburgh project’, undertaken by Mackenzie and Henry).

Although using a range of methodological tools, the basic toolbox from which these were taken is common across the studies, in the sense that the orientation of each was interpretive and inductive, and the different methods employed were all qualitative in orientation and used in order to try to access, understand, record and analyse the subjective experiences and views of police officers going through processes of organizational change. The methods used included one-to-one interviews, participant observation in programme implementation, diary-style observations recorded by officers themselves on dictaphones, focus groups, discussions at knowledge exchange events and seminars, ethnographic non-participant observation, and more. Insofar as the purpose of this article is to draw out the ‘one narrative’ we have identified running through the data emerging from the three studies, the similar ‘appreciative’ sensibilities deployed by the different research teams help to provide a basic platform from which to observe and compare the stories told by the research participants about their changing roles.

The aim of the article is to consider in both a theoretical and an empirical sense how the police ‘do’ reassurance and procedural justice in the context of community-level interactions through community policing. That is, we are interested not only in the normative question of how these ideas should be incorporated into policing, but also in practitioner evidence of how they actually are translated into day-to-day police routines, and how, in turn, these reifications are mediated through the broader contexts of contemporary policing. The ‘one narrative’ of the title describes risks for public reassurance, confidence and co-operation resulting from organizational changes to local policing in Scotland. In common with similar changes in some other jurisdictions, new financial constraints trigger processes of police service reform which unintentionally deplete important aspects of the community policing role. Principally, organizational restructuring that seeks closer integration of the two key delivery functions of reactive (response) and proactive (community) policing, with the aim of achieving more for less in times of austerity, may actually be self-defeating.

Common to all three studies forming the basis of this article is a focus on community policing (CP) and to a lesser extent, reassurance policing. The studies were also conducted at a time when British policing, like many police services internationally, was starting to confront the consequences of the global economic recession. After years of record growth (Zedner, 2006), UK policing has entered a period of contraction. Police budgets in England and Wales are expected to fall by at least 20 per cent in the current spending period (Innes, 2011: 73), with latest estimates suggesting a cut in the workforce of over 32,000 posts by 2015 (HMIC, 2012: 4). In Scotland, the move towards an amalgamated new national force (which will be in place by spring 2013) has been carried through on the back of a rationale which has been predominantly economic (Fyfe and Henry, 2012; Scottish Government, 2011), with pledges to protect frontline officer numbers being met by stripping out civilian backroom functions, including analysis. Different areas of service delivery may be disproportionately affected. In particular, some commentators fear that the cuts may erode the CP capacity of police services (Myhill and Quinton, 2010). Other commentators have argued that the need for the police to maintain efficiency and effectiveness necessitates stronger CP and engagement to ensure that services are better tailored and more responsive to community concerns (Innes, 2011). Indeed an ambition to strengthen CP on such a basis featured in all three of the policing areas examined here.

An emphasis on responsiveness to community concerns contrasts with previous drives in policing that conceptualized effectiveness and efficiency in terms of simply reducing crime (Hough, 2007). Indeed, with performance measures and incentives tending to focus on effectiveness and the technocratic efficiency of the police service (Innes, 2007), ‘softer’ community-facing aspects of policing were often overshadowed (Hough, 2003). In particular, this focus did not sit well with a general public desire to see ever more police officers, on foot, and on the street (Wakefield, 2007). The public’s desire for more traditional ‘displays’ of policing might have easily been dismissed, if repeated sweeps of the Crime Survey for England and Wales during this period had not shown a widening gap between (falling) crime levels and (declining) measures of public confidence in policing (Myhill and Quinton, 2010), with the invisibility of the police at the level of community being implicated in this declining confidence. The police themselves were quick to pick up on these divergent trends, characterizing them as ‘the reassurance gap’ (Association of Chief Police Officers (ACPO), 2001).

Central to the diagnosis of the ‘reassurance gap’ was the contention that people’s insecurities and concerns about crime were not so much influenced by their direct experiences of crime, or their assessment of crime levels, so much as by a more generalized fear about a lack of social control in their neighbourhoods (Herrington and Millie, 2006), or in wider society (Girling et al., 2000; Mackenzie et al., 2010), which was informed partially by their concerns about expressions of social disorder such as anti-social behaviour. To help reassure people, policing needed to be better attuned to local concerns and insecurities, and needed to provide a more visible presence in communities to help restore people’s confidence in the presence and strength of social controls. Central to this was a need for officers to be visible in communities, getting out on foot (Millie and Herrington, 2005) and – as emphasized in the police inspectorate’s own Open All Hours report (Povey, 2001) – being ‘visible’, ‘accessible’ and ‘familiar’.

The visible ‘bobby on the beat’ therefore became the cornerstone of what became known as ‘reassurance policing’ (RP). As Innes (2005b: 160) remarked, this was an acknowledgement that the symbolic or ‘dramaturgic’ properties of policing were of prime importance: ‘The emphasis upon reassurance recognizes that much of the popular appeal of uniform patrol and other policing activities lies in how they symbolize the presence of protection.’ This embracing of reassurance as of central importance to policing was evident in the establishment in 2003 of a National Reassurance Policing Programme in England and Wales, and with the Home Office in 2008 departing from its prior set of crime reduction targets, and instead promoting a new single national measure of ‘public confidence’ in policing as the key gauge of police effectiveness (Millie, 2010). ¹

Early criticism of reassurance policing focused on the potential superficiality of a policing approach that is centred on visibility. Reassurance would be hard to justify if it did not also help achieve other important outcomes such as reductions in crime and disorder (FitzGerald et al., 2002). Indeed it would be reasonable to hypothesize that if reassurance was pursued at the expense of such reductions, with reassurance won against a backdrop of rising crime and disorder, the association between seeing a police officer, and having confidence that they symbolized a healthy level of social control and order, would rapidly weaken.

This may beg the question as to whether, if CP is done well, a separate model of reassurance policing is required at all. However, CP models do not always emphasize police visibility, and Millie and Herrington’s (2004) concern has been that a pragmatic focus on resolving community problems may make it difficult to differentiate reassurance from other CP approaches, with this blurring resulting in reassurance becoming colonized by, and ghettoized within, CP departments. For Millie, reassurance should be a ‘golden thread’ ‘running through all police work’ (Millie, 2010: 230). CP initiatives have also often struggled to base their activities and problem solving on data of sufficient quality to allow them to identify accurately and target community concerns (Bullock and Tilley, 2003). Indeed a feature of CP has often been a relatively weak engagement with real communities (Rosenbaum, 1994; Schneider, 1999). Perhaps reflecting these weaknesses in existing community policing practices, a bespoke programme to test out the principles and practices of reassurance policing, was launched in England and Wales in October 2003.

Setting the Scene: The National Reassurance Policing Programme and Signal Crimes

The National Reassurance Policing Programme (NRPP) was a Home Office funded initiative which aimed:

The Home Office evaluation of the NRPP found that the package of NRPP policing measures was associated both with reductions in crime and anti-social behaviour, but also with reduced ‘worry’ about crime and increased public confidence in policing (Tuffin et al., 2006: xv), thus effectively demonstrating that the ‘reassurance gap’ could in fact be lessened. In line with expectations, it was also found that when the impact of different measures were disaggregated, general foot patrols and community engagement were the key drivers of increased public confidence and reduced worry, while problem solving and targeted police patrols were associated with reductions in crime and disorder.

The NRPP, in attempting to address community concerns and provide visible reassurance, was strongly guided by the work of Martin Innes and his Signal Crimes Perspective (SCP). Focusing on how people interpret risk, Innes critiques earlier criminological models that present a linear and causal relationship between local disorder, fear of crime and actual crime levels (e.g. principally the ‘broken windows’ thesis of Wilson and Kelling, 1982). Innes (2004: 341), in demonstrating that measured ‘fear of crime’ often fails to correspond to a particular level of crime, nevertheless highlights the importance of often ‘trivial’ local incivilities in informing people’s judgements about risk:

Whilst such things may be temporarily unsettling, unpleasant and damaging to an urban aesthetic, physical and social disorders only rarely pose an objective threat to security. Rather their significance lies on how they encode messages about levels of unwanted risk and social control in an area.

Borrowing from symbolic interactionist sociology and semiotics, Innes (2007) develops an approach for unpacking how people interpret crime and disorder, focusing on how a particular crime or disorder incident is ‘expressed’ (how somebody describes the crime or disorder), its effect (how it impacts on them in terms of behaviour, thinking or feelings) and finally its content (how the incident informs their sense of risk or threat). These elements together constitute a ‘signal’, and for Innes identifying, analysing and targeting the most prominent signals within a community is the key to successful reassurance policing. What is required to combat ‘signal crimes’ and ‘signal disorders’ are effectively tailored ‘control signals’, which are – as with crime and disorder signals, forms of communicative action (typically deriving from police actions) that if tailored well may have a positive impact on people’s sense of security (Innes, 2004). So, rather than focusing on crimes or disorders that may be prominent in police statistics but which may not be prominent in informing a particular community’s sense of order and control – or conversely focusing on vague measures of ‘fear of crime’ – one should focus on signal crimes and disorders, which are inherently grounded and ‘citizen focused’ (Innes, 2005a), targeting police resources on incidents that are most impactful to communities.

So SCP provides a framework for generating a particular sort of community intelligence that in turn allows the police to target their efforts on those issues that are having a disproportionate impact on a given community. However SCP is, as Innes (2007) himself stresses, a formal and systematic methodology, which might easily be simplified or corrupted by under-resourced or poorly implemented practice. Therefore, to facilitate and ensure the robust implementation of this approach, Innes and colleagues designed software and a database especially for the purpose of collecting, mapping and analysing signal crimes and disorder. This IT-steered approach, termed ‘intelligence-orientated Neighbourhood Security Interviews’ (or i-NSI) formed the basis of the Strathclyde pilot which is one of the studies reported on here.

Reassurance Policing in Scotland

In Scotland, ‘reassurance’, as a function of local policing, was given explicit recognition rather later than it was in England and Wales. The Association of Chief Police Officers in Scotland (ACPOS) published their Public Reassurance Strategy in 2007. It largely mirrored thinking about reassurance policing in England and Wales and was found to still be influential in all three research sites – particularly in its commitment to ‘mainstream the practice and ethos of public reassurance policing with engaging criminality, response policing, and other policing activity’ (ACPOS, 2007). In other words (Millie’s, for example), reassurance was understood by ACPOS as being a necessary ‘golden thread’ that should run through all policing. Nevertheless CP was given the most attention as being the obvious ‘engine’ for delivering reassurance. It quickly became apparent, however, that the nature of CP itself, not to mention whether it did or did not capture notions of reassurance, was ambiguous to say the least in the Scottish context. The Justice Committee of the Scottish Parliament conducted an inquiry into Scottish CP practice in 2008, finding examples of existing good practice, but also noting the apparent lack of a clear or consistent definition or understanding of what CP was. It was as a direct result of this inquiry that the Scottish Government published the Community Policing Engagement Principles (2009) the following year. These principles gave emphasis to visible policing, community consultation and communication, responsiveness, local accountability, partnership working and a commitment to problem solving as core components of the CP concept, again with explicit recognition of the idea that reassurance was in any case an objective of all policing. The principles were drafted as a guide to activity rather than as a set of prescriptions, in that they did not attempt to impose a single model of CP on Scottish police forces. This is apparent in the on-going variation in the form and practice of CP that is documented in the three studies examined here, although, as we shall see, officers’ own narratives about how they experience CP programmes show striking commonality of theme. ²

The Strathclyde Community Intelligence Project

Though lacking in detail as to how precisely public reassurance should be achieved, the ACPOS strategy emphasized that better engagement and communication processes were needed to facilitate effective reassurance. Strathclyde Police took the lead on this in 2010 developing – in collaboration with the Scottish Centre for Crime and Justice Research, the Scottish Institute for Policing Research and the Universities’ Police Science Institute (UPSI) at Cardiff University – a project to pilot a more systematic approach for collecting and analysing community intelligence in accordance with the signal crimes methodology.

The pilot was run in 2011–12 with the principal aims of testing whether this approach improved the capture of community information on those crimes and disorders that had the strongest ‘signal’ in terms of being key drivers of community anxiety, and to assess in turn whether this information could be usefully collated and analysed to better inform policing responses (or ‘control signals’). These aims were to be pursued through:

The data-collection element of the i-NSI software is designed to facilitate the eliciting of information in a semi-structured way, without requiring any great level of expertise on the part of the interviewer. The software at the top level is also highly flexible, allowing respondents to identify and locate incidents that are of concern to them. Respondents are not forced, in a closed conceptual format, to respond to a list of pre-defined incident types. Rather, they are free to talk about any locally grounded incident that concerns them (in SCP terminology this is ‘the expression’), and GIS (geographic information system) maps of their local area allow them to locate where an incident (or a linked series of incidents) of concern might be located, while providing quite flexible geographic markers with which to map the incident (e.g. an area of concern might be a single address point on the map, or it might be a larger defined area within a public park, or a stretch of street). In addition, the software prompts officers to try to elicit both the ‘effects’ and the ‘content’ of incidents, allowing one to assess seriousness in terms of how it impacts on respondents emotionally and behaviourally.

Inputted information can be stored and then analysed in i-NSI to identify those incidents, or geographic locations, that are prominent, or in Innes’s terminology, to identify those signals that are particularly strong or ‘coherent’. A strong signal is primarily determined by the number of times a specific type of incident, or a geographic location (or better still a combination of the two) are raised by different respondents. The software allows analysts to produce area and/or problem profiles that are in some respects similar to conventional police products of that name, namely mapping and ranking incidents and areas of particular concern. But what is unique about i-NSI is that it supports the ranking of incidents not simply on the basis of frequency of occurrence, or spatial concentration, but it also triangulates these sorts of indicators with how respondents actually feel (e.g. fear, anxiety, etc.), think and act (e.g. subsequent avoidance behaviour of a particular area). This approach ensures that policing responses are not simply based on police-constructions of local problems, or conversely on a myriad range of subjective perceptions some of which may or may not be linked to real local conditions. Rather i-NSI aims to elicit ‘information from interviewees about what problems locally are generative of personal and collective insecurity’ (Lowe and Innes, 2012: 299).

Twelve community police officers administered the i-NSI questionnaire over a period of several weeks. They received a full day’s training in the use of the software, and each officer was given a laptop with the software loaded onto it. Two further days were set aside for UPSI staff to accompany officers on initial interviews, and to help them hone their interviewing and computer skills. At the end of this period the data were passed over to police analysts, who, in addition to attending the initial training day, received two further full days of bespoke training. Running in parallel to this, the researchers were present at all the officer and analyst training events in a participant observation role. The researchers also undertook some i-NSI interviews alongside officers, and explored the experience of officers in using i-NSI through a subsequent focus group and through administering a short, anonymous, self-completed questionnaire.

In terms of the approach to sampling members of the community, officers were instructed to follow the methodology designed for i-NSI, which was:

Within the pilot areas data collection took longer than anticipated, but officers almost completed the target number of interviews, with 111 interviews being conducted across the two areas. A natural limitation of the approach was that under-16s were not interviewed, and given that community concerns about crime and incivility often focus on the activity of ‘young people’ the exclusion of a large proportion of that group clearly needs to be recognized. The gender split in the sample showed a slight skew towards male respondents, with officers being reluctant to interview female respondents in their own homes. One final limitation of the sample was that the officers were not able to interview evenly across the geographic-units, and 11 output areas (out of a total of 52) had no interviews conducted in them at all, whereas a corresponding number were over-sampled.

In terms of the analysed results the pilot interviews threw up a fairly well-known set of community issues, ³ with the focus tending to be locational, with multiple concerns clustering around these locations, rather than being problem-specific. The types of incidents also tended to be mostly low-level, focusing on incivilities such as litter, youths hanging around and threatening behaviour, though under-pinning these issues in both sub-areas were concerns with drug dealing and more serious forms of public violence. The level of associated detail was not particularly promising in terms of facilitating the sorts of targeted problem solving envisioned by Innes though this may, in part, be put down to the pilot status of the project, with officers having to familiarize themselves with the methodology and the technology.

Overall, officers did believe, based on their own existing local knowledge, that the information collected was valid, and that the methodology did reliably capture the emotions, thinking and actions associated with identified crimes and incivilities. Yet in spite of this, officers held generally negative views of the i-NSI software. This is not in itself particularly noteworthy given that the introduction of new IT technologies in the police or other organizations are rarely greeted warmly (Manning, 1992). However, officers had specific concerns about why the i-NSI approach, particularly within the context of organizational reforms that were underway in Strathclyde at that time, did not represent an improvement over pre-existing CP approaches. Briefly these concerns centred on the following issues:

Against this backdrop, a number of experienced community officers held the view that the i-NSI method represented a solution to a problem in-the-making, namely that once established (and by implication more substantive) CP contacts and structures had been dismantled, i-NSI would serve deskilled community officers well in ensuring some level of structured contact, and some targeted level of (albeit superficial) police response, while at the same time generating the sorts of visual and quantifiable indicators of community concerns that would satisfy police performance managers. This was not, to be fair, at all how the developers of the software intended it to be used (see Lowe and Innes, 2012), but how officers represented it as being likely to be used against the backdrop of a claimed deterioration in the CP role.

Police Reorganization in Glasgow and Edinburgh

The second and third studies, undertaken independently of the first, were rather more ‘general’ studies of developments in CP in Scotland between 2009 and 2011. Their findings sit comfortably nevertheless with those of the i-NSI pilot, providing more detailed evidence that supports and extends the observations of the pilot site officers on the pressures currently shaping CP practice. The second study was a small exploratory project focused on the impact of the recession on policing, based in three police stations within a different policing division in Strathclyde to that of the i-NSI pilot (this division hereafter referred to as ‘Glasgow’, although it is one of many divisions within the city). In particular it examined how community and response police officers were adapting to a range of organizational changes, all of which were driven in part by the need to police more efficiently. In addition to reviewing relevant police policy documentation, the study involved 20 in-depth semi-structured interviews with a purposively selected sample of police staff (four of whom were civilian staff), of mixed ranks and different lengths of service. The third study was a three year Knowledge Exchange programme, funded by the Arts and Humanities Research Council (AHRC), which established close lines of on-going communication between an academic research team and community police personnel at all levels from beat officer up to divisional commander in a division of the Lothian and Borders (L&B) police covering parts of the city of Edinburgh (this division hereafter referred to as ‘Edinburgh’). It quickly became apparent to the academic team that in order to explore possibilities for knowledge exchange in the context of CP they had to understand the everyday practice of CP, what it meant to officers and the kinds of tacit knowledge and skills that they valued and deployed. In short, we had to understand what CP actually ‘was’ in practice, and this required us to undertake research. This took various forms: focus group meetings with different interested constituencies of police officer (including strategic management, operational managers and CP officers themselves); individual interviews with officers; and the use of Dictaphone Diary interviews with CP practitioners to explore the detail of their working lives through their own words (see Henry and Mackenzie, 2012 for further details).

The current financial challenges faced by Strathclyde and L&B police are common to police forces across the UK. The Justice Committee – Draft Budget Review planned for budget cuts of ‘24%, in real terms, over the four years of the spending review’ (ACPOS, 2010: 2). The review recognized that this could not be achieved without substantial reductions in police staff, thereby potentially impacting on both police effectiveness and public confidence in policing (ACPOS, 2010: 2). Strathclyde police, for example, is Scotland’s largest police force and while its net revenue budget for 2010/11 was £443 million, it is anticipated that this will fall to £420 million by 2013/14 (Strathclyde Police Authority, 2010). Consistent with Neyroud’s (2010: 1) argument that ‘protecting the frontline’ has become a major political theme in policing in the UK, Strathclyde has attempted to focus on achieving efficiencies in ‘back office’ functions rather than cutting frontline policing. In keeping with its commitment to Scotland’s public reassurance strategy Strathclyde police has also attempted to increase the numbers of officers deployed within communities in terms of high visibility policing, and the key vehicle for achieving this has been the introduction of the force’s ‘Integrated Service Delivery Model’ (ISDM) in October 2010. L&B has taken a similar approach, with their model of ‘Safer Neighbourhood Teams’ which in some respects implement a similar model of role amalgamation to the Strathclyde ISDM model described below, while formally retaining a distinct and separate specialist CP role.

The principle behind the ISDM is to have two specialist roles within the same team – CP and response officers no longer function as separate units, but have points of integration for briefing, cross-coverage for absences and adherence to the same shift patterns, shared administration tasks and co-location where possible. Where they differ is that response officers will respond to priority calls within a designated area operating from strategic bases covering a whole division. CP will take on a partial response role and respond to the lesser priority calls within their local geographical areas either on foot or cycle patrol (described by community police officers in our research as ‘slow response’ policing); they also still take the lead on delivery of the public reassurance strategy (Strathclyde Police, 2011). Creating more flexible, pooled resources within a single team is intended to reduce the number of senior supervisory officers required (thereby making savings that allow frontline officer jobs to be protected), while ensuring that frontline police resources are used as efficiently as possible, maximizing their responsiveness, visibility and effectiveness. The model, far from intending to diminish CP resources, in fact on paper dedicates more officers to CP duties, while reducing the size of response teams (albeit with CP teams having that new partial response role).

The changeover to the ISDM, however, has not always been met with enthusiasm. Some CP officers complained that although on paper they had more officers out ‘walking’, most of them were probationers, while experienced officers, and especially those with skills such as response drivers, were sent to shore up response policing now that their numbers had been cut. Some CP officers thought that placing so many probationers in CP would have a detrimental effect not only on the effectiveness of CP teams, but also with regards to the passing on of knowledge. For example, in the past tutors usually had at least five or six years’ service, whereas now they were averaging two, and as one officer argued ‘they can’t possibly know the job to pass on that knowledge’ (Glasgow officer). ⁴ This perceived skills deficit was exacerbated by the tendency for probationers, having gained some minimal experience of CP work, to then move quickly on to response or other supposedly more exciting units.

Echoing comments made by officers in the community intelligence pilot, CP officers complained that the content of their work was being distorted as the force under ISDM had become increasingly performance driven. For them it was not always possible, or desirable, to issue fixed penalty notices or conduct stop searches in the course of conducting CP duties: ‘If you just have a reactive approach and stop and search everybody in the street – all you’re going to do is put everybody off the police’ (Glasgow officer). This situation was seen as being exacerbated to a degree by the greater number of probationers in CP who lacked the experience of dealing with the public, as one CP officer argued: ‘Unfortunately some of the newer cops that are coming in are just doing people for anything, there’s no discretion’ (Glasgow officer). CP officers also argued that although on paper they had more officers out walking, they now covered a bigger geographical area and still had to contend with abstractions alongside taking on a partial response role by responding to lesser priority calls.

Despite apparently different models of CP on the face of it, in terms of what beat officers report about the impact of recent changes on CP on the ground, the situation is markedly similar in Edinburgh. There too, community beat officers who were previously allocated to ‘their’ area on a one-officer-one-beat model have been reorganized into larger teams working across areas that encompass several of the old beats. The increase in numbers has been achieved by collapsing Community Beat Officer (CBO) and Neighbourhood Action Units (comprised of officers funded by the local council) into Safer Neighbourhood Teams (SNT) along with additional resource taken from the ranks of response teams. This apparent rise in community officers disguises the fact that with diminished response teams, more community officer time is taken up with dealing with the response calls that the ‘response officers’ cannot now handle: ‘They should have bolstered the (response) teams so that they could answer all the important calls, and left the CBOs to do what they were doing’ (Edinburgh officer).

Similar testimony was made in Edinburgh as has been outlined above in Glasgow in relation to the destructive effects of recent restructuring on the institutional memory which can be so important to community policing, where new recruits would ideally inherit a manageable sized beat from an experienced officer who had spent many years engaged with the local community and was therefore well placed, and had the time, to pass on detailed local knowledge. Now resource and restructuring pressures mean new community police officers get very little mentoring or other information transfer from their predecessors, if any at all: ‘it’s pretty much a case of just get on with it, and find out as you go along’ (Edinburgh focus group).

In Strathclyde police a senior officer acknowledged that there was a perception that the amount of time ‘cops’ would be able to dedicate to ‘public reassurance, community engagement type stuff, could be significantly reduced because they were answering more calls’ (Glasgow officer), but the officer argued that with a substantial increase in CP resources delivered under ISDM, this should not detract from being able to engage with the community. Conversely, response officers complained about having to cover bigger geographical areas with fewer resources, fearing it could impact on response times. However, as was reported in Edinburgh, the view from the ground was that stretching of response policing resources, far from resulting in a real gain for CP resources, had resulted in CP officers – in the absence of the ring-fencing that used to protect them in their role – having to be generally available to pick up those incidents that the depleted response teams could not handle:

We were left to our own devices quite a bit to deal with the community and target the problems that people brought to our attention. We would set up our own action plans to target those things, we didn’t get the same amount of calls, the core shift took most of the brunt of the calls. (Glasgow officer)

Now officers found that their ‘chasing of the radio’ made it difficult to engage with the general public: ‘by changing that I think we’ve lost the community, well liaison with the community on a regular basis’ (Glasgow officer).

The creation of single team was seen in both Forces as having substantially removed the degree of autonomy that CP officers had for carving out their role, with CP resources being diverted into response work. This seems to reflect a general requirement in policing to prioritize emergency reactive work over other forms of proactive work. The demands for reactive policing can be characterized as inexhaustible (Hope, 1995), and without a firewall to protect CP officers from having to respond to all but the most serious emergencies, the pressure to bolster the reactive performance of the wider ‘team’ may be irresistible. This pressure is likely to be all the greater given those deep-rooted, and well-documented, aspects of police occupational culture, which display a preference for action, with CP concomitantly often not being seen as ‘real police work’ (Paoline, 2004; Reiner, 2010; Westmarland, 2008). The Divisions studied were certainly no exception, with CP officers stating they are treated as ‘cannon fodder’ (Glasgow officer) and viewed as ‘lazy’ (Glasgow officer) and that they were somewhat resigned to this role as second class citizens in the force: ‘response officers don’t, whatever you say, value the community officer role, and it’s only once you’ve done the community officer role that you actually see how valuable that it is’ (Edinburgh officer). A related issue that emerged strongly in Edinburgh was that CP officers felt that the organization’s preference for quantifiable performance indicators further undermined the perceived value of more amorphous community engagement and problem-solving tasks that for them were crucial to the CP role (Harkin, 2011).

In spite of these apparent concerns about the effects of recent CP reforms, officers were nonetheless supportive of some of the under-pinning ideas, in particular the notion of breaking down the barriers (both ‘practical’ and ‘cultural’) between response work and more proactive policing. Moreover, both the Glasgow and the Edinburgh reforms were still in their infancy, and some of the observed problems may in time be resolved as the initiatives bed down. That said the research clearly highlights some potential problems with the amalgamation approaches represented by ISDM and SNT. In particular, the emphasis on performance and performance indicators does not sit easily with a commitment to supporting more substantive CP work (Collier, 2006; Golding and Savage, 2008; Hough, 2007). Performance measures, such as stops and searches, response times and various forms of ‘tickets issued’ all reinforce the more reactive elements of police work. Moreover, while both models emphasize the importance of visibility and the links to reassurance, taken together with other elements of the models, this conception of visibility may rapidly collapse into little more than an extension of response policing in lieu of the more substantive sort of contact associated with CP. As one senior officer commented, ‘What they want (the public) is cops on the street; they want yellow jackets on the street, that reassuring presence’ (Glasgow officer).

Discussion

The findings from these three pieces of research outline local issues that may, if they prove more broadly generalizable, be indicative of a trend that threatens to undermine the integrity of CP and RP approaches. Certainly there is evidence from other jurisdictions of similar pressures and distortions. For instance, Terpstra (2009) conducted a study of CP officers working in three different forces in the Netherlands, who have a similar CP model to the UK, and found that their role had become more ambiguous in recent years. For example, there is an increased emphasis on ‘core business’ tasks and more time being spent on ‘real police work’ at the expense of CP. In practice they had to cover a large area and 50–70 per cent of their time was devoted to other activities including administration. They, much like officers in the UK, were being pushed to meet performance targets, which subsequently affected the amount of time officers spent on those elements of policing associated with the CP model and reassurance, such as community engagement and high visibility policing. Similarly, Peterson (2010) documented an attempt to import reassurance policing to three different police areas within Stockholm. Here too, a range of problems were experienced. Peterson (2010: 41) noted that ‘the core response functions continue to hold a dominant sway over what is considered “real” police work’. As a consequence, community work often lost out in a ‘tug-of-war’ for prioritization and resourcing. For instance CP officers in one of the areas, who as in Strathclyde were co-located with response officers in the same station, tended to find themselves pulled over onto response duties. In another area, visibility and community ‘contact’ had become increasingly characterized by a fairly simple uniformed police presence with engagement often being limited to the issuing of tickets.

This returns us to the issue of what sort of police activity is required to provide reassurance and engender confidence in the police. Is ‘visibility’, albeit targeted at the right areas at the right times, enough to constitute a ‘reassuring contact’ with the police? Povey (2001) has previously argued that simple visibility is not sufficient and that more meaningful contact is required, including the need for familiar and accessible officers who are known in communities rather than just being ‘scarecrows’. Indeed, evidence from the piloting stage of NRPP demonstrated that not only was visibility not enough, but it could in fact generate insecurity rather than reassurance: ‘too much police visibility conveyed a message that there must be a problem in the area to warrant such a police presence’ (Innes, 2007: 136).

Again, the NRPP evaluation demonstrated that it was not simply visibility in terms of foot patrols that was required. Meaningful contact both in terms of general engagement with the community, and contact between the police and crime victims, was associated with increased confidence (Tuffin et al., 2006). Here the literature on procedural justice provides evidence on precisely what constitutes the sort of quality contact that is most likely to bolster confidence in the police, and more generally enhance their legitimacy. An extensive programme of empirical research in both the USA and the UK has demonstrated that ‘contact matters, and things can go wrong for the police as well as right’ (Bradford, 2010: 14).

Procedural justice models hold that public confidence in policing relates less to any assessment of their performance, or the services they provide (Bradford, 2010), but rather to their success in fulfilling an expressive function, namely conveying ‘images of order, justice and stability’ (Jackson and Bradford, 2009: 497). In terms of the quality of contact, the police treating people fairly and with respect and following correct procedures transparently (Bradford et al., 2009; Sunshine and Tyler, 2003) are all associated with increased confidence in policing. Analysis of UK data in the form of the Metropolitan Police Services Crime Victim Survey also adds an important measure of how people rate a contact in terms of how reassuring they find it, with Bradford’s analysis of the dataset showing that reassurance was an important element of an individual’s overall rating of the quality of contact, and with the quality of contact in turn informing an individual’s overall confidence in the police (Bradford, 2010).

While much of the available sociological literature would emphasize that individual insecurities about social order are not solely rooted in local experiences of crime and risk, but often relate to broader judgements about the ‘state of society’ (e.g. Hough, 2007; Loader, 2006), SCP is premised on the contention that local disorder does inform an individual’s sense of insecurity. Therefore, if one accurately identifies local drivers of insecurity and targets reassurance efforts at these drivers, the level of reassurance should be maximized and confidence in local policing increased. This premise has been lent some empirical support by Myhill and Beak’s (2008) re-analysis of British Crime Survey data that showed that community engagement and problem solving are independently associated with greater confidence in policing. However, in the view of Jackson and Sunshine (2007: 230), even if local signals are not always influential in informing a community’s sense of insecurity, RP-type policing strategies that adhere to the maxims of procedural justice may still impact positively because they may

persuade members of the public that the police share their concerns, that the police are a strong and active symbol of the morals and values that underpin community life. This can be achieved by exercising authority in a manner perceived to be fair […] and by re-engaging as an active, visible and accessible part of community life.

Returning to our data from the i-NSI pilot, the immediate concern would be that quality of contact was not at the forefront of the reassurance strategy. On the contrary, superficial visibility, aligned if anything with increasingly adversarial contact with certain members of the community, was seen as the direction of travel. Such a trajectory was clearly not part of the vision for the force’s ISDM reforms, but resource pressures combined with strong cultural preferences for more response-type policing appeared to be pulling them in this direction regardless.

Officers in the i-NSI pilot were aware of the constraints that the ISDM reforms were placing on their practices in the community. Moreover, they doubted their current capacity to engage in substantive problem solving even if key community concerns were identified. Their capacity both to engage in problem solving ‘in-house’, and in partnership with others, was under threat from resource constraints and the demands of response policing. Beyond the pressure to prioritize response duties, systematic reforms restricting the availability of overtime, coupled with a programme of large scale civilian redundancies, was squeezing capacity. The contraction of any capacity to problem-solve in partnership may seem counter-intuitive, in so far as one might expect that partnership working would offer an attractive route to achieve efficiencies during a recession (e.g. through pooling resources), but in effect the challenges presented by recession may have the opposite effect, pushing services to focus back on ‘core business’ at the expense of partnership. A recent survey undertaken by the UK Drug Policy Commission into local partnership working would seem to support this possibility. The survey of local police divisions and force headquarters, encompassing 29 of England’s 39 police forces, found that: ‘Uncertainty about partner agencies is high and less partnership working, and work with community groups is expected. This is of concern given the evidence of the importance of partnership working and community engagement for effective drug-related policing’ (Beck, 2011: 1).

Given that a key Westminster government strategy for weathering the recession in the face of significant public sector expenditure cuts is precisely to rely more on basing service delivery on partnership work with NGOs and community groups (the so called ‘big society’ initiative, see Norman, 2011) this finding is problematic. Equally problematic for Scottish CP is a further insight provided by procedural justice research, namely that good quality police contact not only increases support among the public for the police, it also increases compliance and co-operation (Skogan, 2006; Sunshine and Tyler, 2003). Thus, recessionary pressures that result in poorer quality police–public interactions may, in turn, undermine the very foundations of any ‘big society’ push to address problems of crime and disorder in partnership with the public.

In conclusion, whether unintended or intuitive, it would appear that the police’s organizational adaptations to the recession in the studies reviewed here may be self-defeating in terms of what are certainly sincerely professed goals for supporting the resources and capacity of community and reassurance policing. Paying closer attention to the lessons contained in the relevant bodies of reassurance and procedural justice research – particularly the evidence that enhancing the quality of police–public encounters can produce substantial dividends in measures of public confidence – may be the most practical way forward for getting the best out of public-facing CP initiatives that will continue to operate in a climate of fiscal restraint for some time to come.