Introduction

Third Party Policing (TPP) involves a partnership between the police (as the first party) and an external entity (the third party) where the legal powers of the third party are used to prevent or control a crime problem (Buerger and Mazerolle 1998; Mazerolle et al. 2013; Mazerolle and Ransley 2005). It is an approach that expands, and potentially optimizes, the capacity of police to target and focus their resources on the geographic, situational, and/or individual factors that underlie crime and disorder problems. In times of fiscal restraint (Ayling et al. 2009; Independent Police Commission 2013), TPP is a policing approach that is, I argue, sustainable over a long period of time and a cost-effective way to control crime.

The importance of considering and measuring the long-term effects of crime prevention and policing is gaining momentum. It is not surprising, therefore, that the Journal of Experimental Criminology recently published a collection of papers scrutinizing the long-term outcomes of a range of different programs evaluated under experimental conditions (see Farrington and MacKenzie 2013). To date, 15 criminological experiments have published long-term (10 years or more) follow-up results (see also Farrington and Welsh 2013). Only one policing experiment—the Milwaukee Domestic Violence Experiment (MilDVE)—offers long-term (10 years or more) follow-up results (see Sherman and Harris 2013a, b), and only four other policing experiments have published (to date) results of experiments with follow-up periods of more than 2 years.

The lack of studies that explore the long-term impact of policing interventions is a clear knowledge gap. Joan McCord, some 50 years ago, recognized the importance of longitudinal follow-up in social experiments. Yet, in policing, long-term follow-up is rare. In this paper, I use the example of the ABILITY Truancy Trial to explicate the characteristics and mechanisms of TPP as a long-term crime control strategy. I describe the longitudinal nature of the trial and argue that a partnership approach to policing is one way that police can sustain short-term crime control gains over the long run.

I begin the paper with a brief overview of Joan McCord’s contributions to experimental criminology. Joan’s legacy is recognized in the Joan McCord Award of the Academy of Experimental Criminology (AEC), which gave inspiration for this paper. I then review the five policing experiments with long-term follow-up results and suggest that partnership approaches in policing might offer an opportunity for police to sustain their short-term gains. I unpack the theoretical model of TPP and then describe the ABILITY Truancy Trial. I conclude by arguing that police, and policing scholars, need to think beyond the short-term suppression of crime problems and start devising ways to imagine and sustain long-term gains.

Criminological experiments and long-term follow-ups

The Joan McCord Award of the AEC serves to honor Joan McCord (President of AEC 2003–2004), and her legacy left after her passing in 2004. The award criteria comprise five parts recognizing the recipient’s (1) contributions to research in experimental criminology, (2) commitment to experimental studies, (3) contributions to policy and practice, (4) contributions to the development of younger colleagues, and (5) work done in the ‘spirit’ of Joan’s legacy, including a multidisciplinary approach to research, advocacy of true randomized experimental designs, a focus on measurement, a commitment to longitudinal follow-up, and the bringing of a ‘spirited approach’ to scientific inquiry. These criteria provided inspiration for the award lecture I gave as the 2013 recipient of the Joan McCord Award.

Joan McCord (1930–2004) was a professor of criminology, the first female president of the American Society of Criminology, and a staunch advocate of longitudinal research as a way of ascertaining the long-term impacts of interventions. In 1957, Joan and her then husband, William McCord, became interested in understanding the long-term effects of the Cambridge-Somerville Youth Study. Richard Clark Cabot initiated the project in 1935 when he, and a number of partners (e.g., schools, police, welfare agencies), identified and recruited 650 “difficult and average” young boys from Boston’s Cambridge-Somerville area (McCord 1978, 1992). In 1939, the boys (mean age 10.5 years) were randomly allocated within matched pairs to equal-sized treatment or control conditions. The treatment is described in the published literature as involving ‘professional attention’ via counseling services, tutoring, medical services, and community programs (e.g., summer camp, Scouts, YMCA) with treatment continuing for an average of 5 years (McCord 1978; Powers and Witmer 1951).

Joan’s longitudinal follow-up is described in a wide range of published books and articles, notably William and Joan McCord’s (1959) Origins of Crime: A New Evaluation of the Cambridge-Somerville Youth Study and Joan’s 1978 American Psychologist article: “A Thirty Year Follow-up of Treatment Effects”. Although initial follow-ups of the boys did not reveal significant differences between the treatment and control groups (McCord and McCord 1959; Powers and Witmer 1951), Joan found that the program had increasingly negative effects over time. Although treatment boys self-reported positive impacts of the program, comparisons of the treatment and control group boys on official outcome data told a different story. Boys in the treatment group had significantly higher incidences of crime, physical and mental health problems, death, and lower occupational status and job satisfaction than boys assigned to the control group. In later analyses, Joan found that the adverse effects were concentrated in the families who cooperated with the treatment and that, as the intensity and duration of the treatment increased, so too did the adverse effects (McCord 1992, 2003). These longitudinal results of the Cambridge-Somerville Youth Study were unexpected and lie at the heart of Joan’s legacy around the importance of assessing the impact of experimental interventions in the long term. While some would argue that the experiment was a failure, Joan argued in 1992:

…the study should be considered a success. It was a success because: It showed the importance of using random assignment to treatment and control groups in order to assess the validity of cherished beliefs about helping others. Despite good intentions, iatrogenic effects occurred…It showed that careful records collected in the process of providing treatment can yield scientifically valuable information…It demonstrated that intervention can have long-term effects (p. 204).

Other criminologists similarly argue for longitudinal follow-up in experimental criminology. For example, David Farrington (recipient of the 2005 Joan McCord Award and 2013 Stockholm Criminology Prize) has, for many years, advocated the importance of longitudinal follow-ups in experiments (e.g., Farrington 1992, 2006). Just recently, Farrington co-edited (with Doris MacKenzie, the 2011 recipient of the Joan McCord award) a special issue of the Journal of Experimental Criminology (JOEX) solely focused on the longitudinal results of a number of criminological experiments (see Vol. 9, Issue 4). From the JOEX introductory article (see Farrington and MacKenzie 2013) and a book chapter by Farrington and Welsh (2013), we know that just 15 randomized controlled trials in criminology have ever reported long-term follow-ups of offending 10 or more years after the experimental intervention. In perusing the results of this small collection of studies, it is clear why longitudinal follow-ups are so important for understanding the long-term impacts of experimental interventions in criminology. For example, Sciandra et al.’s (2013) 10-year follow-up of the Moving to Opportunity Experiment reveals that the impact of randomly offering families a housing voucher to enable relocation to a low-poverty neighborhood has varying impacts on young males’ violent and property crime arrests over time. Two years after random assignment, Kling et al. (2005) reported a significant reduction in violent crime arrests for males receiving the experimental treatment, an effect that became non-significant 4–7 years after baseline measures. Using arrest records 10 years after random assignment, Sciandra and his colleagues found that the differences between youths in the experimental and control conditions continued to narrow.

Other longitudinal follow-ups of criminological experiments have demonstrated that early intervention approaches can sustain crime control gains over many years (e.g., see Schweinhart 2013; Vitaro et al. 2013). Sometimes, however, these gains are exposed only when researchers expand their repertoire of outcome measurement. In the Perry Preschool Program, for example, young African American children were randomly assigned to either a control group or a preschool enrichment program. Initially, the intervention achieved its intended aim by improving the IQ of the children in the experimental group, but when the effects decayed as the children reached age 8, the intervention was considered a ‘failure’ (Westinghouse Learning Corporation and Ohio University 1969). In contrast to the Cambridge-Somerville study, however, the benefits of the intervention emerged over time and were uncovered only after the follow-ups were expanded to include alternate outcome measures (see Farrington and Welsh 2013 for a review). In the Perry Preschool example, children assigned to the preschool program, when assessed at age 40, were significantly more likely to have tertiary education, be employed, and have higher earnings (Schweinhart et al. 2005). In terms of offending, by age 40, those assigned to the preschool program had significantly fewer arrests for drug, property, and violent crimes than those in the control group. Moreover, cost-benefit analyses reveal that 88 % of the impressive monetary benefits of the Perry Preschool Program could be attributed to the program’s effect on crime prevention (see Schweinhart 2013).

Unlike longitudinal experimental studies of early intervention programs, only a handful of policing experiments offer insight into the long-term effects of police interventions. The Milwaukee Domestic Violence Experiment (MilDVE) is, to date, the only policing experiment with follow-ups of offending measures after 10 or more years (see Sherman and Harris 2013a, b). Evaluations of the short-term effects of MilDVE, published in the early 1990s, showed short-term benefits (30 days post-random allocation) of the experimental mandatory arrest intervention. Mandatory arrest by police of misdemeanor domestic violence perpetrators significantly reduced repeat domestic violence incidents compared to those in either of the two control conditions (warning or ‘short arrest’) (Sherman et al. 1991). Yet, in the medium term (6–12 months post-arrest), cases assigned to the mandatory arrest condition showed a significant escalation in domestic violence incidents (Sherman et al. 1991). Explaining these backfire effects, Sherman and colleagues (1992) showed that the impact of mandatory arrest in the medium term varied by subgroups, whereby a deterrent effect persisted for white, employed, and educated offenders who tended to have higher stakes in conformity.

At the point of their 23-year follow-up, Sherman and Harris (2013b) report that mandatory arrest “has no long-term, discernible effect on repeat domestic violence” (p. 493), irrespective of offenders’ stakes in conformity. Moreover, by expanding the outcome measures to death records, Sherman and Harris (2013b) show that the MilDVE may, in fact, have contributed to long-term harms: suspects assigned to the mandatory arrest condition are three times more likely to have been murdered than offenders in the ‘warning’ control condition. Sherman and Harris (2013b) also illustrate how longitudinal follow-ups can uncover ‘sleeper effects’: They demonstrate that “even a 20-year follow-up period would have failed to detect an important difference in suspect homicide victimization” (p. 504).

Longitudinal experimental studies in criminology collectively highlight three things. First, longitudinal follow-ups of policing experiments are very rare (n = 5) in the criminological literature. Second, longitudinal experimental follow-ups are clearly very important for gaining a holistic understanding of the impact of criminological interventions. If Schweinhart and colleagues (2005; 2013) had not conducted longitudinal follow-ups, we would not know that early intervention approaches, such as the Perry Preschool Program, can generate long-term crime control gains. If Sherman and Harris (2013a, b) had not conducted their 23-year follow-up of the MilDVE, we would not understand the backfire effects of police arresting domestic violence perpetrators. Third, it is not enough to just examine the impact of a program: Collecting a wide range of outcomes measures as well as studying these outcomes over time is important for building a comprehensive understanding of the longitudinal impact of criminological interventions. Sherman and Harris’s (2013b) long-term examination of MilDVE death records, for example, highlights the importance of looking broadly to assess the long-term potential harms of police interventions. The Perry Preschool experimental evaluation shows how looking broadly revealed the long-term benefits of the program.

Policing experiments and the lack of long-term follow-ups

Significant reviews of the policing literature conducted under the auspices of the National Academies of Sciences (see National Research Council 2004; Weisburd and Eck 2004), the Campbell Collaboration (see Telep and Weisburd in press), Sherman’s recent Crime and Justice essay (see Sherman 2013), and the ‘What Works’ Maryland Report (see Sherman and Eck 1997; Sherman et al. 2002), as well as the variety of studies assessing the contributions of police to the worldwide crime drop phenomenon of the 1990s and into the 2000s (e.g., see Eck and Maguire 2006; Levitt 2004; Weisburd et al. 2013), identify ways that police can be effective in suppressing, controlling, and preventing crime and disorder. Yet, many of these short-term effective practices do not last over time. This may or may not be an issue; it is plausible that there is value in policing interventions that simply contain or suppress short-term spikes in crime or disorder problems with no resolve to influence change over time. However, even two of the most popular of policing interventions—hot-spots policing and foot patrols—offer limited benefit beyond the initial suppression of the crime problem. We know, for example, that violent crime problems reemerged when foot patrols were taken away at the end of the Philadelphia Foot Patrol Experiment (see Sorg et al. 2013). In hot-spots policing, we know that crime and disorder problems ‘bounce back’ after police attend hot spots for 10–15 min during directed patrols (see Koper 1995).

The failure of police to sustain even their short-term suppressive benefits is even more limiting, given that reviews of the experimental policing literature (see Braga et al. 2013; Lum et al. 2011, 2013) show that follow-ups of policing experiments rarely extend beyond 6–12 months. Sherman and Harris’s (2013a, b) 23-year follow-up of the MilDVE is the only policing RCT with a follow-up period of more than 10 years. Another four policing RCTs include follow-ups of more than 2 years (but less than 10 years). These follow-up studies in policing experiments are instructive and deserve brief mention here.

First, Rose and Hamilton’s (1970) 30-month follow-up of the Manchester Juvenile Cautioning and Supervision experiment revealed no significant differences between supervised and cautioned boys in terms of recidivism across 6-monthly intervals from the time of random assignment. Second, Klein’s (1986) follow-up of a juvenile offender diversion experiment found statistically significant differences between the experimental and control conditions at 6, 15, and 27 months after random assignment. Specifically, a smaller percentage of juveniles assigned to the arrest-and-release condition were rearrested at 6, 15, and 27 months in comparison to the referral or judicial petition conditions. Third, Sloboda et al.’s (2009) follow-up of the ‘Take Charge of Your Life’ Drug Abuse Resistance Education (DARE) RCT, using self-report data collected from students at baseline and annually between grades 7 and 11, found no statistically significant differences between the experimental and control groups on measures of underage alcohol consumption or marijuana use. The program did, however, have a significantly positive effect for youths who reported using marijuana at the baseline measure. Fourth, Strang’s (2012a, b) preliminary 10-year follow-up of the Reintegrative Shaming Experiment (RISE) conducted in Canberra (Australia) revealed some long-term positive results of police-led restorative justice conferences (RJC) for both victims and offenders. For example, substantially more offenders assigned to the RJC condition reported that they were pleased with how their cases were handled, that they were able to repair the harms of their crime, and that their case processing enabled them “to obey the law in the future” (Strang 2012b). Tyler et al.’s (2007) 4-year follow-up of the RISE drinking and driving experiment revealed that police-led restorative justice conferencing did not significantly reduce recidivism compared to the standard judicial process. However, offenders who participated in a restorative justice conference were significantly more likely to perceive the law as legitimate at the 2-year follow-up, and have beliefs in legitimacy, and those who perceived laws as “more legitimate” were significantly less likely to have recidivated at the 4-year follow-up. So far, however, the RISE experiments have not reported long-term crime-based outcomes.

Despite the difficulties police have in sustaining crime suppression gains, statements about the role of police in contemporary society typically imply that the police have (or should have) responsibility for sustaining crime control benefits over the long run. The Report of the Independent Police Commission (Policing for a Better Britain; released in late 2013), for example, explicates a long-term, social justice role for police in preventing and suppressing crime. If reforms are adopted throughout Britain as a result of the Commission Report, police would be expected to build social cohesion and legitimacy, forge collaborations and partnerships, and consider how short-term interventions might play out in the long run. For British policing, the social justice model of policing is no longer aspirational, but rather a foundation for policing at a time when “…the police have experienced sharp budget cuts and face a period of fiscal constraint that is likely to continue for the foreseeable future” (p. 13). Weisburd and Telep (2014) similarly argue that “…causal mechanisms underlying developmental patterns of crime at a place…suggest that strategies that focus on long-term social change should be added to the tool box of crime prevention at places” (Weisburd and Telep 2014, p. 13).

Clearly, however, the dearth of longitudinal follow-ups in policing experiments limits our capacity to fully assess the long-term effectiveness of policing interventions. The dearth of longitudinal follow-ups in the experimental policing literature, in combination with Sherman and Harris’s (2013a, b) unexpected results from the Milwaukee Domestic Violence Experiment, highlights the urgent need for longitudinal follow-ups of policing experiments. Currently, our corpus of policing experiments is unable to assess which types of policing interventions work in the long term, and which ones do not. The policing literature is silent on how short- and long-term impacts might differ. Moreover, the range of data collected in evaluations of policing interventions is generally inadequate to assess a broad range of potential long-term outcomes, typically relying on official administrative data and rarely on attitudinal or behavioral outcomes. In short, experimental criminologists, particularly in the field of policing interventions, still have a lot to learn from Joan McCord’s legacy.

Third Party Policing

Partnerships in policing are not new. Police have always sought out and formed partnerships with a range of different entities to tackle a myriad of different types of problems. What is new for police is the reformation of Police Acts that explicate the collaborative nature of police work. The Police and Fire Reform (Scotland) Act (2012) and the vision for British Policing (see Independent Police Commission 2013, pages 14–15 and Chapter 2) are examples of how partnerships in policing are shifting from being encouraged to being required across most policing functions.

Third Party Policing (TPP) is just one form of partnership policing. In TPP, partnerships are formed to build police capacity to control and prevent crime. The key mechanism of TPP is activating or escalating ‘latent’ legal processes. These legal levers are often under-utilized or dormant (at least from a crime control perspective) without police initiating, fostering, reminding, and encouraging their partners to use their available legal levers. One of the central tenets, therefore, of TPP is that third parties that possess a legislative mandate (i.e., an existing legal lever) are likely to make better crime control partners than partners that lack access to a legal lever.

Whilst police can foster productive partnerships in many different ways, whether or not they are sustainable in the long run depends on the stake that the partner has in the problem, and their mandated, legislative responsibilities. In TPP, the availability of legal levers is a necessary condition and represents one type of policing partnership that has the capacity to endure and offer residual crime control benefits that extend beyond the initial intervention. Figure 1 illustrates the key components and processes of TPP. As Fig. 1 shows, TPP comprises three component parts: (1) police (‘first party’), (2) crime or disorder problem (‘second party’), which could be a problem place, person, or situation where criminogenic places, times, and people converge, and (3) an external entity (‘third party’) that police partner with to control or prevent the crime or disorder problem.

Fig. 1
figure 1

Third Party Policing model (Mazerolle et al. 2013)

Full size image

In TPP, the ‘first party’ is defined as the public police. As Fig. 1 illustrates, public police work in partnership with a third party for the purposes of controlling or preventing a crime and/or disorder problem. Partnerships may be forged in an episodic manner (see Mazerolle and Ransley 2005), through a program of crime control activities (e.g., Pulling Levers Policing, see Braga et al. 2012), or because the partnership is mandated by law (e.g., the UK Crime and Disorder Act, 1998; the Scottish Police and Fire Reform Act, 2012).

The ‘second party’ in TPP is defined as the ultimate crime control or prevention target (see Buerger and Mazerolle 1998; Mazerolle and Ransley 2005). In line with Routine Activities Theory, the ultimate target of a TPP intervention can be a problem person (a motivated offender), a problem place (an amenable place), or a problem situation (a suitable target, absence of suitable controllers) (see Cohen and Felson 1979; Eck 1994; Felson 1995). In essence, TPP interventions aim to focus police resources on one or more criminogenic factors that either allow crime problems to flourish, or prevent crime problems from emerging or escalating.

The ‘third party’ lies at the center of the TPP intervention approach. A third party is an entity—a person, an agency, organization, or business—operating within a legal framework and with legal powers and responsibilities not directly available to police. The third party is valuable to police because they have access to a legal lever that is (or could be) applied to control or prevent a crime or disorder problem (see Mazerolle et al. 2013). The third party is thus the partner and agent of crime control within TPP. A third party can be an individual (e.g., a bar staff member, property owner), an organization (e.g., Pharmacy Guild), a business (e.g., a bar), a regulatory authority (e.g., liquor licensing authority, local council, school), a government department (e.g., education department), or a network of collaborating agencies (e.g., see Green 1996).

The key defining feature of TPP is that police indirectly, rather than directly, target crime and disorder problems, and they do so through a partnership with a third party and through the legal levers available to that third party (see Mazerolle et al. 2014). In the following sections, I explore how different crime control partnerships between police and third parties, coupled with different types of procedural processes of legal levers, are likely to influence the effectiveness of TPP for addressing crime and disorder problems in the long term. First, I describe the relationship and legal lever components of TPP and note how these two mechanisms work together to distinguish TPP from other types of policing approaches. Second, using theoretical and empirical evidence, I will explore how the partnership and legal lever mechanisms are expected to help police sustain long-term crime control gains.

The power of Third Party Policing partnerships

The power of policing partnerships is realized when partnerships and legal levers interact in productive and legitimate ways to tackle crime and disorder problems. Like the chemicals nitrogen and glycerine, which are largely benign on their own, legal levers and partnerships that operate outside of a TPP intervention lack the interactive qualities and thus the power to control and prevent crime problems in the long term. As with the nitration of glycerin to produce the explosive ingredient of dynamite, I propose that it is the process of forming legitimate partnerships in policing that enables activation (and escalation) of legal levers and ultimately produces effective and sustainable crime control outcomes. In short, partnerships with access to legal levers lead to a ‘whole’ that is more than the sum of the parts. The mechanisms that foster these powerful policing partnerships are unpacked in the following paragraphs.

Building productive TPP partnerships

TPP partnerships exist when police partner with one or more third parties to work together to control or prevent a crime or disorder problem through the initiation and/or escalation of third parties’ legal levers. The varying characters of these partnerships are depicted in the TPP Partnership Matrix (see Fig. 2), which categorizes TPP partnerships along two continuums: number of third parties and type of engagement strategy (see Mazerolle et al. 2013). That is, TPP partnerships can involve single or multiple third parties and the techniques used to initiate or maintain TPP partnerships can range from collaborative to coercive on the ‘Engagement Continuum’ (see also Mazerolle et al. 2014, for an in-depth illustration of the TPP partnership dimension).

Fig. 2
figure 2

Third Party Policing Partnership Matrix (Mazerolle et al. 2013)

Full size image

As Fig. 2 shows, the bottom left quadrant represents TPP partnerships that involve multiple third parties and the use of collaborative engagement techniques to forge and maintain legitimate partnerships. Although this type of partnership is likely to be complex due to multiple stakeholders (e.g., see Fleming 2006; Meyer and Mazerolle 2014; Walters 1996), they may be particularly effective for addressing crime and disorder problems because they promote a multifaceted approach to targeting criminogenic factors through productive and sustainable partnerships. An example of a TPP intervention involving multiple third parties working in a collaborative manner is the Specialised Multi-Agency Response Team (SMART) initiation (see Green 1996; Mazerolle et al. 1998, 2000). In this TPP intervention, police partnered with SMART, which was comprised of representatives from several regulatory agencies (third parties), such as housing, health and safety, and fire authorities. A police officer and SMART initially attempted engagement with property owners (perceived to be part of the crime problem) to remind them of the legal requirements around property management under housing, safety, fire, and health legislation (legal levers) and offered a free course on property management. Where the drug problems persisted or property owners did not cooperate with SMART and the police, the legal levers were gradually escalated to compel cooperation (e.g., issuing code violations, fines, property closures, or initiating court proceedings). For reasons I explore below, these types of collaborative TPP partnerships are likely to generate and sustain long-term crime control gains.

Figure 2 also identifies TPP partnerships that involve a single third party and the use of coercive techniques to forge and maintain the crime control partnership (see Fig. 2, top right quadrant). For these types of partnerships, police often try to work with less-than-willing partners, convincing (or threatening) them to work with police in order to deal with a crime and disorder problem. Read and Tilley’s (2000) description of the relationship between police and the UK Health Authority (third party) during efforts by police to address the problem of doctors being ‘generous’ with their prescriptions of methadone suggests a coercive type of partnership. The outcome of the partnership was that the UK methadone prescription regime was tightened, with strict guidelines to control the crime problem through monitoring pharmacies dispensing methadone, and with reprimands against doctors who continued to overprescribe the drug.

The type of TPP partnership is likely to influence the effectiveness and sustainability of the TPP intervention. Claiborne and Lawson (2005), for example, show that collaborative partnerships that are built when agencies work together “in response to special mutually dependent needs and complex problems” (p. 2) are likely to last over time. Other research shows that interagency partnerships that lack mutually beneficial goals are hard to sustain beyond the short term (e.g., see Roussos and Fawcett 2000; Sloper 2004; Sridharan and Gillespie 2004).

The emerging theory of relational coordination helps unpack the dynamics of these working partnerships. Bond and Gittell (2010), for example, posit that “coordination that occurs through frequent, high-quality communication supported by relationships of shared goals, shared knowledge, and mutual respect enables organizations to better achieve their desired outcomes” (Bond and Gittell 2010, p. 119). Empirical tests of relational coordination theory show that multiagency partnerships characterized by a high degree of relational coordination achieve greater levels of efficiency and outcome quality (e.g., Gittell 2001, 2009; Gittell et al. 2000, 2008; Havens et al. 2010). Similarly, TPP partnerships built in collaborative ways, such as those suggested by relational coordination theory, are likely to foster a sense of mutual legitimacy. Consistent with Sunshine and Tyler’s (2003) findings, mutual legitimacy garners long-term voluntary cooperation from third parties to assist police with their crime control goals. For policing, proactive engagement techniques that foster mutual perceptions of legitimacy are critical for establishing and sustaining long-term crime control.

Activating legal levers

TPP approaches not only rest on partnerships but the ‘whole’ is also made more than the sum of the parts through the activation or escalation of a third party’s legal levers. The processes that dictate initiation and escalation of these legal levers both define TPP and distinguish the approach from other policing interventions, such as Problem-Oriented Policing (Goldstein 1979, 1990; Spelman and Eck 1987) and Intelligence-Led Policing (see Ratcliffe 2002). Legal levers are broadly defined as the legal powers possessed by third parties that create a crime control or crime prevention capacity that is otherwise dormant, under-utilized, or unavailable to police (see Mazerolle et al. 2013; Mazerolle and Ransley 2005). Police use TPP partnerships to access, influence, and/or activate these legal levers in TPP interventions. It is the activation of these latent legal levers that is expected to control or prevent the crime problem targeted by the TPP intervention. Example legal levers include conduct licensing (e.g., alcohol, firearms), mandatory reporting (e.g., chemical sales, child abuse), orders to control behavior (e.g., gang or domestic violence injunctions, truancy regulations), orders under regulatory codes (e.g., building, fire, health and safety, noise codes), and property controls (e.g., drug nuisance abatement).

Legal levers define and shape TPP interventions by delineating potential third parties available for a police partnership and the procedural aspects of a TPP intervention. These legal levers define and mandate processes, specify legal outcomes, and stipulate legal consequences for noncompliance. Legal levers in TPP are situated within an overarching legal framework that aims to regulate social, economic, or functional activities such as health and safety, licensing, banking, and/or transportation and third parties possess the legal power to regulate these activities (within their jurisdiction) through implementation or enforcement of legal levers. For example, a bar owner implements legal levers around responsible service of alcohol in order to regulate the behavior of patrons (e.g., staff training, alcohol serving times, age restrictions). Alternatively, a liquor licensing authority enforces compliance with licensing conditions, also aimed at regulating behavior of patrons and licensed establishments (e.g., fines for serving underage patrons). In TPP, police assume that conditions that allow crime problems to flourish can be controlled when (or if) third parties use their legal levers to regulate behavior. In TPP, police and their partners target the behaviors of individuals or groups of individuals, or the people frequenting crime-prone places or geographic areas. In sum, police partner with third parties to modify the criminogenic conditions underlying a crime problem, thereby indirectly controlling or preventing the problem through implementation, or enforcement, of available legal levers.

Unlike problem-oriented policing, where the process of intervention is driven by the SARA (Scanning, Analysis, Response, and Assessment) model (Eck and Spelman 1987), the law depicting legal levers dictates the processes of TPP interventions. In TPP, these laws are largely drawn from the increasingly complex web of regulatory laws in the ‘new regulatory state’ (Braithwaite 2000), where the emphasis is not so much on post-event use of formal legal sanctions, but rather on articulated and graduated actions that ultimately seek voluntary cooperation (see Mazerolle and Ransley 2005 for a review). In explicating the theory of responsive regulation, Ayres and Braithwaite (1992) and Braithwaite (2006, 2011) describe this system of graduated sanctions as a regulatory pyramid (see Fig. 3). The pyramid captures how regulators respond to each successive act of noncompliance by progressing through a hierarchical range of sanctions in a systematic and increasingly punitive way. As Ransley (2014) suggests, the range of legal levers that could be used in TPP interventions is extensive. A preliminary review of TPP literature (see Mazerolle and Ransley 2005) shows that most legal levers utilized in TPP interventions align closely with Braithwaite’s (2006, 2011) concept of the regulatory pyramid. That is, legal levers are activated by initiation of benign consequences to encourage compliance (e.g., education, warning letter) and then sequentially escalate to more punitive consequences to coerce compliance (e.g., infringement notices, to fines, to license revocation). This codified and stipulated process for regulating conduct defines the stages in a TPP intervention, where breaches are dealt with by sanctions of increasing severity, with the ultimate sanctions at the tip of the pyramid.

Fig. 3
figure 3

Regulatory pyramid (adapted from Ayres and Braithwaite 1992; Braithwaite 2011; 2006)

Full size image

Braithwaite (2006, 2011) suggests that regulation of social, economic, or functional activities through the pyramid structure is both efficient and effective, provided regulators are willing and able to consistently initiate and escalate sanctions in response to transgressions. For Braithwaite (2011), the range of sanctions aligns with the array of capabilities and motivations that underlie noncompliance. Figure 3 illustrates this alignment. As Fig. 3 shows, if an ‘offender’ is a rational actor who is motivated by deterrence, the use of persuasion or education at the bottom of the pyramid is unlikely to be effective for obtaining compliance. However, in the responsive regulatory model, an offender motivated by deterrence could be coaxed into compliance through activation of deterrent-based sanctions (e.g., warning letters or civil actions) that sit at least half-way up the pyramid. Provided that citizens believe “…in the inexorability of escalation if problems are not fixed” (Braithwaite 2011, p. 489), most escalations should not proceed far beyond the lower levels of the pyramid. Moreover, Braithwaite (2011) suggests that escalations are unlikely to compromise perceptions of legitimacy pertaining to the law or the regulator if the regulatory process begins with approaches that align with the principles of procedural justice. As a result, people who are targeted within regulatory frameworks are more likely to voluntarily comply in the long term without the threat of punishment or coercion than those targeted using business-as-usual laws, particularly criminal laws at the arrest end of the spectrum (but see Weber 2013 for analysis of how these processes and laws are often poorly coordinated and understood).

Generating the power to sustain crime control gains

Crime control gains can be sustained over the long term when productive partnerships (that are based on mutual respect, relational coordination, and legitimacy) use and escalate legal levers that are available to the third party partners. In TPP, the sustainable crime control benefits accrue through police forging partnerships with third parties that can active or escalate legal levers. This pathway to long-term crime control is explained below.

Why is a partnership not enough? A partnership between police and another entity on its own is unlikely to generate the capacity to control crime over the long run without the third party having access to a clearly articulated legal lever. Bond and Gittell’s (2010) study is a case in point. In their study of re-entry programs, partnerships between criminal justice and substance abuse agencies did not predict significantly higher levels of collaboration, and in some, albeit not all of the models, higher levels of relational coordination were associated with increased, rather than reduced, recidivism rates. Bond and Gittell (2010) concluded that the lack of formal systems in place for fostering the partnerships was one likely reason for why some partnerships failed (see also Liddle and Gelsthorpe 1994; Meyer and Mazerolle 2014). In TPP, therefore, partnerships are most likely to be effective and sustainable when existing legal provisions underpin the foundations for collaboration.

Why are legal levers not enough? When legal levers exist within regulatory frameworks, they should already work to generate voluntary compliance in the way hypothesized by responsive regulation theory. Yet this is not always the case. In policing and crime control, many regulatory legal levers are not consistently activated in ways that make them effective for controlling or preventing crime and disorder problems in the long run. The street-level bureaucrats literature highlights how those on the ‘front-line’ of policy, regulatory, and legislative implementation often lack the knowledge of the legal provisions available to them, and, even if they are aware of the provisions, they often know little about the procedures for activating the legal processes and/or lack the capacity to do so (see Baldwin and Black 2007; Gofen 2013; Lipsky 2010; Weber 2013). As a result, we observe a patchwork of individual attitudes, levels of knowledge, and beliefs among those responsible for activating legal levers. This, in turn, influences the way they implement laws in day-to-day practice. Bergen and While (2004), for example, demonstrate that the likelihood of nurses adopting new government regulations and policies was influenced by the clarity of policy guidance and the alignment of the policy with their personal and professional values (see also Tummers 2011).

For policing, existing legal levers are likely to generate crime control gains in the long run when they are activated within the context of a collaborative partnership. The formation of legitimate and functional crime control partnerships alters a third party’s awareness and attitude toward their own legal levers. These changed, more positive orientations to the third party’s legal levers increase the likelihood that third parties will use, escalate and continue to utilize the full range of legal lever sanctions (from education and persuasion through to prosecution) in the way originally envisioned by Braithwaite’s regulatory pyramid. Long-term sustainability of crime control gains thus occurs when a TPP partnership fosters the capacity and willingness of partners to use the full range of their legal levers in a consistent manner that then becomes routine and entrenched within the third party agency operations.Footnote 1

The ABILITY Truancy Trial: experimentally testing TPP and its long-term impacts

The ABILITY Truancy Trial is an explicit test of TPP that has been designed, from the outset, as a longitudinal experimental test of a policing intervention. The genesis of the ABILITY Trial began when a Queensland Police Service (QPS) Superintendent expressed concerns about youth truancy in her policing district. She lamented the overrepresentation of truanting youth in the district’s crime statistics, the potential for truancy to compound youths’ already existing criminogenic risk factors, and the lack of police rapport with schools in the area. One of the local school principals recalled that, prior to the start of ABILITY, police were pelted with tomatoes and lunchbox leftovers any time they walked inside the school grounds. In late 2009, police in this particular QPS district decided to focus their enforcement efforts on truancy, starting with a desire to forge a working partnership with the schools. The police recognized that the schools were a vital partner for the police because they had a legislative responsibility to address truancy problems. The police reached out to the Education Department and school principals and suggested that they work together to address the truancy problem in their district.

At the same time that the idea of a police-led truancy project was forming, the police Superintendent personally approached me to discuss my interest in evaluating the intervention program. Multiple working meetings, ethics applications (UQ Ethics Approval No. 2010000500), and funding applications later (see ARC 2010, FL100100014), the police-led program was named the “ABILITY Program” with agreement to test the short- and long-term effectiveness of the program under randomized field trial conditions. Based on previous work with the QPS (e.g., see Mazerolle et al. 2012), the police appreciated the advantages of experimental research for assessing the impact of their practices. Between December 2010 and June 2011, the pilot of the ABILITY Trial intervention component was conducted (see Mazerolle et al. 2011), and by October 2011, the ABILITY Truancy Trial was launched. In the sections that follow, I describe the key components of the ABLITY Trial and highlight how the ABILITY Trial aligns with Joan McCord’s legacy.

ABILITY Trial methodology and intervention description

The ABILITY Truancy Trial is a randomized field trial that aims to test the effectiveness of TPP for addressing crime problems over the long term. The ABILITY intervention is a police-led, TPP approach that seeks to reduce truancy and crime problems associated with families with truanting youth. Specifically, the ABILITY Intervention involves (1) a police partnership between the QPS and 11 target schools who (2) possess legal levers that create a legal mandate for schools, and the Education Department more generally, to address a crime and disorder problem (in this case, truancy and associated criminogenic risk factors). Between October 2011 and May 2013, the schools and police together identified and recruited 102 truanting youthFootnote 2 and their responsible guardians from the 11 participating schools (see section below for eligibility criteria and sample description). Students and their families who agreed to participate were randomly assigned either to the control group or to the ‘engagement’ group, which received the experimental intervention. The control group received business-as-usual action (arrests, cautions) by police for those truants caught committing offenses, coupled with school-issued warnings that rarely escalated through the regulatory pyramid. The control group cases also received a resource package that identified potential service providers that were available to all families in the school districts.

The experimental component of the ABILITY Intervention comprised activation of truancy laws through the formation of a partnership between police (QPS) and the Queensland Department of Education, Training and Employment (DETE) school staff from the 11 participating schools. Figure 4 summarizes the escalation of the legal levers pertaining to truancy in Queensland within the format of a regulatory pyramid.

Fig. 4
figure 4

Queensland Department of Education, Training and Employment truancy legal lever initiation and escalation process

Full size image

The ABILITY Trial specifically targeted truancy problems because, at the time of inception, police in the target district intuitively believed that truancy was symptomatic of disorder and other risk factors (e.g., see Henry 2007; Sutphen et al. 2010) and causally related to future delinquency (e.g., see Farrington 1989, 1996; Henry 2007; Henry and Huizinga 2007). In the ABILITY Trial, police and schools worked together to conjointly address the truancy problem through activation and escalation of the school’s legal lever as a way to indirectly target the underlying issues of each student’s truanting behavior. The police–school partnership was formed with a view to generate long-term crime control gains.

The experimental partnership between police and the schools was most salient during the Family Group Conferences (FGC). These conferences provided a forum for school representatives and police to come together with the truanting young people and their guardians to discuss a number of key matters: first, the forum provided an opportunity for the school representatives to explain to the experimental group young people and their guardians the steps that schools need (by law) to follow if and when young people fail to comply with the lawFootnote 3; second, the FGC used restorative practices to elicit the ‘affect’ of the truanting behavior from all forum participants; third, using a child-centered approach, the FGC facilitator empowered the truanting young people and their guardians to articulate an Action Plan and work together to reduce the truanting problem. Using a dialogue of procedural justice (see Bottoms and Tankebe 2012) and restorative principles (see Braithwaite 2003; Dowling 2010a, b), the FGC forum sought to solidify the working partnership between the schools and police and, at the same time, foster a genuine motivation for the young people and their guardians to willingly comply with the law.

Prior to each FGC, the police representative and trained conference facilitator collaboratively engaged with school staff, the student, and the student’s family to gain an understanding of the problem and issues underlying the student’s truancy. In the context of the FGC, a trained facilitator, the truanting young person, a responsible guardian (usually one or more parents), a representative from the school, and a uniformed police officer came together to discuss the truanting matter. Representatives from relevant social services agencies were invited to attend and participate in the FGC on an as-needed basis. The FGC facilitators used a specially crafted Conference Plan that: (1) communicated the legal lever initiation and escalation process (see Fig. 4); (2) incorporated the key ingredients of procedural justice (giving voice to all FGC participants, treating people with dignity and respect, being neutral in decision making, and conveying trustworthy motives); and (3) emphasized restorative processes, especially by empowering the group within the FGC circle forum to collectively devise a plan for resolving the truanting problem. Facilitators used a specially designed script that crafted restorative language into communicating the consequences of truanting. Using this script, the facilitators collaboratively led the participants through a discussion that focused on the truanting young person and centered on the events contributing to truancy, the impact of the truanting behavior for the young person and others, and the legal consequences of truancy for young persons and their families in the short- and long-term (legal levers). With the help of the facilitator and other FGC participants, the young persons and their guardians developed an Action Plan that aimed to reintegrate the young persons into a positive learning environment.

Following the conference, the ABILITY police representative engaged with the students, their families, and school staff to identify progress or transgressions from the Action Plan. Communication between police and the schools was critical: If police subsequently identified that a student continued to truant, the school staff were encouraged to initiate or escalate the truancy legal lever in accordance with the Education Act and related policy and procedures. Approximately 6 months after the FGC, experimental participants took part in an Exit Meeting to discuss progress (or lack of progress) with their Action Plans.

The experimental condition of the ABILITY program encapsulates the key ingredients of TPP (see Figs. 1, 2, 4). First, it includes a collaborative partnership between police and the schools. Second, it harnesses and activates the legal levers pertaining to how and when the truancy legislation could be used. Third, it brings these two TPP conditions (partnership and legal lever) together within the context of the FGC as a means to solidify the partnership. Importantly, the use of procedural justice dialogue and restorative processes within the FGC sought to foster perceptions of the legitimacy of the truancy laws and empower the young persons and their guardians to willingly alter their patterns of truancy.

Outcome measures

Consistent with Joan McCord’s legacy and the general call for experimental longitudinal studies in criminology (see Farrington 2006, 2013), my research team designed and obtained ethics approval to conduct baseline measures and 12-week, 6-month, 1-year, 2-year, and 5-year follow-ups with the ABILITY youth and their responsible guardians. We also obtained ethics approval and informed consent from participants to obtain longitudinal official data from the QPS (e.g., arrest, cautions) and participating schools (e.g., attendance, grades). Ethics approvals and funding permitting, we also intend to conduct a 10-year follow-up of the trial participants. My research team developed a range of measurement instruments that aim to evaluate the impact of the ABILITY Trial across a number of different outcomes. Wherever possible, preexisting scales with strong psychometric properties have been selected. Table 1 provides a measurement summary for the evaluation of the ABILITY Trial, heeding Joan McCord’s legacy by committing to a robust, broad, and comprehensive measurement model.

Table 1 The ABILITY Truancy Trial Outcome Measurement Summary
Full size table

ABILITY Trial schools and participant sample

The 11 schools participating in the ABILITY Trial are located within one of the most highly disadvantaged metropolitan areas of Queensland’s capital city, Brisbane. As of 2011, seven schools in the trial fell below the average index of Community Socio-Educational Advantage (ICSEA) (see Australian Curriculum, Assessment and Reporting Authority (ACARA), 2013), nine schools were deemed Low Socio-Economic National Partnership schools (see Australian Government 2013), and six schools were rated as falling substantially below the Australian average across a majority of literacy and numeracy domains (see ACARA 2013). Moreover, 48.9 % of families in the participating schools catchment area had no working parent (Australian average is 19.8 %) and 36.9 % of families in the catchment areas received Centrelink welfare benefits (Australian average is 16.8 %) (using census statistics available at beginning of trial; Australian Bureau of Statistics 2006).

To be eligible for inclusion in the ABILITY Trial, students had to be aged 10–16 years with 85 % or less attendance in three previous school terms and no legitimate explanation for their absences from school. Based on preliminary data obtained from the QPS and participating schools, we know that 91 % of the families in the ABILITY Trial had some form of contact with police, 42 % of the participating families had police contact due to child protection concerns, 54 % of the young people had some form of police contact including shop lifting, street checks, child protection concerns, and being named as a witness to an incident, and 62 % of the youths’ responsible guardians had some form of police contact including child protection concerns, domestic violence, and drug-related offences. In short, these data show that there was substantial crossover between the families posing a problem for schools in terms of truancy and families who were coming into contact with police.

Table 2 below compares the experimental and control groups on some basic demographic characteristics. As Table 2 shows, the simple random allocation process led to a high level of equivalence between the experimental and control groups. Indeed, no significant differences exist between the two groups on key demographic variables at baseline. In terms of truancy levels between the two groups, the range for all 102 participants was from 15 to 62 % absenteeism (across 3 school terms), which is an average of 23.72 % and equates to approximately one unexplained absence in every school week. The experimental group average was 24.93 % (SD = 12.08) and the control group average was 22.51 % (SD = 7.50), with no significant difference at baseline (t100 = −1.22, p = .227).

Table 2 The ABILITY Truancy Trial sample at baseline
Full size table

Conclusion

Joan McCord left an enduring legacy on the field of criminology in general and on experimental criminology in particular. She advocated for the use of experiments to test the effectiveness of criminal justice interventions, argued for policy makers and practitioners to adopt evidence-based practice, and worked to develop the skills and careers of younger colleagues. She was committed to multidisciplinary approaches to research, paid serious attention to the art of measurement, and demonstrated the importance of longitudinal follow-up in the years following an experimental test of crime and justice interventions. Yet the dearth of longitudinal follow-up in the population of existing policing experiments signals a significant gap in what we know about the effectiveness of policing.

In this paper, I have drawn on Joan’s legacy to contextualize the purpose and theoretical elements of the ABILITY Truancy Trial. The ABILITY Trial is designed as a longitudinal experimental test of TPP. The efforts of police to build working relationships with the schools (the third parties), who have both a stake in the problem (truancy) and possess legal levers that can control the truancy problem, are hypothesized to create long-term capacities for maintaining the crime control gains. The TPP approach works through collaborative partnerships with entities that have access to regulatory legal levers. These partnerships create long-term crime control gains, we argue, when the third parties communicate and activate (or escalate) their existing legal capacity to regulate criminogenic factors causing or facilitating crime problems, with the ultimate aim of long-term voluntary compliance with the law. The longitudinal nature of the ABILITY Trial will allow us to track the residual crime control gains of the TPP approach, which we expect to last well beyond the initial efforts to encourage the young people to willingly attend school.

In the spirit of Joan’s legacy (e.g., see McCord 2003), we recognize that an intervention like ABILITY could have some backfire or harmful effects. What if parents are unable to get their children to attend school? What if the school environment is so negative for the child that the pressure on the child to attend school ends in a deleterious outcome for the child, or for another attending child? Desmond and Valdez’s (2013) study that itemizes the negative outcomes of TPP in Milwaukee is a case in point: their study depicted how nuisance property ordinances can negatively impact women living in the inner city, highlighting the importance of tracking and reporting harmful impacts of TPP, even when police do achieve crime control gains. In the ABILITY Trial, we are collecting a range of outcome measures—and collecting these outcome measures over long periods of time—paying attention to the possible backfire effects of the ABILITY Trial.

The ABILITY Truancy Trial is designed as a longitudinal experimental test of TPP that pays close attention to gathering data on the holistic and long-term impact (both the positive and negative) of the ABILITY intervention. The Trial contributes to the small number of policing experiments that track long-term outcomes. The cooperative nature of the partnership, coupled with the legitimacy of the legal levers, builds long-term capacity for crime control. This, I suggest, is consistent with the vision of the British Independent Police Commission, which states:

“Good policing requires the police to foster and sustain collaboration in ways that galvanise social action against crime without either extending the reach of police or overriding the purposes of other agencies” (p. 47).