The $24m 'illusion' of programmes to stop family violence
Monday, 15 June 2026
Every year, New Zealand state agencies make more than 9,000 referrals to stopping-violence programmes.
Two new studies asked victim-survivors what actually happened when their abusers attended those programmes.
More than 70% reported their abuser's behaviour deteriorated in at least one way during the programme or shortly after.
Taxpayers spend tens of millions of dollars on the programmes every year, yet agencies admit they lack evidence on what works. Efforts are underway to address that.
A landmark new body of research suggests New Zealand may have spent decades measuring the wrong thing – and asking the wrong people – about family violence. Paula Penfold investigates.
Fourteen men are seated in chairs around the edge of a small, ordinary room somewhere in Auckland, confronting their violence.
Some are here because they’ve been sent by a judge or probation officer. Others were referred by police. Some walked through the doors themselves.
For two hours once a week, for 20 weeks, two facilitators will lead them through facing the reality of their actions; why their partner or children are scared of them.
Jayden* says his behaviour was “pretty bad” and led to one of several stints in jail. He says after a “wretched” upbringing he didn’t know any better. The first time he came here, he wanted to leave straight away.
But he realised he was a “walking time bomb,” so he came back, and began to speak to the group. “I just think this course brings out the vulnerability in men,” he tells Stuff. “Men aren’t normally vulnerable, we hold walls up. You could be the toughest guy in the world, but we all still bleed red and we break.”
Jimmy* was assessed as not being suitable for group sessions, so he’s here for one-on-ones, after a “lot of yelling and abuse” led to a protection order against him. Part of the order was that he had to come here.
“I was pretty egotistical. I was like, ‘I don’t need this. They’re not going to tell me anything I don’t know. They don’t know me’. I was very defensive.”
He says he’s learning not to be. “Just being able to understand my anger. Anger is not the actual emotion. There’s all these underlying factors. Being able to see myself on the tension scale and recognise when I’m starting to get heated up and frustrated, and being able to cool myself down.”
The organisation where we’re talking to these men has been providing stopping-violence programmes for more than 20 years, and it meets all the tests of what the sector says leads to good outcomes: the facilitators are qualified; participants can get one-on-one as well as group therapy; feedback is sought from their partners.
But not all programmes are created equal.
Every year, New Zealand state agencies send more than 9,000 people into stopping-violence programmes. Men account for nine out of every 10 referrals.
Across the country, there are dozens of providers and hundreds of programmes, and for decades they have been a key state response to family violence.
Last year, mandated programmes funded by the Ministry of Justice and the Department of Corrections cost more than $24m.
Yet inside official documents is a bureaucratic admission.
At the end of the programme the facilitator will fill out a form – an FVPP05, a completion report.
The number of sessions attended will be recorded, along with whether the participant acknowledged responsibility for their behaviour, had undertaken sufficient “empathy and victim impact work” and had a safety and relapse prevention plan.
The form will then be sent back into the belly of the justice system and can be considered in decisions such as day-to-day care of children, and sentencing.
But a crucial voice can be missing.
When an abuser is referred to a programme, providers are required to proactively contact the partner for a “victim-informed assessment” of their safety before, during and after a programme.
There is no central collection of data on those assessments.
But several senior people in the sector told Stuff programme providers can spend “countless hours” chasing up agencies for victims’ contact details, only to be told to sign off a completion certificate without the assessment.
The box was ticked. It didn’t mean the person who was the subject of the violence ever got a call.
The issue of missing victim-informed assessments matters because it’s symptomatic of a wider problem. For decades, the underlying assumption of both the public and the judicial system has been straightforward and seemingly logical: that stopping-violence programmes stop violence.
But how is that measured? And what if the system has been asking the wrong people for proof?
Who we forgot to ask
The stakes could hardly be higher.
Stuff’s Homicide Report found that family violence featured in the majority of homicides involving adult women, and that half of adult female victims were killed by a current or former partner.
A 2019 study found 55% of women had experienced one or more types of abuse from partners in their lifetime.
Yet despite the millions spent on interventions aimed at changing the behaviour of violent abusers, there’s remarkably little effort to ask the opinions that arguably matter most: the women living with the consequences of violence.
Instead, success has largely been judged by rates of reoffending, and by what abusers themselves say about whether their behaviour improved.
Longtime family violence researcher Dr Neville Robertson says both tests are flawed. He’s been saying so for nearly 30 years.
Back in 1999 Robertson wrote a paper which posed the question: do stopping violence programmes improve the safety of the victims, or do they produce better educated abusers?
His answer then was the same as it is now: evaluations that rely on self-assessment are “useless” because abusers under-report their violence, and reoffending rates cannot measure the true scope of the problem because most assaults do not end up in the justice system.
It’s rare, Robertson says, for there to be woman-centred evaluations of stopping violence programmes.
“And that is the gold standard, as far as I’m concerned.”
In the absence of that ongoing oversight, enter two new landmark pieces of research.
Life and death
For years, Deborah Mackenzie, co-founder of charity The Backbone Collective, has been wanting to address that critical gap. So late last year Backbone surveyed 471 victim-survivors of family violence. The result is a just-published report called “Just ticking the box”.
The key group was 172 respondents whose abuser attended a stopping-violence programme, most referred by the state. The findings were stark:
- 73% said their abuser's behaviour deteriorated in at least one way during the programme or in the three months afterwards.
More than half reported worse violence and/or abuse during that period.
Only 10% reported their abuser showed only better behaviour during the programme and in the three months afterwards.
Just 4% reported positive behaviour change across all forms of abuse for at least a year after the programme ended (and didn’t credit the programme as being the sole reason).
Half of those whose abuser reduced physical violence said he increased another form of abuse.
More than eight in 10 felt unsafe while their abuser attended the programme and for up to a year afterwards.
Four in 10 felt very unsafe after the programme.
Six in 10 would not recommend the programme their abuser attended to others.
For Mackenzie, the research revealed an uncomfortable paradox. “Abusers were actually benefiting from being referred to these programmes far more than victim-survivors were benefiting in terms of their safety.”
‘The illusion of compliance’
Where official data is lacking, The Backbone Collective research goes some way towards filling the vacuum.
Nearly three-quarters of respondents said the provider of the programme their abuser attended did not try to contact them. Only 20% said the provider had.
Alongside Backbone’s online survey, the Papakāinga Trust carried out qualitative research – wānanga and talanoa with Māori and Pacific women.
Led by Dr Daysha Tonumaipe’a (Te Arawa, Tainui, Ngāti Hine, Cook Islands Māori/Aitutaki), a research fellow at the University of Auckland, the findings pointed in the same direction. Tonumaipe’a calls it “the illusion of compliance”.
“The general feeling and experience was they [programme participants] are just getting better at the language,” Tonumaipe’a says. “The biggest message we’ve had is that accountability should be based on safety, and not completion of these programmes.
“Why are you assuming that attendance equals safety? Please measure safety.”
An unlikely consensus
This is probably the part of the story where you’re thinking programme providers will righteously defend their own work.
That’s what I thought, too, when I rang Merran Lawler, chief executive officer of Te Kupenga Whakaoti Mahi Patunga, the National Network of Family Violence Services.
She found The Backbone Collective report “depressing reading, but also completely unsurprising”.
There’s a reason she’s not defensive.
“At one level the report could be read as attacking the credibility and the viability of stopping violence services and programmes,” she says, “but it’s far more nuanced than that.”
She cites a key date: 2014, the year Programme Approval Panels were abolished. Made up of victim-survivor advocates, alongside organisations and experts in stopping violence, their job was to approve programme design and content.
Their abolition was a philosophical decision, Lawler says, and a financial one. “To get the cheapest possible services from a wider range of providers.
“It’s fair to say that it has resulted in the dumbing down of the sector. The safeguards were much better at the turn of the century than they are today.'
Lawler argues there is now less focus on qualified facilitators, and more on lived-experience practitioners – people who formerly perpetrated violence.
“Groups of guys getting together and being able to share what’s happening for them, and removing any of the discomfort that necessarily goes with any type of behavioural change.”
And too often without their victims being asked if their abuser’s behaviour has actually changed at all.
What the system says
You might also have expected government agencies to be defensive in the face of the Backbone findings.
None of those contacted by Stuff were. In fact, the said, a cross-agency review of the effectiveness of stopping-violence programmes was delayed in order to allow the research findings to be incorporated.
The review is part of Te Aorerekura, the Government’s 25-year family violence strategy, and has already delivered a sobering finding consistent with the Backbone report: that victim and survivor perspectives are not prioritised as highly or as consistently as they should be.
The Ministry of Social Development’s general manager for Safe Strong Families and Communities, Mark Henderson, went a step further, acknowledging that the state fundamentally lacks evidence on what works.
“We are contributing to efforts to address this,” he said. “… the overall evidence base for stopping-violence interventions is uneven and lacks the mechanisms required to assess sustained effectiveness.”
There was another key admission: “Yes,” said a spokesperson for the Executive Board for the Elimination of Family Violence and Sexual Violence, which is leading the review of stopping-violence programmes. “We are aware that people who use violence may use course completion as evidence that they are safe, despite continued use of coercive control and other forms of violence.”
The Board said that highlighted “the need for more comprehensive responses to people who have used violence”.
Principal Family Court Judge Jackie Moran said the Backbone research was welcomed by the judiciary. “The report is complex and raises several issues which deserve careful consideration.”
One such issue, that of programme completion being signed off without victim input, was acknowledged by chief operating officer at the Ministry of Justice, Carl Crafar.
“We are continuing to work with sector partners to better understand these challenges.”
In the meantime, he said programme completion was only one piece of information considered, and facilitators with concerns about ongoing or escalating risk could formally communicate those to the courts at any stage.
Decisions from the cross-agency review on priorities for stopping violence services are expected to hit ministerial desks between July and August this year.
So what would work better?
Women who responded to the Backbone survey described what safety felt like to them.
Neither Backbone nor Papakāinga is arguing for the abolition of stopping violence programmes; they say that inference would miss the point entirely.
What they do want is for the state to “stop expecting them to be the magic pill,” as Mackenzie puts it. “It’s not helping abusers, and it’s putting women and children in greater danger.”
The Backbone report makes 20 recommendations, including:
Instead of programme attendance being used as a proxy indication of behaviour change, an evidence-based family violence risk assessment tool is developed, giving victim-survivors input.
A requirement for programmes to be delivered by specialists.
The state should identify repeat abusers and escalate management if necessary, rather than repeatedly sending them to the same or similar programmes.
Ongoing and frequent monitoring of abusers.
The state should protect victim-survivors from abusers whose behaviour has not changed.
You can read the report here.
I put it to Lawler all that sounds expensive.
'They only need to do an analysis of one family violence case that involves the victim ending up being killed,” she says, “to realise that actually they can't afford not to invest. The cost pales into insignificance when we compare it to the actual cost of the loss of life, and imprisoning somebody for years and years for having committed a heinous crime.”
When the intervention does work, it can be life-changing.
Jimmy and his partner managed to keep their relationship together. He gets to parent his two boys.
After arriving convinced he didn’t need help, he now comes back every second week, long after his court-ordered sessions finished.
“I’m just so thankful,” he says. “It’s probably the best achievement of my life.”
Some of those who have worked in this sector for a long time and have seen those results are worried that even with widespread acknowledgement of what’s not working, there’s no urgency.
At almost 75 years old, Robertson has come to a conclusion about that.
“I don’t think we care that much about what happens to women, frankly,” he says. “When I started in this business I hardly understood what misogyny was. But the further I went, the more I came to understand that we live in a deeply misogynistic society, and we place bugger all value on women.”
Once, he hoped the problem of violence against women could be fixed in his lifetime.
“Now, I’m hoping it will be in my grandchildren’s lifetime.”
*Names have been changed.
Where to get help
Rape Crisis 0800 88 33 00, click link for local helplines.
Victim Support 0800 842 846.
Safetalk text 4334, phone 0800 044 334 webchat safetotalk.nz or email support@safetotalk.nz.
The Harbour Online support and information for people affected by sexual abuse.
Women’s Refuge 0800 733 843
Male Survivors Aotearoa Helplines across NZ, click to find out more (males only).
If you or someone else is in immediate danger, call 111.
If you or someone you know is in a dangerous situation, click the Shielded icon at the bottom of this website to contact Women's Refuge in a safe and anonymous way without it being traced in your browser history. If you're in our app, visit the mobile website here to access Shielded.