The shift in focus from victims to the most serious perpetrators of domestic abuse

Abstract

This article contends that there has been a recent shift in focus towards the most serious perpetrators of domestic abuse in England. Police forces employ different ways of both identifying and managing high-harm high-frequency domestic abuse perpetrators. However, there is little evidence as to which management strategies are most effective. We use research findings from a recently conducted academic/police research partnership project to understand how the new focus on perpetrators is configured within current and proposed police practices and strategies in England.

Since the 1980s, victims of domestic abuse have slowly moved closer to the centre of policy debates, both symbolically within public and political discourse and also within criminal justice processes themselves, as this article will show. Victims now have more control over the processes that follow report of an incident to police than they have done for over a hundred years, although victims’ rights groups and the Victims’ Commissioner (2022) advocate for more. Nevertheless, we suggest that there has recently emerged a much stronger focus on offenders, particularly onto the small proportion of domestic abuse perpetrators who cause the most harm most frequently. The new emphasis on offenders does not, and should not, displace the need for greater rights and empowerment for victims, but it does raise questions about how police can address the persistent and significant problem of serious domestic abuse. In this article, we use data from a recently conducted study in England, to ask three key questions about the implications of the shift in focus from victims towards perpetrators. How are the most serious threats to victims of domestic abuse currently categorised in police systems? What strategies are employed to manage these offenders? And, finally, can the strategies currently employed cope with the numbers of people considered to offer the most serious threat? The article explores these questions by analysing data from our study, including statistics on the most serious offenders, using police records to contextualise these data. We then construct vignettes to illustrate offending trajectories. Finally, we present the conclusions of our study, including recommendations for future police practice in addressing the most serious offenders.

The victim’s policy journey

In the late 19th century, victims were responsible for initiating a court case, arranging for the arrest of the suspect and paying for the costs of the trial. However, by World War II victims played little part in the prosecution of offenders, other than acting as witnesses. As the police took over the process, acting as prosecutor in magistrates’ courts and appointing counsel in the Crown courts, the victim’s marginality to the process grew (Godfrey, 2008). By the time that academics became interested in victims, their centrality to the system had waned considerably, influencing the scope of academic research. Research from the 1940s to the 1970s was mainly concerned with ‘typologies’ of victims, and how they had contributed to their own victimisation (Hentig, 1948; Hindelang et al., 1978; Mendelsohn, 1963; Wolfgang, 1958). It is not surprising that this approach became increasingly problematic when critical criminology emerged in the 1980s and 1990s (Newburn and Stanko, 1994; Schafer, 1968; Stanko, 1988; Walklate, 1992). By that time, activists had established charities such as Women’s Aid, secured the provision of domestic violence refuges and were continuing to agitate for greater co-ordination between statutory and voluntary agencies to support victims (Cook and Jones, 2007; Hoyle, 2007). The authorities slowly moved in the same direction with a key shift signalled in the 1990 Home Office Circular which urged the police to treat domestic violence as seriously as other violent crimes (Hoyle and Sanders, 2000). The 2004 Domestic Violence, Crime and Victims Act then established the Code of Practice for Victims of Crime and created the role of Victims’ Commissioner, in an attempt to monitor criminal justice agencies’ response to domestic abuse.

In response to criticism from Her Majesty’s Inspectorate of Constabulary and Fire and Rescue Services (HMICFRS) about the falling levels of arrests in domestic abuse cases, a positive action policy (defined as taking steps to ensure the safety of victims, usually, but not always, by arresting a suspect) was introduced to increase the arrest rate (College of Policing, 2015b). Following this policy, the average arrest rate across the United Kingdom hovered around 30% (Office for National Statistics (ONS), 2022). The 2021 Domestic Abuse Act introduced new measures designed to support victims, including eligibility of domestic abuse victims for special measures in courts (for example, giving evidence from behind a screen or by video link), and giving priority need to victims (for example, in relation to homelessness caused by escaping their abuser). It also created a statutory definition of domestic abuse which encompassed emotional abuse and coercive and controlling behaviour. Domestic Abuse Protection Notices and Orders were introduced to protect victims and, finally, the 2021 Act required the Home Secretary to publish a domestic abuse strategy which focusses on both victims and offenders. The Violence Against Women and Girls (VAWG) strategy was launched in the same year to prioritise the safety of women and girls (Home Office, 2021). Domestic Abuse comprises a significant proportion of incidents reported to police in England and Wales (between 15% and 20%). For the year ending March 2022, 68% (910,980) of the over 1.5 million reported domestic abuse incidents/crimes were classed as crimes. However, just 7% (67,063) of these were referred to the Crown Prosecution Service. Of that number, 72.7% (53,207) were prosecuted and from those prosecutions 76.4% (40,650) were convicted (less than 5% of the total number of reported offences). ¹

Turning the gaze towards perpetrators

There is a laudable aim to empower the victim and make them a part of decision-making processes (asking victims of incidents what kind of action they think should be taken, whether they wish to prosecute, the facility to retract statements before trial, use of victim personal statements in court, the Victims’ Code providing recourse to review of unduly lenient sentences and so on). However, this has placed greater and greater responsibility for initiating and sustaining a criminal justice response to their victimisation (Davies, 2018; Hoyle, 1998, 2007; Mills, 1998). Enabling and supporting victims so that they have enough confidence in the criminal justice system to continue with the prosecution process are challenging for police. The number of prosecutions for domestic abuse has fallen each year since 2017 and, due to the length of time a trial now takes to come to court, the likelihood is that greater numbers of cases will be dropped through victim attrition (Godfrey et al., 2022). The most recent figures show an average (mean) of just under 700 days from offence to completion for Crown Court cases. ² The need to offer further support for victims is recognised by victim support charities and advocate groups. At the same time, the police seek ways of continuing with investigations and prosecutions when victims do withdraw their support. One response has been to prepare files for court which can be turned into ‘Evidence Led Prosecutions’ so that, if a victim does not appear in court on the day, footage from police body-worn cameras, res gestae statements and so on can be used as evidence (Godfrey et al., 2022). However, criticism of low prosecution rates and the inability to control violent repeat offenders continue to receive public attention and press criticism (Barr and Topping, 2020; Cowling and Forsyth, 2021). This became particularly acute when the Covid emergency raised fears of a ‘Shadow Pandemic’ (Mlambo-Ngcuke, 2020). During the lockdown periods, shoplifting, public order and alcohol-related offences all decreased, giving the police time and staff resources to quickly progress domestic abuse cases already under investigation, and to focus on households where domestic abuse was prevalent (Godfrey et al., 2022; Godfrey and Richardson, 2023; Hohl and Johnson, 2021; Johnson and Hohl, 2023; Walklate et al., 2021, 2022). However, frustration that perpetrators escape police attention continues to be prominent in both public and political discourse (Quinn and Adu, 2023). Serious and prolific offenders loom large in media and political debates that seek to capitalise on and push forward more punitive sanctions for offenders who are perceived to be predatory, especially those who cause the most harm, most frequently.

Who are high harm offenders?

There are published annual statistics on the number of domestic abuse incidents, crimes, prosecutions and victims, but very little data are available on perpetrators. One figure, now 10 years old, put the number of domestic abuse perpetrators in England and Wales at around 400,000 (Robinson et al., 2014). The Government’s 2023 Strategic Policing Requirement includes details of how domestic abuse should be tackled, with the key message being that ‘the relentless pursuit and disruption of adult perpetrators should be a national priority for the police’, with measures proposed to enhance police capacity to achieve this aim (Home Office, 2023a). However, it is unlikely that the police could easily cope with this number of perpetrators, if they could at all. It has been estimated that a quarter of domestic abuse perpetrators have committed more than one offence against current or former partners (Drive, 2020). Of a sample of 40 high-harm serial VAWG offenders identified as part of an inspection, nearly half had offended against two victims, 14 against three or more, and a few against eight or nine different victims (HMICFRS, 2021). The government stressed to Chief Constables that they must focus on ‘the most dangerous and repeat perpetrators’ and needed to ensure that first responders, public protection teams and local neighbourhood teams were all aware of the most dangerous perpetrators in their force areas so that they could proactively address their offending. A 2023 report conducted in the West Midlands police force reiterated that the need to identify the most harmful perpetrators was critical to effectively target resources (Crest, 2023: 10).

Police forces currently use a variety of mechanisms and criteria to identify the most serious domestic abusers. Definitions of ‘Repeat’ and ‘Serial’ offenders were created to facilitate better perpetrator management both within an individual force and between forces, although there is still considerable blurring between those terms. Repeat offenders are defined by Safe Lives (2018) as people in intimate relationships who offend against the same victim within a year of their last referral to a Multi-Agency Risk Assessment Conference (MARAC), which is approximately a third of all people currently on MARAC, according to Safe Lives. In 2015, the College of Policing (2015a) stated that definitions of repeat perpetrators were unhelpful because

the number of times a report has been made to the police is inherently unreliable due to the way in which domestic abuse is reported/not reported. Most victims and perpetrators have experienced or engaged in domestic abuse multiple times before a first report is made so they are, in reality, already repeat cases.

In other words, ‘the best predictor of future domestic abuse perpetration is a history of domestic abuse perpetration’ (Cordis Bright, 2022: 9). So, whereas ‘repeat victim’ remains a meaningful police term, the term ‘repeat perpetrator’ has lost currency, and there are no available national or local statistics of the number of repeat offenders. The term serial is also problematic, with many competing definitions (four just within the four police forces in Wales, Robinson et al., 2014); the College of Policing have worked hard to standardise their definition (two or more victims over any time period across all forces). However, in practice there is little difference in offending behaviour between repeat and serial offenders (although one study found that serial perpetrators have a more disturbing risk profile than other perpetrators, Robinson et al., 2014).

The most widely used bureaucratic tool employed by English police is the Recency, Frequency, Gravity and Victim matrix (RFGV), a demand-management tool based on how often, recent, harmful and serious is the offending, and how many different victims are involved in complaints against the same perpetrator. This system helps police to prioritise their activities. It is important to distinguish between demand-management systems tools (RFGV-fuelled Serial Domestic Abuse Perpetrator (SDAP) databases, sometimes termed top 10 or top 20 lists) and risk-management systems (threat-to-life databases), which try to predict which offenders will go on to commit fatal/near miss catastrophic offending. Both of those are relevant to domestic abuse perpetrators, but problems arise when demand-management tools are used as risk-management tools. Academics and police forces have attempted to close gaps between the two systems, or to use one system to inform the other, testing perpetrator identification tools (Clancy et al., 2014; Robinson et al., 2014), using statistical modelling to predict high-risk offenders prospectively (Bland and Ariel, 2020) and analysing potential progressive steps to domestic homicide (Chantler et al., 2020; Crest, 2023; Monckton-Smith, 2021). In addition, one English police force is currently analysing one-off domestic incidents to correlate factors present in those cases with subsequent fatal offences. Preliminary results of unpublished analysis by that force indicate that those who cross the divide from non-fatal to fatal offending may have previous histories of using strangulation in a relationship, possession of weapons, and kidnapping. Although having only been involved in one reported incident of domestic abuse, the presence of these factors may indicate a possibility of escalation in the future into serious domestic abuse and fatal or ‘near-miss’ fatal offending. Refining that piece of analysis, and realising the potential to combine demand- and risk-management tools, could ultimately produce the most sophisticated analytical tool for the management of more serious domestic abuse perpetrators in the future.

As this article will show, there are problems with demand-management tools. First, the spreadsheets they create quickly become swamped with data, making them operationally unmanageable with hundreds of people (and for one English force, over a thousand) in their top 20. Practices vary across forces concerning entry to, maintenance and exit from demand management, but there are still large numbers of offenders who entered but never left databases (although they are not actively offending). Some forces place people who have not recently offended on a ‘dormant’ list while other forces remove them. However, the main point is that, due to limited resources, only those at the top of these lists can be actively managed, so the tools can only work on ‘comparative’ harm, rather than ‘absolute’, and even with a focus on the more serious, rather than all, domestic abuse perpetrators, the numbers are unmanageable.

Another problem is that demand-management systems are not sufficiently connected to the strategies used to mitigate the risk posed. In essence, the system is being designed to predict how much resource should be devoted to managing individuals based on the recency, frequency and gravity of their offending, largely unconnected to any system that assesses which police strategy reduces re-offending for each of those perpetrators.

These problems persist, as HMICFRS found in 2021. They asked forces in England and Wales to identify individuals who posed the highest risk to women and girls in that area, later reporting that they ‘were concerned that of the 40 individuals that the forces identified to us, 34 hadn’t already been recognised and managed by the force as their most prolific repeat offenders’ (HMICFRS, 2021). Moreover, over half of the 40 individuals had antecedent histories of offending against women and girls who had lasted for over 5 years without effective police strategies being in place. Clearly, the system of identifying and managing perpetrators who pose the most serious threat needs refinement. Our research, therefore, explored which strategies are currently available to reduce re-offending by serial and repeat domestic abuse perpetrators and attempted to match strategies against patterns of re-offending.

Methods

We used a multi-method approach in this study, comprising documentary analysis, interviews and secondary analysis of police data. The first part of the study was an evaluation of the policies and procedures relating to domestic abuse perpetrators, through analysis of documentation from five geographically dispersed police forces in England. We then conducted semi-structured interviews with 18 officers from those five forces, who had responsibility for strategic and operational management of domestic abuse perpetrators. Interviews were conducted online via secure platforms, recorded and transcribed in accordance with University of Liverpool data-management protocols. Funding and time constraints only enabled us to access and analyse quantitative data from one northwest police force. Access to the data analysed in this article was granted by this force, following their information sharing processes and after ethical approval by University of Liverpool Ethics Committee (February 2022, reference 11056). The interviews and documentary analysis allowed us to understand current policies and practices in each force for identifying and managing high-harm domestic abuse perpetrators.

We analysed data kept on the police data-management system (Niche) from one force in northwest England, using their register of SDAPs. ³ We obtained data for the top 10 individuals on this database, compiled using the RFGV matrix, for December 2022. For each incident reported to police, the data records the complainant (anonymised for the purposes of the research), where the incident took place, what police action was taken and the criminal justice outcome. Using the information, we produced a timeline for every individual, showing their offending behaviour/victimisation over time, who they offend with, and against, and periods of imprisonment/community orders. Where possible, this offending timeline was matched against the strategies being used to manage that individual. This enabled us to build up a more detailed, longitudinal picture of each offender, which we present in vignette form. This approach builds on previous work using life-course longitudinal data (see, for example, Godfrey, 2011; Godfrey et al., 2007, 2010, 2017; for overviews of life-course criminology see also Jolliffe and Auty, 2023; McAra and McVie, 2017), with the aim being to form a ‘what works’ approach to reducing harm rather than developing a typology of offenders. ⁴

Findings

Our sample of the 10 highest priority offenders on the SDAP database (the ‘Top Ten’) were involved in a total of 2346 incidents, averaging between 3 and 28 incidents per person per year for 10 years stretching back more than 20 years. The highest number of offences committed in 1 year for each of the top 10 perpetrators per year ranged from 17 to 109. The offending profiles and trajectories were mixed – they appeared in the data as suspects (in 75% of incidents) and as complainants (in 25%) across a range of offence types – of which the largest were violence (48%) and property (15%). Of all the incidents where SDAP Top Ten offenders were suspects, 6% were domestic abuse incidents, while 94% were non-domestic-related. It has been suggested that practitioners are reluctant to identify a link between domestic abuse and non-domestic abuse offending and tend to perceive there to be a prevalence of domestic abuse ‘specialists’ (Crest, 2023). Yet, in line with recent research, we found that, in prolific offenders at least, domestic abuse was accompanied by many other types of offending (Verbruggen et al., 2022). From our study, the lives of Top Ten perpetrators are significantly involved in alcohol and drug-related offending, neighbourhood disputes and disturbances.

The following two vignettes illustrate how the data were used to construct life-histories of serial and repeat offenders:

Male, heterosexual relationships, born 1984 (340 incidents over 20 years)

Offending began in 2003 when he was 19 with his indecent assault on a girl under 16, then one shoplifting charge in 2004, followed by a gap in offending until 2005 when he was a suspect in cases of ‘possession with intent to supply’, public order and anti-social behaviour. He was convicted of threats and harassment, and also threatened arson against someone who he thought had passed information about him to the police 2 years earlier. He then became increasingly involved in drug-related offending, threatening violence to others, while he was a victim of aggravated burglary by rival dealers and threatened by a man with a gun. He reported to police that his partner had committed a section 47 assault (head wound) against him in 2007 when she found out about an affair. Throughout 2008, he was involved in drug-related violence including the use of weapons, both by him and against him. In 2012, he was imprisoned for 6 years for arson endangering life (having set fire to an abandoned house). Released on licence, he was convicted of harassment of his ex-partner and returned to prison. Release was followed by further offences, recall, further offences and so on. In 2016, his offending was erratic and regular. He was the subject of police investigation 8 times, with two resulting in charges. He was a disorderly offender (threatening shopkeepers, instigating a feud with an environmental health officer, shouting at passers-by in the street, breaking windows and so on). His life was also becoming increasingly chaotic, offending continually (low-level public order, violence, threats) although he was not charged by police. By this time, he was deeply involved in the drugs trade, and a repeat victim of drug-related violence and burglary. He was also subject to frequent mental health episodes. Between 2019 and 2022, he was involved in a plethora of low-level offences, including shoplifting, threats, possession of cannabis, fighting, intimidating pub staff, spitting at police officers, criminal damage and sending malicious communications. There followed a bout of offending against ex-partners, threats to police, mental health workers, ambulance staff, as well as being a general nuisance in his local town. In 2021, his poor mental health was evident. He was homeless and living in a tent, found naked and rambling in the street, committed ‘nuisance offences’ around town and was frequently hospitalised. He was still heavily involved in drug use in 2022. Although he is considered by police to be a serial and repeat domestic abuse perpetrator, his offending against current and ex-partners makes up only a small proportion of the offences for which he was a suspect or charged.

Female, heterosexual relationships, born 1980 (239 incidents over 13 years)

Her first offence was a section 47 assault on her 7-year-old child in 2009, followed by her involvement in a number of neighbourhood disturbances, which were never charged. Following the assault and the break-down of her marriage, she moved to rented accommodation, and subsequently formed a new relationship. Her new neighbours made complaints about her house being noisy and used for parties and prostitution. She was evicted in 2012 and until 2015 was involved in sporadic and infrequent offending (a string of alcohol and drug-related offences, assaulting police and public disorder). Her mental health worsened, and she was admitted as an outpatient at a mental health institution, during which time there were frequent assaults on ward staff. In 2016, she was living at the local YMCA, becoming more vulnerable to sexual exploitation, and involved in frequent public order and alcohol-related offending. In 2016, she committed serious assaults on both a neighbour and her partner (followed by malicious communications when that relationship ended). In 2017, her offending related mainly to alcohol and arguments with current and ex-partners, until a shoplifting spree in 2019 resulted in 18 weeks custody. In 2021 and 2022, she was repeatedly arrested for public disorder, shouting, breaking windows and causing a nuisance to neighbours. She was also arrested for a section 47 assault on her new partner. Her offending history was considerable, as was her record of being a victim.

Her house was burgled in 2009, and again in 2015. In between those events, she had her property damaged by neighbours, and had a drink thrown over her in the pub. She reported her estranged husband to the police for assaulting their 3-year-old daughter in 2011, but no action was taken. Similarly, she reported being a victim of an assault in 2015, and the subject of coercive control in 2019 but neither resulted in a prosecution. During this period, with family relationships irreparably broken, she was reliant on sleeping on other people’s sofas which made her more vulnerable (she was robbed while asleep in 2020 and had indecent photos taken by a ‘family friend’ while asleep on his sofa). Returning to housing association property in 2022, her house was referred to as a ‘crack den’ by neighbours, and she had frequent fights with other drug-users; there were problems concerning paying or rather not paying debts to drug-dealers (resulting in smashed windows, assaults and so on). During the last decade, her life has been in constant crisis.

Incidents involving these kinds of offenders, domestic-related or not, consume a huge amount of police time and resources. However, despite police attendance, and positive action usually being taken, there is often no criminal justice outcome. A third (34%) of all domestic abuse incidents in our police data from one force that involved a Top Ten perpetrator resulted in a charge (HO outcome code 1). In two-thirds (66%), no charge resulted because the victim did not support (HO outcome code 16, 38%), because the victim was not identified (HO outcome code 18, 20%), or because of problems with the evidence (HO outcome codes 14 and 15, 8%). Therefore, as far as the victim (and perpetrator) was concerned, there was no criminal justice ‘result’ in the majority of cases. Positive action by the police may have removed one of the parties, thereby calming down the situation, and offering temporary respite for the victim, but did not result in the perpetrator facing prosecution. Notes made on the police data-management system (Niche) can provide the context within which decisions are often made, as the two cases below illustrate:

2018 Closure Summary: I have advised complainant to keep a note of any negative contact she has with ex . . . she is struggling to cope with what she perceives to be his harassment of her. Her mother will stay with her until things settle. No new offences disclosed–Harassment has already been recorded and this is further evidence of same–Original Incident Report: My husband has just picked the children up from school walked in the house 15:25 and given me a lot of abuse, he forced his way in, was swearing at me and was abusive to my 70 year old mum. This was all in front of a workman I had in, and our children. Last night he was revving his engine outside and I refused to let him in the house. He was ringing saying he is going to destroy me. He has now left the house, but I am scared and he is demanding a key to the house and comes as and when he pleases.

Closure Summary: An arrest hasn’t been made because no evidence has been provided/is available to prove the offences. This is a reported Malicious Communications/harassment between two vulnerable individuals with Mental Health struggles, who have been in a relationship. . . . Neither [x] nor any family were willing to provide statements and they would not show me the images. Without any statements or screenshots of the threats, we are unable to evidence the offences and there would not be enough for a res gestae led prosecution. Due to both having vulnerabilities and neither having any history of domestic violence other than a reported rape that was investigated and NFAd [no further action taken] . . . I am satisfied that:- Suitable safeguarding has been put in place to prevent further harm – Positive action taken, both parties advised re further contact, neither are in a relationship any more. The investigation is complete, and the incident can be closed with no further tasking.

Given that most incidents did not result in a charge, and that offending continued for years, we examined which police-led strategies had been put in place to reduce or stop re-offending. The first strategy was bureaucratic, with offenders placed on a demand-management system. However, following entry onto an SDAP or other priority database we found that there were no subsequent standard routes or structures for management of these types of offenders (Godfrey and Richardson, 2023). Different forces have created process systems which attempt to meet their needs (partly determined by size of force, internal organisation and available resources).

One strategy which has been evaluated positively by academics, and consequently heavily invested in by Police and Crime Commissioners, the use of perpetrator programmes was largely absent in our data sample. By 2023, £36 million had been given to Police and Crime Commissioners by the government to fund perpetrator programmes (Home Office, 2023b). Participation in specialist perpetrator programmes delivered by the Probation and Prison Service can be imposed as part of a court order following conviction, or people can voluntarily access programmes delivered by the charity sector through self-referral, police referral or referral by outside agencies (Bates et al., 2017). Kelly and Westmarland (2015) found that perpetrator programmes could be effective, but less than 1% of offenders begin a programme and there is a significant drop-out rate (Respect, 2021). In plain terms, most perpetrators would never experience one unless the capacity of perpetrator programmes was massively expanded (which would be a seismic shift in the management of domestic abuse requiring significant investment). The importance of managing the 99% of domestic abuse offenders who do not undertake programmes has become very evident, and, for them, the most used strategy, according to our interviews, although it is not always easy to discern in the quantitative data, was ‘disruption and incapacitation’.

Disruption and incapacitation strategies extend from normal police practices and desires to frustrate offending and offenders through police patrols and quickly responding to incidents when they occur, to a set of strategies designed to proactively target the activities of high-risk offenders, particularly those who do not engage with perpetrator programmes or with other offender-management structures:

Is this a person that we can potentially manage? And that doesn’t necessarily mean manage them in a positive way in terms of, you know, I will do all the nice stuff which is sort out their alcohol issues, sort out their drug issues, sort out some of their mental issues or whatever other issues have, but it can often, I mean we will keep a much closer eye on you, and we will jump on you using all the statutory tools which are available to us . . . to manage your behaviour . . . If you don’t want to engage with us, that won’t stop us working with you. But it will be from a completely different viewpoint. We will hound you about your restraining order, any breaches, highlight that you weren’t in at your curfew time to the local police unit, go round and arrest you, knock on your mother’s door – we’ll be a general pain. . . (Reducing Re-offending Team Lead, Godfrey and Richardson, 2023)

Disruption can also include known perpetrators being tagged as ‘local targets’ for tasking and co-ordination by uniformed officers at local, area and force level, placing markers on vehicles, keeping a suspect in custody while completing enquiries, striving to meet the thresholds for remand rather than bailing suspects) and being responsive (designing and implementing individual ‘Trigger Plans’ so that officers attending incidents know how to proceed, frequent stop search, vehicle stops). Since serial and repeat domestic abuse perpetrators have a mixed profile of offending (with the largest proportion of their offending not being domestic abuse–related), this provides another opportunity for disruption. For example, one force’s ‘4P Plan’ (as part of ‘Op Dart’ Domestic Abuse Reduction Tactics) takes into account all forms of criminality, not just domestic abuse, for offenders already designated serial and repeat perpetrators:

The perpetrator should be targeted through a broad range of tactics and this may include targeting opportunities relating to drug offences, acquisitive crime, driving offences etc. aimed at maximising disruption and enforcement. (Operational Order, no date)

Officers should take the opportunity when responding to non-domestic abuse offences to disrupt offending trajectories. This seems a much more targeted and productive response than some of the net-widening suggestions recently offered (Crest, 2023). However, so far there has been very little evaluation of the effectiveness of disruption strategies, and evidence suggests that incapacitation – usually the imposition of short-term prison sentences – has little impact on re-offending for domestic abuse offenders (Gibbs, 2018). There is little evidence that they are effective in the medium or long term (Godfrey et al., 2007, 2010).

Many Top Ten were subject to MARACs. In MARAC, various relevant partner agencies are invited to talk about the risk of future harm in relation to children, intimate partner victims and vulnerable adults, and to draw up an action plan to help manage that risk. Perpetrators’ patterns of offending were considered only peripherally as they are more concerned with safeguarding victims and their children. Therefore, in some areas, a parallel system of Multi-Agency Tasking and Co-ordination partnerships (MATACs) was also established (Davies and Biddle, 2018). The MATAC framework is similar to MARAC but with a focus on serial perpetrators of domestic abuse rather than victims. When subjects for targeting have been determined, profiles are created, and monthly (usually) MATAC meetings are held to determine the most appropriate method of targeting, each perpetrator. Studies have found that being subject to MATAC did bring about a change in offending behaviour, but they have only been adopted by a small number of police forces, and for only a very small number of perpetrators (Cooper, 2020).

Limitations of the data

The Public Protection Unit Lead of one force stated, in regard to all of the strategies described above, ‘there is no sense of what actually works’ (Godfrey and Richardson, 2023). As with all criminological research relying on official data, the quality of the data determines the accuracy of the output. First, police data only record reported offences, and there may be other incidents which were never reported, and therefore do not enter the police database and therefore are not included in our analysis. Second, it is difficult to determine in the data which strategy was in place for distinct periods of time (during which time re-offending may or not have taken place). For example, one person seemed to be subject to an Anti-Social Behaviour Order, imprisonment, temporary repeated confinement in a mental health ward, detention on a mental health warrant and court-imposed community orders. They were also subject to police disrupt strategies, but the strategies seemed to overlap with others, such as MARAC. Our calculations of when particular strategies were in place were based on a best guess. We noted that there was considerable variation, overlap and blurring of different strategies for the same perpetrator, and little evaluation of the success or failure of any strategy that had been applied.

A further limitation with our study is that each location we looked at had a different approach to identifying and managing serial and repeat domestic abuse perpetrators. This is a limitation in the sense that it makes direct comparison more difficult. However, these different approaches are also one of the key findings of our study.

The consequences of the new focus on perpetrators?

The new focus on perpetrators may well inhibit offending, reduce the number of victims and impact heavily on the most serious and prolific perpetrators as it is designed to do; however, currently, it is difficult to draw that conclusion without a number of issues being addressed.

The primary issue concerns the evidential base used to identify the most serious offenders. While there is still a heavy steer towards categorising offenders into a large and varied number of diverse types (see Crest, 2023, for example), as far as identifying and monitoring perpetrators, and tailoring police strategies, there are only three groups. The first are the offenders who are never reported to the police. Figures from Refuge (2024) suggest that less than a quarter of domestic abuse crime is reported to the police. The victims of violence will be supported by charities and non-police agencies, and the police should take steps to increase trust so that victims will contact them. The second group are the perpetrators who have been reported once or twice for their actions. These people may stop offending, have their offending addressed through perpetrator programmes, commit a catastrophically serious offence or may continue their offending to become part of the third group, who we would term high-harm high-frequency (HHHF) perpetrators. The demand- and risk-management systems currently in use inadequately identify offenders in the second and third groups. There will have to be better attempts to join these systems using more reliable data, and clearly identifying all risk factors. For example, there needs to be far greater co-ordination between the NHS, GPs, mental health services and domestic abuse charities and the police in order to identify risk and also address domestic abuse trajectories at an earlier stage. This is not an easy task, since different agencies have different responsibilities, resource bases, ethical frameworks and data-protection systems. However, multi-agency partnerships of this kind could use relational databases to assess risk of catastrophic domestic abuse, but also provide opportunities to intervene earlier in the offending careers of HHHF perpetrators.

Once identified as HHHF, and an offender-management strategy adopted, the success of each approach should be evaluated (in terms of re-offending). Although perpetrator programmes have been shown to work for small numbers of perpetrators, there will need to be analysis of the effectiveness of disrupt and other strategies. This is especially as some forces are now seeking to extend the operational remit of these strategies. Disparate and divergent practices are emerging, including the use of tactics more normally adopted to address organised crime groups and terrorism. These include enhanced covert surveillance of suspects and victims, embedded undercover officers and other ‘lawfully audacious’ strategies. These tactics may well be necessary, and effective, but will need to be properly evaluated and analysed in terms of their effectiveness, but also to establish their legal and moral legitimacy.

Policing HHHF perpetrators using robust tactics may also be very popular with the public – for many people it is what they think the police ‘should’ be doing – but the public reception to this policy shift towards perpetrators should be given serious consideration. The criminal justice system has gone some way in the last 30 or 40 years to more fully address victims’ concerns and needs. With police reorientating their focus more and more towards perpetrators, potentially overriding the wishes of victims, it is possible that victims lose even more confidence in the criminal justice system. However, it is important to the public that the police and other agencies do take domestic abuse, and the perpetrators of it, very seriously (Hudson, 2002; see also Coker, 2004). It is important that the impression is not gained that the police and policy makers see victims as ‘inadequate’ in some way – responsible for the low conviction rate in not supporting prosecutions, for example – but equally important that serious offenders have their behaviour addressed, challenged, reduced and curtailed. Assessing risk, identifying the most serious perpetrators and knowing which strategies work most effectively are vital for reducing re-offending, managing offending behaviour, improving victim–survivor engagement and reducing operational policing costs. However, the new shift towards the most frequent and most harmful offenders is only welcome if it can be achieved without again pushing victims into the margins.