Shifting Power at the Front Door: State–Community Decision-Making Partnerships in Child Protection

Abstract

Inequities for Indigenous people in child protection systems are well established. One avenue for addressing these inequities is at the ‘front door’ of child protection, when reports are first made to statutory child protection services. This article reports on a formative evaluation of a shared decision-making forum in a small city in Aotearoa New Zealand, where a community Māori organisation meets to make initial decisions about reports together with the statutory agency. The aim is to improve information quality by bringing local, relational knowledge to the decision and provide a service response to those cases that are ‘closed’. The findings are that initial enablers were the policy context that emphasized community devolution, consensus on problems and aims, relationships between leaders in both organisations, and high community investment. Early challenges were a reluctance from some workers to engage in the process, lack of agreed processes, and fears of simply replicating the statutory agency in the community. Current enablers following a period of establishment were relationships of trust, the development of practice processes, commitment to review, increased information sharing, community location and leadership, and an alignment with practitioners’ values. Challenges were conflicts about moderate risk situations, lack of other key services, inconsistent attendance, and authority conflicts over legal mandates and information sharing practices, especially relating to high-acuity situations. The implications are that organisational, policy, and resourcing level changes and relationships from front-line workers to leaders are essential for moving institutional logics.

1. Introduction

Efforts to devolve child protection systems, reduce disparities for Indigenous people, and improve early, supportive responses to families facing difficulties are challenging in many countries. Both decolonial and abolitionist advocates argue for a radical approach that dismantles the interconnected web of statutory agencies that rely on surveillance, bureaucratic compliance, investigation, and child removal and instead instigate community-led systems of care that provide early supports, poverty reduction, and relational care for the whole family (Hyslop and Pease 2025; Waitangi Tribunal 2021; Dettlaff et al. 2020). While less radical in nature, reforms that address the outcomes of poverty and move the bulk of resources into the community to enact a public health approach to child protection have also been continuously promoted as a way to effectively address the causes of child harm and reduce inequities (Lonne et al. 2021; Pryce et al. 2019). There are in practice some substantial crossovers between these approaches, particularly at the initial phases of change (Pryce 2025; Keddell 2025). Detailed descriptions of how these aims could be achieved are less common, with little attention to the strategic, staged process of change that might bring such ideas into fruition nor to the important contextual considerations that shape their eventual implementation.

One place to begin to address disparities and disrupt such norms is at the point reports about children’s care are first made to the child protection agency. In this context, decision-making about reports of concern is important, as they are the initial entry point of the family into the system and often represent some kind of emerging family issue or problem, even though most do not meet the threshold for legal intervention (Font and Maguire-Jack 2015). They also begin the creation of disparities for Indigenous people, racialised minorities, and those living in high-deprivation areas, beginning the expression of social inequities within the child protection system (Hood et al. 2019).

Far from being neutral, the processes of deciding how to respond to reports contain sense-making processes that inevitably reflect particular ways of construing risk and interpreting information and have defined consequences. Decisions are often reliant on superficial information and are made under time pressure, have a narrow bureaucratically defined range of possible outcomes, and seldom release timely, welcomed, and supportive resources. There have been many critiques of this assessment process as superficial, expert-driven, bureaucratic, and procedurally focused (Lonne et al. 2021; Font and Maguire-Jack 2015). Initial decisions have also attracted critique for perpetuating racist stereotypes and implicitly promoting Eurocentric parenting norms, with a focus on risk rather than protective factors, including those relating to cultural context and continuity (Boulton et al. 2018; Lonne et al. 2021; Williams et al. 2019; Wilson et al. 2016; Wright et al. 2024 Keddell and Hyslop 2019).

There is often a disjuncture between espoused polices and guidance and how they are applied in practice. Institutional logic theory provides way of understanding how differences in values and organisational norms come to dominate decision-making, often overriding explicit ideals (Lounsbury and Boxenbaum 2013). Dominant institutional logics are achieved in practice through procedural compliance, recording processes, greater attention to particular sources of information than others, and the preferring of particular cultural and theoretical narratives (Thornton et al. 2008). Institutional logic theory helps explain how disparities and oppressive practices can persist despite rhetorical attempts to reduce them and highlights the need to disrupt organisational cultures if change is to be implemented successfully.

The resistance of Indigenous people to state over-intervention, along with broader critiques emphasising the need for community partnerships and preventive resourcing in child protection systems, has led to collaborative decision-making processes aimed at broadening decision-making control at the initial decision point (Chamberlain et al. 2022; Pryce et al. 2019; Waitangi Tribunal 2021). The aim is to incorporate Indigenous and community knowledges, disrupt knowledge hierarchies, and improve access to early support services for struggling families as a first step in devolving power and resources. This article reports on one example of a collaborative decision-making process implemented in a small city in Aotearoa New Zealand, where a community Māori organisation partnered with the statutory agency (Oranga Tamariki) to establish a shared decision-making process to improve initial decisions and responses to whānau (extended families). Understanding the contexts, enablers, and challenges to shifts in power and practice is important to help provide the evidence base for reconstructing child protection systems. This is needed in order to better understand what contributes to successful systems change as ‘non-reformist reforms’ (those that lead to more substantial reforms) and disrupts institutional logics (Asher BlackDeer and Harty 2025; Keddell 2025).

1.1. Disparities and Responses

Disparities in the child protection system for Indigenous people are well documented internationally and locally (Bussey and Lucero 2013; Haight et al. 2019; Chamberlain et al. 2022; Boulton et al. 2018). Disparities are driven by underlying inequities relating to colonisation, including over-surveillance, assimilationist ideologies, land alienation, poverty, cultural denigration and institutional racism, and the resulting social determinants such as income inequality and a lack of access to economic and social resources (Paki Paki et al. 2024; Keddell and Hyslop 2019; Keddell 2022; Waitangi Tribunal 2021; Information Gateway 2024). Inequities have negatively impacted whānau life and traditional social structures such as hapū and iwi and created biases in system responses (Abuse in Care 2024; Keddell 2022; Waitangi Tribunal 2021; Pihama et al. 2017). Research both in Aotearoa and in other settler-colonial states notes that compared to other families in contact with the child protection system, Indigenous families often face serious social challenges combined with few effective and available services (Cram et al. 2015; Paki Paki et al. 2024). Due to a history of over-intervention and racism when accessing public services, trust and engagement in existing services may be low (Haight et al. 2018; Kaiwai et al. 2020).

Reports to statutory organisations, and their outcomes, are also shaped by the resource environment. Reports are more likely to proceed to further system intervention when service accessibility outside the statutory system is challenging and there is a lack of culturally resonant and well-resourced preventive services, particularly for Indigenous families (Hood et al. 2020; Fluke et al. 2010; Font and Maguire-Jack 2015; Hood et al. 2019; Haight et al. 2018; Paki Paki et al. 2024). Providing effective assistance to whānau who do not reach the threshold for statutory intervention, yet are reported multiple times, is a key aspect of preventing harm, re-reports, and care entry (Toikko et al. 2023; Trocmé et al. 2014). Re-reported families often have persistent social problems such as poverty impacting their parenting capacity that are not addressed by re-reporting, and multiple reports may further estrange them from services (Esposito et al. 2025). Reduced workforce capacity within the statutory agency exacerbates this, with evidence that reduced workforce size within statutory services also increases re-reports (Esposito et al. 2025). The workforce of Oranga Tamariki within Aotearoa NZ was reduced by 6% in 2024 due to ongoing public service cuts at the same time that significant cuts were being made to community services (Hanly 2025; Gibbens 2024). Report ‘churn’, in which multiple reports of concern are made over time about a whānau, may also mean that those making reports of concern lose faith in the system where they see no action being taken, leading to disillusionment (Keddell et al. 2024).

1.2. Aotearoa New Zealand Context

The child protection system in Aotearoa New Zealand follows the notify–investigate system of other anglophone and settler-colonial nations, with a centralised, government-run, statutorily empowered child protection agency and a wider network of community-based NGO organisations that are also mostly funded by government. Some of these are run by Māori entities, whether iwi, hapū (tribal), or pan-Māori. Reporting concerns about child abuse and neglect to the central agency (Oranga Tamariki) is not mandatory, but all organisations in contact with children are legally required to have reporting guidance in place that encourages reporting to Oranga Tamariki for situations involving risk of harm to children. Similar to other settler-colonial countries, disparities have been extensive for Māori yet have had undulations over time due to changing social and policy environments, with current disparities at the point of first report being 3.2 times the non-Māori rate (not controlling for socioeconomic or other factors) (Waitangi Tribunal 2024; Ministerial Advisory Committee 1988; Keddell et al. 2022; Oranga Tamariki 2023a). Between 2020 and 2023, following legislative, policy, and practice changes, both entries to care and disparities for tamariki and rangatahi Māori began reducing (Oranga Tamariki 2024b). Since then, there has been a fluctuating policy environment, including the repeal of s7AA of the legislation, (that required decisions and practices to recognize the Treaty of Waitangi, allowed for partnerships between the state and Māori, and created a legal onus on the state to measure and report on disparities for Māori), cuts to community service funding, and general undermining of the Treaty of Waitangi in all legislation. Reports to child protection and care entry have begun to rise again (Wysocki et al. 2025). However, before these changes began, some possible causes of the reduction in care entry and reports was an increased emphasis on community-led decision-making; a stronger policy and practice emphasis on whānau preservation within Oranga Tamariki; resourcing of some community organisations to work with whānau utilising Māori practice approaches; and formal iwi (tribal) and Māori organisation partnerships under the Oranga Tamariki Act 1989 and enabling community initiatives (Oranga Tamariki 2021, 2023b).

Partnerships of various kinds with iwi and other Māori organisations have established new practices aimed at maintaining whānau connections, reducing disparities, and improving early help pathways for whānau who are reported, and this may be changing decision-making practices at the ‘front door’ of the child protection system (Roguski 2020; King-Howell et al. 2025). In a number of trial locations, iwi, hapū, or Māori organisations are utilising negotiated decision-making practices with Oranga Tamariki as soon as a report of concern is received. These practices aim to shift or share early decisional power with iwi/Māori organisations and enable access to effective prevention services that already have whānau connections with people reported (King-Howell et al. 2025). This process shifts practice towards tino rangatiratanga in a practical way through beginning the process of restoring authority to iwi Māori over decisions early in the decision-making continuum. A further aim is to improve pathways to effective services for families who do not meet the threshold for legal intervention at the time of reporting.

Emerging evidence of community-led decision-making has been promising. One study found that iwi-led family group conferencing enhanced the trust of whānau, improved whānau participation, provided a trusted mechanism for the authentic participation of young people, and led to improved outcomes for children (Roguski 2020). Other factors affecting positive outcomes were a strong relationship between the local site office of Oranga Tamariki and the iwi, particularly where this was supported by committed champions from both organisations. This ensured the protection of iwi autonomy and promoted rangatiratanga (Roguski 2020). A process evaluation of mana mokopuna, the partnership between Waikato Tainui and Oranga Tamariki found that it was preventing care entry and widening the pool of potential whānau caregivers in the Tainui region (tribal area). Fundamental to this was strong and open leadership and collaborative planning. For whānau, a greater focus on whānau-led practice was evident, leading to a greater sense of empowerment in a system that was historically marginalising. It also helped to shift entrenched attitudes and practices of Oranga Tamariki staff (AIKO 2018; Morunga 2022). This partnership has gained significant funding, and in 2021, 203 out of 261 reports exited from Oranga Tamariki, while in 2022, of 173 removal applications that went to the family court, only 4 resulted in removals (Love et al. 2019; Morunga 2022). From 2019 up to 30 June 2024, in the Waikato Rural North region, where the partnership is most embedded, the number of children in care decreased by 51% (King-Howell et al. 2025). Ngāti Toa entered an s7AA partnership with Oranga Tamariki under the Oranga Tamariki Act 1989 in 2020. As a result of this partnership, reports of concern, ‘further actions’, and removals have declined, and a family group conference process was co-designed with Oranga Tamariki that draws on Ngāti Toa tikanga and mātauranga Māori (practices and knowledge) (Ngāti Toa 2024, 7 August). While there has not yet been research regarding effectiveness, feedback provided to the iwi by both whānau (families) and kaimahi (workers) suggests that this process has helped to maintain children within their whānau and assisted in upholding the mana of whānau and the iwi itself (Ngāti Toa 2024, 7 August). Meanwhile within Oranga Tamariki, the new positions of kairāranga have been increasing the focus on Māori concepts of whānau and the capacity to find whānau members, which together with the new practice framework have been helping drive the wider practice shift of the statutory organisation towards whānau preservation and devolution to community and iwi/Māori services (Love et al. 2019; Stanley 2024; Oranga Tamariki 2024a).

Report decision-making is often limited to the administrative information recorded about a family, combined with the report information, which can lead to poor information quality. This effect is exacerbated in centralised systems where intake workers are not embedded in the communities within which the reports are received from, as is the case with the intake centre in Aotearoa New Zealand. A lack of relational and relevant knowledge of the community may reduce accurate appraisal of the family situation and restrict te ao Māori concepts relating to whānau and wellbeing from the knowledges used to construct the family issues, strengths, and best solutions (Boulton et al. 2018; Chamberlain et al. 2023; Kaiwai et al. 2020; Pihama et al. 2017). To prevent all these impacts, more effective responses at the front door of child protection systems need re-design (Hood et al. 2019; Lonne et al. 2021. Addressing these issues at the point of report should help strengthen whānau, reduce child harm, and reduce the need for legal intervention. This could also improve confidence among potential reporters that their reports will lead to supportive action for the whānau involved (Haight et al. 2018). However, there is also the potential for barriers to devolution at this point, including underdeveloped practice processes, uneven implementation, lack of capacity, loss of community trust, and unclear mandates or legal responsibilities.

This research project examines one of these trials of changed decision-making at the point of report. Tai Timu Tai Pari (a collective of several Māori community agencies) alongside Oranga Tamariki have constructed a new decision-making process at the point where reports of concern are received by local site offices, post-triage at the national intake centre. Referred to as ‘the tēpu’ (literally, the table), Tai Timu Tai Pari and Oranga Tamariki come together multiple times each week to negotiate decisions and responsibilities, either via a statutory response from Oranga Tamariki or a community-led response via Tai Timu Tai Pari and its community partners. Drawing on expert stakeholders and whānau who have experienced this decision-making model, this process evaluation describes in detail the context, enablers, and challenges of this form of decision-making.

1.3. Research Questions

How does the new triage decision-making process affect decision-making and outcomes for children and whānau?

What are the barriers and enablers of community-led decision-making and more effective, whānau-centred outcomes for tamariki and whānau?

2. Methodology

This project uses a qualitatively driven mixed-methods approach to formative evaluation, nested within a methodology comprising Kaupapa Māori and participatory evaluation principles and processes. It has a qualitative core, meaning the qualitative data are most prioritised for their explanatory value, while the statistical element highlights changes to the pattern of reports, adding an additional lens to the qualitative findings (Hesse-Biber et al. 2015). This article reports qualitative findings only. Kaupapa Māori research principles prioritise power-sharing over research aims and processes and focuss on community partnerships and relationships, as well as values of mana, manaakitanga, (mutual support) whānaungatanga, rangatiratanga (sovereignty), and utu (reciprocity) (Goings et al. 2023; Smith 2022; Walker et al. 2006). Kaupapa Māori research methods focus on research driven by Māori aspirations and processes, utilise Māori tikanga in research design, and leave all parties better off for having participated in the project (Pipi et al. 2004). Participatory research in community settings expresses similar values, with an emphasis on community-defined needs and research that is embedded in the process of service design and implementation, positioning community members as the experts and leaders (Wallerstein et al. 2018; Koster et al. 2012). As Koster et al. note, research in Indigenous community contexts should continue to reject research methods that are extractive, position the researcher as the expert, and reflect traditional research methods that work ON Indigenous communities rather than “WITH and FOR them, based on an ethic that respects and values the community as a full partner in the co-creation of the research question and process, and shares in the acquisition, analysis, and dissemination of knowledge” (p. 195).

The expression of these principles and practices in our study is evident in the invitation from Tai Timu Tai Pari to University of Otago researchers, based on pre-existing relationships, to undertake the evaluation to help further their own aspirations of systemic change within child protection systems and to develop community-led child protection services. Principles of reciprocity and whakawhanaungatanga (relationship building) were established through multiple hui (meetings), through which the research aims and processes to be used were established. As the tēpu was a pilot that was ongoing, the participatory process evaluation model was ideal. Due to the relational nature of kaupapa Māori research, there are often blurred boundaries between researchers, research partners, and participants, as we all co-collaborate in knowledge production (Oates 2020). Research processes were iterative, with several opportunities for researchers and participants to reflect on and shape the findings. These facets were reflected in team members occupying multiple roles as researchers and participants and iterative analyses, including a co-analysis hui, relationship-building hui, formal interviews, focus group, and ongoing publication hui. Role clarity assisted with managing real and perceived conflicts of interest in the analysis of findings. The research team comprises people from within the university, Tai Timu Tai Pari, and Oranga Tamariki. The research team are a combination of Māori and non-Māori and bring expertise across a diverse range of areas, including social work management, policy, practice, research, and local knowledge of whānau and hapū. Ethics approval was obtained from the University of Otago ethics committee, the Research Access Committee of Oranga Tamariki, and the consultation process with the Ngai Tahu research consultation committee was completed. Informed consent was obtained from all participants involved in the study.

After relationships were established, professionals and whānau involved in the new decision-making process were invited to take part in focus hui and interviews that took place in 2024. The focus group incorporated tikanga as appropriate and finished with shared kai. Interviewers were both Māori and Pākehā. All interviews and the focus group were recorded and transcribed and then returned to participants for checking. Analysis of the qualitative component of this study was by primarily deductive thematic analysis framed by our research questions, with some limited inductive analysis, for both professional and whānau member data. Following initial analysis of the emerging themes, two online co-analysis hui were held with participants to enable further reflection and development of the themes with participant input in late 2024 and early 2025. Further, an online feedback tool via padlet was enabled and shared with all participants to enable asynchronous feedback if they could not attend the hui. Ongoing hui following these also contributed to the analysis process beyond these formal meetings to embody iterative sense-making. This helped promote researcher accountability to the community and participants.

Participants

Participants were twenty-one employees of either Oranga Tamariki, Tai Timu Tai Pari, or one of the ‘member organisations’ of Tai Timu Tai Pari who provide community workers to attend the tēpu. Adult whānau members who had been through the new triage process and who had also had involvement with Oranga Tamariki were contacted by community workers operating under the umbrella of Tai Timu; however, despite multiple efforts by many workers and flexibility in interviewer, place, and medium of interview, only one whānau interview was completed, making a total of twenty-two participants. The worker participants held a range of positions from frontline workers to supervisors to managers. Fifteen participants were Māori, six were Pakeha, one was Pasifika, and one identified as ‘other’. Sixteen were interviewed, and thirteen attended the focus group (some undertook both).

3. Findings

Findings are divided into initial context enablers and challenges and current enablers and challenges. Initial context enablers and challenges are those elements that contributed to or challenged the tēpu’s inception, while the current enablers and challenges refer to what was found at the time of the project, some eighteen months later.

3.1. Initial Enablers and Challenges

As can be seen in

Table 1. Enablers and challenges to the tēpu starting.

Enabler: Shared Problem Definition	NFAs Having no Support Re-Reported Families Unnecessary Escalation to Oranga Tamariki Poor Information Quality
Translated into: shared aims	Improve responses to NFA whānau Improve information quality Share decision-making power Reduce disparities through empowering community
Other key enablers	Socio-political context—s7AA Local context relationships and leadership Sense of investment in whānau
Key challenges	Lack of shared process Resistance from some staff Concern about becoming another Oranga Tamariki

, a key contextual factor was having agreed problem definitions and shared aims for the tēpu between two key stakeholders: Oranga Tamariki and Tai Timu Tai Pari leadership. Agreement between state and community participants that the problems were that closed cases received no immediate support, leading to re-reports and escalation, and that reports often contained very limited information about the family, leading to poor decision-making, were clear areas of consensus.

This consensus is evident in these two examples, one from a statutory worker and one from a community participant:

“We had a very high NFA rate, which is no further action. We also had a lot of whānau returning to Oranga Tamairki quite often, but it wasn’t necessarily meeting our threshold. So there was a need for someone to come in, a project to assist”

P11 Oranga Tamariki.

“…they saw themselves they didn’t have enough staff to go out and see all these whānau, some of the whānau getting NFAed were really screaming out for help”

P10 Community

Another point of consensus was that decisions made at the point of report were often based on superficial information based only on the report of concern and other information Oranga Tamariki knew about them. This could lack depth of knowledge of the whānau, their relationships and whakapapa, and be out of date. Oranga Tamariki making initial decisions alone also consolidated their power to make those decisions and limited community power and information input:

“because who better to help these vulnerable whānau than our own village? From my understanding it was a way for us as the iwi, as community, to help our own people in a way that Oranga Tamariki couldn’t. Because we know our whānau best”

P13 Community.

A higher-level aim was to reduce disparities for Māori in the child protection system by reducing unnecessary escalation to Oranga Tamariki:

“I think the aim is to reduce disparities for tāngata whenua, to give iwi an equal say in what’s happening to their tamariki and whānau”

P8 Oranga Tamariki.

These shared aims required some specific contextual conditions that enabled them to bloom into the tēpu being established. A key contextual factor was the political and social context of the time. Multiple inquiries into Oranga Tamariki and critique of its over-intervention, particularly for Māori, the Hawkes Bay case, the Waitangi Tribunal inquiry into Oranga Tamariki, and the introduction of s7AA into legislation all set the conditions for practices such as the tēpu to emerge (see Waitangi Tribunal 2021; Keddell et al. 2022). Tai Timu Tai Pari was created partly as an outcome of the adoption of the He Kahu Aroha report within Oranga Tamariki, which set a clear focus on community devolution (Oranga Tamariki 2021; Waitangi Tribunal 2021).

Beneath this policy environment, the specific organisational environment in the local area was also important. It is a small city with only three child protection sites, a large Māori population, and a web of pre-existing links between the various hapū organisations and staff within Oranga Tamariki. Many leaders and kaimahi within all organisations had other pre-existing relationships, either professionally, personally, or via whakapapa and whānau. These relationships provided a foundation of goodwill and trust between key actors.

Several high-profile community leaders were helpful to enable the emergence of the initiative because they held high trust and mana in the community, ensuring community investment. Particular leaders within Oranga Tamariki were also influential, pulling the impetus towards community power-sharing despite others within the organisation being less willing. These champions helped drive the initiative forward and created an initial bridge of trust:

“the profile of Tai Timu Tai Pari within the community is really positive as well, and there have been instances where other services have said, ‘how do we become a part of Tai Timu Tai Pari ?’. So there’s that push within the community seeing this as a positive process, which it is, and wanting to get on board … And I think that’s massive”

FG1

Finally, this sense of shared mission was not only related to professional relationships but also contained a strong sense of investment in local community and the whānau:

“Our passion, I think. All of the people that are at the tēpu that are a part of Tai Timu Tai Pari, not just people at the tēpu, Oranga Tamariki regional office, and their support and drive to get it done, has been huge. It’s been a couple of years in the making…the people who do that mahi, their passion to get it up and running was one of the biggest enablers. I think also being a small community, we are all very passionate and connected to our community. So we have that investment in our community and our partners, and in those whānau.”

FG

A consensus on the problems of ‘closing’ cases without offering help to whānau, and limited information, within the context of a devolutionary policy environment and multiple positive relationships among leaders provided fertile soil for the tēpu to gain traction. A strong sense of community investment and care for whānau helped form a strong basis for action.

3.2. Early Challenges

While these contextual factors contributed to the tēpu starting, there were also some immediate challenges. As with many new organisational forms, there was not initially an agreed process of how the tēpu would operate, which at times led to confusion. There was a reluctance from some front-line Oranga Tamariki staff to engage and share information and power with the community workers. Some explained this could be because they were not adequately informed about what the tēpu was and how it would work, while others suggested it was more about resistance to the changing power relationships and lack of trust in community capacity. Lack of attendance from both Oranga Tamariki and community workers also challenged the initial consistency needed to progress decisions:

“Heaps of barriers. Heaps of barriers from Oranga Tamariki staff, some sites, when they didn’t want to… they didn’t want to play in our sandpit, just ‘they own the sand pit’, they thought… They just were not coming on board…they just wouldn’t turn up, or they didn’t think it was important enough, or they wouldn’t bring everything that we agreed to …they were … stalling…but there were also barriers within ourselves. So, in terms of our own collective providers within Tai Timu Tai Pari who were supposed to be sitting at the table. So we had barriers as well”.

P12 Community.

3.3. Current Enablers and Challenges

Following this period of establishment, at the time of the research, the tēpu ran with a related, but different, set of enablers and challenges, as outlined in

Table 2. Current enablers and challenges of the tēpu.

Key Enablers	Key Challenges
Relationships as glue and balm	Threshold clashes about moderate risk, and particular case types.
Development of practice processes: consistency, consensus, bring backs, joint working	Inconsistent attendance
Review	Authority conflicts around information sharing and legal mandates
Information sharing and recording	Lack of services
Community location and leadership
Alignment with practitioners’ values

. Some initial teething problems faded away, while other issues came to the fore. The following section covers the current enablers and challengers.

3.3.1. Relationships and Whanaungatanga as Glue and Balm

The centrality of relationships, based on whanaungatanga, operating at multiple levels is a key enabler of the tēpu to operate in its current form. This includes relationships between organisational leaders and between the kaimahi who work directly at the tēpu. For example, this person comments on the ability of leaders to communicate and troubleshoot emerging issues:

“Not so much involved, like sitting at the tēpu, but the likes of (Chief Executive Tai Timu Tai Pari) and (Chief Executive Te Uri ō Hau (one of the contributing organisations)) and all that, and having that relationship with Oranga Tamariki, if something’s not going well or something needs changing, to have that good relationship and be able to voice that with Oranga Tamariki. And likewise…”

P3 Community (parentheses ours)

The role of trusted facilitation and how it contributes to relationships and depth of discussion was particularly noted:

“we have a really fantastic relationship with x, who leads the tēpu. I think the difference because we already have that relationship, we have that respect for each other. We’re not always perfect, but we do have the respect for one another to be able to beg to differ, but see each other from each other’s point of view”

P8 Oranga Tamariki

Relationships were central to both strategic leadership and direction, as well as front-line practice. Similarly to establishing the tēpu, relationships assist the tēpu to have robust conversations and work collaboratively, developing over time.

3.3.2. Developing Practice Processes: Consistency, Consensus, Bring Backs, and Joint Work

The development of practice processes enables the table to have agreed ways of working, overcoming initial uncertainty. Facilitation and consistency of membership are key to the quality of the tēpu decisions and practice development:

“I had to set up the process with Oranga Tamariki, and all us providers, being clear about, and actually naming, who is going to be at the tēpu and on which dates? … who is going to facilitate? All that sort of stuff, I had to hold them to account. Even with our partners (community orgs), like, you have an obligation. This is part of what we agreed to, to sit at the tēpu.”

P12 Community

This basic requirement is intertwined with the previous theme of the importance of relationships and trust, as these are the fruit of consistent attendance. Appointing a facilitator, who was from the community, helped strengthen this, as that person was well-respected, and this choice also clearly established the tēpu as a community-led initiative.

Consensus decision-making has also developed over time as the tēpu norm. There are efforts to incorporate views from those who wish to share them and a commitment to continue with discussion until consensus is reached, recognising their shared responsibility for the outcome. This is not always easy, but participants expressed this was the best way to make robust decisions:

“We’ve learnt to just have our say, and we have to be able to say it with honesty, back it up, and just have a different viewpoint, and know that neither of us on our own can make a decision. We have to be able to talk about it, and stop our excuses”.

P6 Community

Another emerging practice process is that of ‘bring backs’, where a case can initially go the community, but they may ‘bring it back’ to the tēpu and to Oranga Tamariki if they feel it is too serious for them to work with, without making a new report. This helps reduce the risk of taking on a case but then discovering it is beyond community capacity, which encourages community risk aversion:

“if there’s a case that they’re going to go and have a cup of tea, we’ll call it a bring back. So they’ll come back, and talk about what they’ve found—does it need to be escalated? Or, will they be able to manage it, or vice versa… Whoever’s going to go and get it, is going to get it, and bring it back, so it can be looked at twice”.

P15 Oranga Tamariki

A ‘joint working’ approach has also developed in grey or unclear cases, where if it is not clear who should respond, Oranga Tamariki and community workers will initially visit the whānau together to establish who should work with the whānau. Joint working can lead to some tensions between the community role and perception in the community, something community workers were keenly aware of:

“I like the relationship between Oranga Tamariki and community, and keeping that really quite solid, but we’re also not there to make them look wonderful. We’re not there to get them in the door and say, ‘these guys are perfect and wonderful and let them in’… We don’t want to say to the whānau ‘they’re going to be super nice to you’. So sort of a bridge maybe more so?”

P6 COM

The development of consensus decision-making processes, expectations around consistent attendance, bring backs, and joint working have been key to developing the tēpu approach over time, enabling resolution of conflict and more effective responses to whānau.

3.3.3. Commitment to Review and Development

There has been an ongoing commitment to review and development, including a formal review, to address issues and clarify processes as they arise. An ability to remain open to ongoing change and honestly address blocks to shared understanding and processes has been important to the ongoing development of the tēpu:

“I think we did have a number of challenges, which we covered off in a review maybe a month ago… which was timely, and that we’d not really come together as a collective, to just stop, and reflect, and go back and say, ‘okay, what’s working? What’s not? What has been an enabler? What’s here been an obstacle or a challenge?’, and it was very constructive”

FG

One of the fundamental tasks of reviews is to clarify the key purposes of the tēpu.

“I think some improvements that we’ve made in the past have been around the understanding of what the tēpu is, and the purpose for the tēpu, and that can cause conflict…so, understanding why everyone’s sitting at the tēpu is really important for community as well…”

P3 Community

Commitment to ongoing review and the ability to call reviews has helped refine both the purpose and processes of the tēpu.

3.3.4. Increased Information Sharing and Recording

Information sharing at the tēpu is a key enabler of the tēpu, as it improves the depth of information available to decision makers, and this has a consequential effect on the quality of subsequent decisions. Decisions about pathway and service type are both improved through the combination of information sources. Oranga Tamariki can access administrative information from a range of sources that community workers cannot, while Oranga Tamariki do not often have relational, up to date information about the whānau and their everyday lives. For example, in this instance, the community has asked Oranga Tamariki to go and seek further information before a decision is made:

“If we can’t make a decision, usually we’ll ask them to take it back and get more information. Usually, they can gather more information from police, or MSD, or the notifier, and then we’ll bring it back. It’s kind of like it needs that little bit of time to sort of clear your mind, and bring it back, and look at it with fresh eyes. That extra information does help”.

P9 Community

Accurate information sharing relies in turn on dedicated administrative roles who record all the tēpu discussions, outcomes, and ongoing referrals. As the community cannot access Oranga Tamariki systems, this administrative resource is necessary so community providers can retain information needed to action decisions. While information sharing has improved in scope and more reports are being brought to the tēpu, there are challenges around timing and scope for inclusion (discussed below in challenges).

3.3.5. Community Location and Leadership

The tēpu aim of community devolution is assisted through community location and facilitation of the tēpu. The tēpu has progressed from a community worker observing the three different sites’ meetings to combination into a single group and joining as a member and to community leadership, expanding community membership and bringing it physically into a community building location. This has been a powerful way to shift power into a shared model. It has also reduced the formality of the tēpu and reduced its association with Oranga Tamariki:

“we used to sit in the Oranga Tamariki building, actually moving it here to more of a neutral ground, I think it’s given it a better vibe and made it feel not so… what’s the word… not so, just that government feeling, that sort of, it’s a bit more, not so uptight”.

P9 Community

“and then it came over here. It’s cool. I quite like getting out of the office… It’s going all right, and it is where it needs to be. It’s in the community. This is the community pathway, so why shouldn’t it be in the community?”

P11 Oranga Tamariki

Workers from all organisations noted the value of a community-based process increasingly being located in and driven by the community partners.

3.3.6. Alignment with Practitioner’s Values

Alignment of the tēpu with practitioners’ values such as collaboration, whānau-centred practice and devolution has been a less visible impact of the tēpu and an enabler of it. This facet contributes to buy-in and commitment. Enabling practice to align with values helps support practitioner wellbeing, promotes personal investment in the tēpu, and avoids the moral injury commonly reported in child protection work when practice departs from this. This was noted for workers from both community and statutory agencies:

“Yeah, it does. It really does (reflect our values). Having both community and government agencies work together for the betterment of our whānau, coming up with a plan on how we can best support these whānau. I think it makes me feel good”.

P13 Community

“I study te Ao Māori, Treaty of Waitangi, have read Puao te ata tu, it’s just so in tune with my soul. …it’s what I already know. I was raised that way. So, it’s great for me to work in an organization that is very much aligned to who I am as a person … So that’s why I said, ‘who are we to tell whānau hapu and iwi… it’s about the respect”

P8 Oranga Tamariki

Alignment with practitioners’ values, both those within the community and Oranga Tamariki, assists practitioners in their sense of achievement, investment in the tēpu, and wellbeing important for professional development.

4. Challenges

As with all systems change and development, challenges arise over time. For the tēpu, challenges include managing the mismatch in expectations about moderate-risk situations, lack of basic service availability for whānau, inconsistency of attendance and capacity challenges, lack of clarity around mandates, and tensions around information sharing.

4.1. Threshold Clashes About Moderate Risk

A key challenge in any joint decision-making context in child protection is diverse risk perceptions and thresholds for decisions. There can be a particular mismatch in expectations and understandings around whānau that are neither ‘low-risk’ nor acutely severe. Community worker reluctance to accept these situations can reflect concerns about risk both to themselves and the children and the fact they do not have the same legal mandate as Oranga Tamariki to respond to these. Oranga Tamariki workers hoped community workers would accept more, especially when they felt it did not meet their threshold for investigation.

Statutory workers are exposed to acute cases constantly, which calibrates their risk norm parameters, while community workers are not, reducing their tolerance for risk:

“So it depends how high the risk is…community… and Oranga Tamariki, have different thresholds. So for something that Oranga Tamariki thinks is quite low level, is quite high risk for the community…”

P3 Community

Oranga Tamariki have statutory powers and support from others, such as police if needed, that community workers do not. This contributed to a sense of risk for community workers entering some homes and the types of roles they can undertake (for example, interviewing a child alone and undertaking investigative questioning). Community participants wanted to provide early support only to closed cases to prevent re-reports and address issues early without ‘taking on Oranga Tamariki’s work’, which they felt they had neither the mandate nor resourcing to do.

“So if we think that they’re trying to push some of the other stuff over to the tēpu, we say, ‘no, that’s something you should be keeping. You should hold on to that, because there’s a whole lot of stuff that we can’t get into’. It may be corrections, it may be mental health”

P14 Community

One way through this impasse for some whānau was via the practice of ‘bring backs’ described in a previous section and in this quote:

“I think community does pick up a lot of stuff that could be kept with Oranga Tamariki, and there is some cases where community gets worried about whether they should be picking that up or not, but we like to call those bring backs, so if community goes out first, then they can identify further action that is to be done by Oranga Tamariki, then we’ll just bring it back, and Oranga Tamariki will reopen the case. Yeah, I can see a lot of medium risk definitely going to community”.

P3 Community

The Oranga Tamariki view was quite different, feeling that it was not for the community to determine what their action should be, particularly when they recognised that there could be harm associated with their intervention. They wanted to avoid that if at all possible:

“it doesn’t make any sense to us why it’s come back to us as a CFA (child and family assessment). It’s kind of hard, because we make a commitment that whatever is discussed at the tēpu, and whatever the outcome is at the tēpu is that we will follow that. But there are times when our rationale, and our statutory threshold, is not being met. So why would we put five children through our process when it’s not necessary? …So there is definitely a gap between our understanding of what they actually do in the community, and their understanding of what we do”

Oranga Tamariki P11

Community views on the expectation they would take on many more cases at times were that this was linked to resource capacity within Oranga Tamariki, something they felt was unfairly deflecting families onto them:

“Oranga Tamariki …at the moment, their priorities are changing because they’re wanting to save money, which has become front of their mind. So, they’re wanting to pull back a lot more and send a lot more over to community. They keep talking about... that ‘it’s not their threshold’. So, I asked them, ‘so what is your threshold? … If you’re saying it’s not fitting yours, well, it’s not fitting ours either…”.

P12 Community

Underpinning this tension is workload capacity. As this Oranga Tamariki participant notes, there are simply ‘not enough’ community workers to take on some of the more complex cases; however, Oranga Tamariki are also under workload pressures, exacerbating responsibility friction:

“They can only do that if they’ve got more staff, because they’re in the same spot where we are.”

P15 Oranga Tamariki

A related source of decision pathway tension revolved around specific case types and assumptions about who should take them. In particular, what was considered to be ‘family harm’ was subject to wide and disputed definitions, and the automatic assumption that everything defined as family harm should be taken on by Oranga Tamariki was a norm that has been challenged, resulting in greater clarity around this type of whānau issue.

Fine-grained negotiations over expectations, thresholds, risk, and capacity and type of family issue have been ongoing. While all parties agree that the community would like to take more cases, practical capacities and differences in mandates and risk perceptions mean this is not always possible.

4.2. Lack of Other Services

The issue of family harm cases and who is responsible for them speaks to a wider problem, that of the lack of a range of basic public services in the north, a highly deprived area of Aotearoa New Zealand with significant lack of access to basic health and other services. Oranga Tamariki staff, for instance, also noted that they are not necessarily best-placed to respond to family harm or complex mental health needs:

“I think it’s the cases I get frustrated with are the family harm cases where I don’t believe that’s our work. I believe that there’s other agencies out there, and actually, I could say that for a whole different lot of cases… I don’t understand why they come to Oranga Tamariki, and I don’t understand why mental health comes to Oranga Tamariki when we have mental health specialists.”

P15 Oranga Tamariki

This lack of other basic services such as mental health, addictions, and educational needs was a consistent problem noted by both Oranga Tamarki and community participants, and this hampered their ability to address the causes of potential impingement on parenting capacity.

4.3. Inconsistent Attendance and Its Causes

Inconsistency of attendance at the tēpu is an ongoing challenge. As noted, the relationships and trust that form between participants help with informed decision-making and establishing processes, as well as managing conflict. Time commitment and other pressures, however, have meant that workers cannot always attend consistently, which at times has challenged trust and decision-making. Participants noted that this was partially due to the time commitments involved:

“because of the way that the model is really asking all participants to be present for all of the RoCs from each of the sites. That’s still, it means, the kaimahi from X are being involved in decisions for the other two sites, and vice versa, and all of them. So it is a big time commitment, and because of that, it means that participation is variable”

FG

“also be the time, there was a week where I spent, there was two weeks, and all in all I spent about 16 h, if not more, at the tēpu.”

Oranga Tamariki P8

This has led to at times frustration and challenges relational continuity:

“So that is causing a lot of friction, and it’s causing the Tai Timu Tai Pari to question Oranga Tamariki’s commitment, which is really unfair, because we’re all being lumped in the same basket. But I totally get where they’re coming from, because they’re not seeing the same level of enthusiasm … So yeah, that’s definitely a challenge.”

Oranga Tamariki P8

Inconsistency of attendance challenges the relationships and trust that underpin the tēpu’s decision-making process.

4.4. Authority Conflicts: Legal Mandates and Information Sharing

A number of legal, ethical, and technical challenges arose through the ongoing development of the tēpu, particularly concerning mandates and information sharing. The wider legal context the tēpu was operating in was unclear due to lack of clarity around its status. Initially, Tai Timu Tai Pari had enabling communities status as a proposal, which was accepted by Oranga Tamariki in 2022. However, sometime after this, Tai Timu Tai Pari was informed that they no longer met the enabling communities criteria and that the tēpu was being re-categorised as an example of partnered response. This affected the level of power and authority Tai Timu Tai Pari could exert over the process.1

“So…the first barrier, of course, is a change of government and priorities and the uncertainty that’s developed within Oranga Tamariki with the changes… so now, we’ve learned that the tēpu was partnered response not enabling communities. We saw it as part of our enabling communities moemoea, which is turning off the tap of our tamariki going into state care… if those on the ground and in the regional office, don’t understand what it is, and they don’t understand that they have to let things go to community.”

P2 Community

A related issue is that of information sharing and access to data, which also reflects confusion regarding status. Participants had diverse views on who should have access to reports, which kinds of reports should be shared with the tēpu, and when they should be shared. While an increasing proportion of reports have been brought to the tēpu over time, community participants expected that all local reports made are brought to the tēpu, even very acute ones that they will never take responsibility for directly. Leaders noted there are about 2500 reports in the area each year, but in some weeks, they are only seeing around 10 at the tēpu, indicating many are not being brought to the tēpu:

“We knew what the numbers were because we’d seen the numbers. So we knew that there was either a resistance or lack of communication and understanding internally for that to be able to happen. So we left that for the regional leadership team (within ORANGA TAMARIKI) to deal with, and we continued there, and they set the process together….eventually, the design was that all RoCs had to come to the table (but) …there were always reasons why we couldn’t do stuff…”

P2 Community

Community participants felt the sharing of higher-risk reports is important for several reasons, including to gain an understanding about higher acuity issues, offer their knowledge about those whānau even if they are not leading the work, and being able to start supportive work with that whānau alongside Oranga Tamariki. It was very important to the community participants that they could offer support and advocacy through the FGC and other statutory processes:

“If you’re going to be involved with our whānau, so are we”

Community P12

There were concerns from some Oranga Tamariki workers about sharing people’s information unnecessarily, where that information included highly sensitive material such as children’s disclosures of abuse, information about pending criminal charges, or evidence that may be used to secure convictions in the future. It was felt to be unnecessary because the community would never be taking on these cases. This was particularly in regard to the more acute cases. Their balance of benefits and costs of sharing differed from the community perspective.

Other Oranga Tamariki participants noted that this was not only about information but about mandate and legal roles. Oranga Tamariki have a legally mandated role to protect children from harm, and this can include the use of police, interviewing children alone without parental consent, and other higher-level interventions that are not available to community participants at the tēpu:

“The whānau will end up being way more driven, in direct lines than… those high tariff cases, the tēpu aren’t going to touch. They’re just not, actually, I wouldn’t expect them to. They can’t go in with police. They can’t.”

P15 Oranga Tamariki

The Oranga Tamariki legislation explicitly allows for information sharing with child and family services in order to further child protection and wellbeing more generally (s66 Oranga Tamariki Act 1989). Community participants argued that even in high-risk cases, because they tend to ‘know the whānau better than a government agency’, they still have valuable information to offer and can also provide advocacy for whānau as they go through the Oranga Tamariki process from the start. For example:

“We’re quite strong … from the community point of view, because we want to know what they’ve got on that whānau, although they’re pretty slow in sharing the information. Like I said today, they hold back information. But obviously the community knows more about the whānau than they do, and I’ve tried to stress that point many a time I said, ‘remember, we’re here helping you’.”

P14COM

Part of the advocacy and support community participants are able to provide is long-term engagement and support with whānau, which is not possible in the ‘process and event’ driven system of Oranga Tamariki practice, which may close after investigation.

The timing of access to reports was also an issue. Community participants reported needing access to the reports before the tēpu meets in order to be able to peruse the information before the meeting. Current practice is they have no access to any information until the tēpu meets and information is shared directly by Oranga Tamariki workers at that moment. Differences in the timing of access to information affect the decision-making of individuals and groups, reducing decision-making effectiveness. Still other Oranga Tamariki participants noted that while there have been higher-level conversations about information sharing and access timing, the key issue at the front-line is a technical issue rather than ethical or legal concerns:

“The challenge really is more around getting it in action, around getting the RoC’s to that email address, because there is a lot of red tape for Oranga Tamariki that we need to do that’s higher than our pay grade, I would say, around the confidentiality, privacy, and going through all of that”

FG

In conclusion, there are a number of varied issues relating to legal status, mandates, and information sharing that remain challenging to resolve. These all reflect the changing power relationships that require a re-negotiation of power and authority—whether that be over organisational status or sharing specific types of reports at the tēpu.

4.5. Future Needs of the Tēpu and Threats to Development

Many participants shared insights into perceived future enablers and barriers to the development of the tēpu. These included themes relating to the policy context, building community capacity, and attending to the ‘long game’ of long-term devolution reform.

4.5.1. Policy Context to Continue to Support Devolution

The future of the tēpu relies on the policy emphasis on devolution of child protection to the community, particularly authority to Māori, and resourcing related to the community and Oranga Tamariki staff who take part in the tēpu. Current policy changes are threatening this direction. In particular, the loss of funding of community services, lack of clarity about the enabling communities process of devolution, the repeal of s7AA, and the Treaty Principles Bill, along with the dissolution of key Oranga Tamariki roles, all point to a set of complex policy changes.

“And then losing a lot of things as well at the moment, with everything changing. We’re losing, Ngāti Hine Family Start and stuff like that. So it’s a little bit… unclear at the moment as to what is happening, because… losing a lot of funding and everything, and staff and lots is happening at the moment…So it is hard.”

P4 Oranga Tamariki

In particular, there was an awareness that more, not fewer, resources were needed to be able to continue development of the tēpu:

“I think it holds amazing things. I think it’s going to continue to grow… It needs to be funded, because this work can’t continue without it. That’s the reality. If we don’t get the funding, and this shuts down within a year, I would say it would, we’re going back to NFAs. Which actually in this government, I don’t think they care. We’re just the little people on the ground trying to clear their mess”

P8 Oranga Tamariki

“it would just be, we need more resources. We can’t do it on our own. I just want to shake you right now. I think give us some people. We need people. We’ve got way too much work up here, and we’re doing our very best…We definitely can’t manage. If we have to take it all back, we can’t before we’ll crumble. Kids are going to die. That’s simple as that”

P11 Oranga Tamariki

Ensuring the structural and legal conditions for Māori to reclaim authority over child protection processes, along with retaining and extending workforce resources, is fundamental for the tēpu to continue to operate and to develop further.

4.5.2. Building Community Capacity to Take on More Moderate-Risk Cases as Part of the ‘Long Game’

Along with more funding and workforce, community capacity needs support in order to take on more of the ‘moderate’ cases, that is, those whānau where the presenting issues were moderately serious. Enlarging the scope of cases was linked with the tēpu becoming more independent from Oranga Tamariki over time, as well as supporting community worker’s skills and capacity. The reference to child and family assessments, which are currently undertaken by Oranga Tamariki, in this quote shows this willingness for more whānau to be responded to by the community:

“the aim, if it was up to me, the aim of the tēpu, would be to move forward with the new structure, which is what we’ve got coming in now… I would say that CFA’s (Child and Family Assessments), as long as they’re not high risk, sit with community…the aim is for that is to happen slowly over time…”

Oranga Tamariki P8

While there was broad support for this direction towards further devolution, there was also concern that the community did not simply take on the same practice models as Oranga Tamariki and operationalise them in the community. Instead, as more responsibility is taken on in the community, there was an awareness that new ways, aligned with the community agency kaupapa and Māori practice models, had to be developed. This is a common concern in Indigenous-devolving systems. New practices must be developed alongside the structural devolution process (Cleaver 2025; McDonald et al. 2025):

“…it’s good when people understand that there’s a long game here…. We’ve got a huge opportunity…let’s grip it, and let’s go with it, but let’s not just turn up with a barrel of brown paint.”

P16 Community

More community capacity is needed to be able to progressively move more of the child protection process into the community. Developing Kaupapa Māori practice models and processes, increasing the community workforce, and staff training are all part of this.

5. Discussion

Complexity in child protection systems requires sophisticated strategic actions and a clear understanding of diverse drivers and consequences (Gillen and Canavan 2024). Likewise, in transformative projects such as decolonisation or community devolution, initial reforms in any part of the system must pave the way for more radical change instead of just an easily coopted concession from the existing system (Fitzmaurice-Brown 2023). This requires ongoing, coordinated, and strategic actions, resources, and practices across multiple parts of the system to effect sustainable change. Efforts towards change can be undone or paralysed by the weight of organisational path dependency, political change, institutional logics, poor resourcing, or the replicative nature of colonialism itself (Fitzmaurice-Brown 2023). This study contributes to understanding these complex systemic strategies, building an evidence base relating to systems development towards devolution starting at one possible starting point: the ‘front door’ of child protection.

Findings were that the policy context emphasis on family preservation and Indigenous rights and a small town where many pre-existing relationships existed between leaders were important initial enablers. This context set the scene for consensus between the statutory and community participants around problem definition, particularly that closed cases required more effective early support, and information quality (and therefore decisions) could be improved through combining information. Ongoing enablers were positive relationships at all levels of practice, consistent attendance, the development of key practice processes, commitment to review, information sharing, community location and leadership, and the alignment of the practice with practitioners’ values. Challenges were responsibility tensions in relation to moderate-risk cases, consistency of attendance, and conflicts around information sharing, especially in relation to higher-risk situations and the timing of sharing information. Gaps in other key services needed to address family struggles such as mental health were also evident. Shifting power and affecting genuine change to institutional logics in the child protection system, including decentring power from state to community actors in Indigenous contexts, is complex. Gaining traction through shared problem definition, shared aims, leadership from all participating organisations, and ongoing commitment to review helps both start a shift in power and maintain initial movement. In terms of institutional logics, moving community devolution to Māori beyond simply a symbolic or rhetorical logic to deeper organisational change requires multiple strands of implementation. In this instance, those strands have included organisational arrangements and location, changes in decision-maker group membership, changing decision-making processes, and changing decision outcomes to include the ability to address immediate family need (Lounsbury and Boxenbaum 2013). These are all needed to shift power. Further shifts, however, may require legal delegations to ensure continuing redistribution of power, particularly in order to enable more moderate-risk threshold decisions and rebalance information sharing inequities.

Relationships are fundamental to the tēpu—its genesis, development, and functioning. These relationships mediate the outcomes the tēpu is able to achieve, which is high consensus around most discussions (Ilgen et al. 2005). Pre-existing relationships within the community, including professional, whakapapa, and informal relationships, helped the tēpu establish and develop. The relationships of workers at the tēpu overcame initial mistrust to develop into relationships of trust that help weather the inevitable robust conversations and shifting power. Mutual respect for each other’s areas of strength and expertise, despite initial mistrust, was built through consistent attendance. Particular leaders within both key agencies were influential, with trusted individuals carrying mana building community trust in the process. Oranga Tamariki leaders pulled the impetus towards community power-sharing despite others within the organisation being less willing, a clear example of the ‘diffusion of innovation’ theory of implementation of organisational change, where leaders initially produce change while others may take time to adapt (Bunger et al. 2019).

It is well established that different professionals weight certain information differently when considering decisions about intervening in family life, for example, severity of and patterns of harm; parental reaction to past service offerings; the likelihood of recurrence; children’s disclosures; and family stability may all be weighted differently in decision-making fora by different professionals (Britner and Mossler 2002). In addition to different weighting of key factors when reaching a risk judgement, different professionals also have differing thresholds at which they may make a decision to act (Fluke et al. 2014). These thresholds are influenced by multiple factors, including policy, personal factors, organisational norms, and resourcing (Keddell and Hyslop 2020). While group decision-making expands the expertise, services, and organisational mandates available for families, how people manage considerable differences remains a challenge (Przeperski and Taylor 2020; Alfandari 2019). In this study, professionals were managing key differences between statutory and voluntary mandates, perceptions of professional hierarchy differences, differences relating to organisational cultures, and authority challenges as power shifts away from the statutory agency and workers. Both risk perceptions and threshold conflicts were evident in this study, with evidence of a key gap in threshold between those that were NFAs and clearly were within the remit of community workers to respond to and those that were perceived by statutory workers as not requiring their follow-up (usually in the form of a ‘child and family assessment’). With the logic of the tēpu that all decisions needed consensus decisions that everyone abided by and every family receives some kind of response, either from community or Oranga Tamariki, this gap became a cause for conflict, especially in the context of shrinking resources for both agencies. This reflects the known resource ‘gap’ between the statutory threshold and capacity to respond in the community as well as differences in risk perceptions between the same (Keddell and Hyslop 2018; Keddell et al. 2024). There is an ongoing need for further clarification to establish the ‘shared mental models’ (shared beliefs and agreed values and heuristics) around moderate-risk situations. These are required to develop for any group decision-making process to be effective (Parker et al. 2025). However, this is not simply an issue of decision-making criteria but of material capacity and capabilities to respond.

The sharing of reports and the timing of sharing brings to light a number of important ethical and practical concerns, including proportionate information sharing, privacy concerns and interests, impacts on decision quality and technical mechanisms for information sharing. It is also a key site for the expression of asymmetrical power dynamics and differences in legal mandates between parties to the tēpu. The need for all report types to come to the tēpu enables the community to contribute information about highly complex whānau and allow them to provide continuous support and advocacy to whānau throughout the Oranga Tamariki process. Advocates outside the statutory system can improve the chances of family preservation and engagement. It also reinscribes community investment and accountability, helping shift power and ultimately reduce disparities (Saar-Heiman et al. 2024; Lalayants et al. 2021). Differences in the timing of access to information affect the decision-making of individuals and groups. Exclusion from early access to reports for community workers creates a timeframe effect and a ‘common knowledge’ effect. In terms of timeframe, acquiring the information on the day at the tēpu meeting creates decision-making time pressure for community workers, forcing them to rely more on heuristics and automatic, intuitive thinking, which is prone to biases and short-cuts. Those workers who access the information earlier (Oranga Tamariki workers) can engage in a more considered analytic ‘slow thinking’ style (Alfandari et al. 2022; Kahneman 2011). In a group situation, those provided with information earlier are likely to develop a shared view that can exclude other decision-makers at the table, creating a power imbalance that impacts decision outcomes (Alfandari et al. 2022). This asymmetry can also affect the ‘shared mental models’ that are needed in a group for group decision-making to be effective, that is, the in-common interpretive frameworks that need to develop for any group decision-making process (Parker et al. 2025). These are more difficult to establish when different group members have access to information at different times. Legal delegation of statutory powers to receive and investigate reports under s15 and s17 of the Oranga Tamariki Act may assist with this transfer (possible under enabling communities or the deleted s7AA provisions—see Oranga Tamariki (2024b), supported by a formal information sharing framework policy guidance that makes explicit the information pathway and transfer.

The availability of services beyond what either the community or Oranga Tamariki can offer remains a key concern, highlighting the need for comprehensive, multi-disciplinary support for whānau facing complex challenges. This is especially difficult in the north, where services and support are limited. The Oranga Tamariki legislation creates a legal onus on the state to support children’s wellbeing—not just address care and protection concerns—yet the paucity of support for either the community agencies that participate in the tēpu or the other services that should be supporting parents, whānau, and children show clear gaps in meeting this legislative requirement. Several sections of the Oranga Tamariki Act 1989 are relevant here, including the opening purpose in Section 4, which states that the aim of the act is to “promote wellbeing” through “establishing promoting or co-ordinating services that …promote their best interests and…assist families and whānau … to prevent their children from…suffering harm”.

The tēpu approach to decision-making and service provision is an initial strategy to begin the ‘long game’ of devolving community services to Māori communities. As recommended by the Waitangi Tribunal and Oranga Tamariki’s own ministerial advisory committee, devolution is a way to reduce inequities, improve prevention, and restore authority to Māori (Waitangi Tribunal 2021). This project highlights one approach that starts at the ‘front door’ and identifies the enablers needed to begin this shift in power, including leadership, relationship building, development of decision-making processes, and developing shared understandings around risk, thresholds, and privacy. Addressing the information sharing and resource challenges as well as maintaining the policy environment that is an important context for maintaining momentum is needed to continue this direction of travel. Scaling-up similar initiatives brings further challenges, including balancing between community and iwi priorities and the need for fairness between locations. However, given the now-rising numbers of Māori children entering care, it is clear that strategies such as this warrant further adoption and investment (Oranga Tamariki 2025).