New Zealand repealed the statutory duty to reduce Māori over-representation in state care about 11 months after a Waitangi Tribunal report that carried no obligation with it — 69 percent of children in care were Māori against a 27 percent population share on 2025-06-30

What is happening?

In 2019 New Zealand wrote a duty into the Oranga Tamariki Act. Section 7AA required Oranga Tamariki, the Ministry for Children, to reduce disparity for tamariki Māori through policies and practices with measurable outcome targets, to have regard to the mana, whakapapa and whanaungatanga of every child in its care, to build strategic partnerships with iwi and Māori organisations, and to report publicly every year on what it had done.

That section is gone. The Oranga Tamariki Repeal of Section 7AA Amendment Act passed its third reading on 2025-04-03 and received royal assent on 2025-04-07. What replaced it is a softer clause about developing strategic partnerships with iwi and Māori organisations. The duty to reduce disparity against measurable targets, and the duty to report annually on it, are not in the replacement.

The Waitangi Tribunal had already found that the repeal would breach the Treaty of Waitangi. It ran an urgent inquiry, issued an interim report on 2024-04-29 that recorded findings without recommendations and described the process as rushed and arbitrary while warning of a risk of substantive harm, and then delivered its report on 2024-05-10 holding that the repeal would breach the Article 2 guarantee of tino rangatiratanga and the Treaty principles of partnership and active protection. Royal assent followed about 11 months later.

Nothing in that sequence required anyone to stop. Tribunal recommendations do not bind the Crown. The single binding jurisdiction, created by a 1988 amendment, reaches certain Crown-owned land including land transferred to state-owned enterprises, and this matter does not fall inside it. The Tribunal can find a breach of the Treaty and it cannot require anyone to do anything about that finding.

The numbers the repealed section was written against have not moved. At 2025-06-30, of 4,217 children and young people in Oranga Tamariki care and custody, 69 percent were Māori, while Māori were 27 percent of the population under 19. Aroturuki Tamariki, the independent monitor of that system, records the over-representation as unchanged across three years.

Around 2025-04-15 the two numeric operational targets went as well. The government announced that the 58 percent target for placing children in care with wider whānau and the minimum 23 percent investment target for iwi and Māori organisations had been dropped as outdated departmental targets, under an approach described as looking at need rather than at race.

Whose problem is this?

The body that found the breach of the Treaty is not the body that decides what follows from it, and the body that decides was under no obligation to answer the finding before it voted.

Where does this problem end?

The boundary matters because the repeal is recent enough that nobody can yet show what it changed on the ground, while the disparity it was written against is measured every year and has not moved.

What is the state now, and what should it be?

Now

The state the repealed section aimed at was written into the section itself. It required policies and practices that reduce disparity for tamariki Māori, measured against outcome targets the department had to set and publish, and it required an annual account of progress against them. Two of those targets were numeric and public until April 2025, at 58 percent for placement with wider whānau and 23 percent for investment through iwi and Māori organisations. As of 2026-08-15 no statutory target, no numeric operational target and no annual reporting duty stands in their place, and what remains is a partnership clause naming no quantity and no date.

How big is it?

Between 2,889 and 2,931 Māori children and young people were in Oranga Tamariki care and custody on 2025-06-30. The chain has two terms and both come from the same report. Aroturuki Tamariki counted 4,217 children and young people in care and custody on that date and published the Māori share of them as 69 percent. Because that share is published to the whole number, the true share sits between 68.5 and 69.5 percent, and the two ends of that band are the two ends of the interval. The width is rounding and nothing else.

That number counts a single day. Across the year from July 2024 to June 2025, 5,640 children and young people spent some time in care, about a third more than stood in it on 30 June. The Māori share of that annual flow is not published in any source opened here, so the number of Māori children touched by the system across a full year is larger than the interval above by an amount nobody here has measured. The only evidence on entries says that over the past decade a majority of children newly entering care were Māori, and a majority is not the same quantity as 69 percent, so the flow share cannot simply be assumed to equal the share on one date.

The interval also leaves out the group the repealed duty was written to reach before a removal happens. Children at risk of entering care are not in any count opened here, and neither are the whānau on the other side of each placement.

Under what conditions does it arise?

1. A finding with no remedy attached. The Waitangi Tribunal is a standing commission of inquiry. It can take a claim urgently, hear evidence, and publish a holding that the Crown has breached the Treaty. It cannot compel. The binding jurisdiction added in 1988 covers certain Crown-owned land and reaches nothing in this matter.

2. The duty was ordinary law. Section 7AA sat in an ordinary act. Nothing entrenched it, and nothing in the legislative process required a Treaty breach finding to be answered before a vote, which is why the gap between the finding and royal assent is a fact about the parliamentary calendar rather than about any legal constraint.

3. The measurement outlived the obligation. The disparity is still counted and published every year by an independent monitor. The number section 7AA was written to move is therefore still in public view while the duty to move it is not.

4. The numeric targets were policy, not statute. The 58 percent placement target and the 23 percent investment target sat inside departmental policy, so withdrawing them took an announcement rather than a vote, and no vote is recorded against them anywhere.

A body that can name a breach but cannot order a remedy leaves the decision exactly where it stood before the inquiry began, and that gap between naming and ordering is the structure this document is about.

What has been tried?

Two directions were tried and both ran out at the same place. A duty was written into statute and then removed by the same kind of majority that had written it, and a Treaty breach finding was delivered on time and carried no obligation with it.

What was found?

Why is it still unsolved?

Enforcement absent — the breach was named by the body set up to name it, and no instrument turns that naming into an obligation.

A tribunal that holds an urgent hearing, takes evidence and publishes a holding that the Crown has breached the Treaty looks from the outside like a court. For this purpose it is not one. Its recommendations bind the Crown only over certain Crown-owned land under a jurisdiction added in 1988, and this matter sits outside that jurisdiction. So the finding arrived on time, said what it had to say, and left the decision precisely where it had been standing.

The second part is that the duty itself was ordinary law. Section 7AA carried the weight of a measured racial disparity but none of the protection of an entrenched provision, so the same simple majority procedure that created it in 2019 removed it in 2025. A duty that can be repealed by simple majority lasts exactly as long as the majority wants it to, and no part of the Treaty process changes that arithmetic.

The third part is the quietest, and it is the one that keeps the pattern stable. The counting did not stop. An independent monitor still publishes the share every year, so the gap stays visible, documented and unchanged, while the obligation that pointed at it has been withdrawn. Visibility without obligation is a stable arrangement rather than an unstable one. Nothing breaks, no deadline passes, and each year the same figure arrives to no particular consequence.

What observation would mean it is solved?

Candidates — a. the Māori share of children and young people in Oranga Tamariki care falls toward the 27 percent share of the child population and stays there across several annual reports. b. a binding duty to reduce that disparity, with published targets and an annual account, is restored in statute. c. Waitangi Tribunal findings acquire an effect the Crown must answer by law rather than by choice.

a alone is weaker than it looks. The share can fall because fewer Māori children enter care, or because more non-Māori children do, and those two have opposite meanings for the same number. It can also fall because the threshold for entering care moved, which changes the count without changing what happens to any child. One annual report cannot separate these, which is why the direction has to hold across several of them.

b alone counts paper. Section 7AA was that instrument, it stood for six years, and the over-representation is recorded as unchanged across the most recent three. Restoring it would restore the duty and the annual account. It would not by itself restore anything that made the duty bite, because nothing opened here shows that the original section carried a consequence for missing the targets it required. It is listed as the direct reversal of the change described above, not as a recommendation.

c would change the structure, and it is the least likely of the three. It would convert a commission of inquiry into something carrying compulsion, which is a constitutional question rather than a child welfare one, and no source opened here reports it being contemplated.

The honest reading is that a has to be measured across several years, b has to arrive with a consequence attached, and c is a different order of question altogether.

What is it connected to?

Fills with researchthe official response to the repeal from Māori political parties and from iwi, including whether any judicial review or further Treaty claim was filed after royal assent, was not investigated in this round because of time, and it is the axis most likely to change how this document reads. The relation to Indigenous child removal in Australia, Canada and the United States was also not established; a sibling document covers the Australian case, but relation type and evidence grade were not confirmed here. One opened source places this repeal among a wider set of law changes affecting Māori across 2025, and that set was not pursued.

What these sources do not say

• What the repeal changed on the ground. Royal assent was 2025-04-07 and the monitoring cycle runs from July to June, so the first full year of data after the repeal has not been published. No source opened here can support a claim that the disparity widened because the duty was removed. This is a gap in time rather than a gap in research, and it will close on its own schedule.
• The exact wording of what the Tribunal recommended. The report itself could not be opened. The official Tribunal site and the law society submission both defeated automated retrieval, so the holding described here comes from a Tribunal news page and from reporting, and no recommendation is reproduced in this document in any form.
• How the 58 and 23 percent targets were performing when they were withdrawn. The one opened source that records the withdrawal confirms that the targets existed and that they went. It does not give the actual whānau placement rate or the actual iwi investment share at the time, so it cannot be said whether the targets were being met, missed narrowly or missed badly.
• What the replacement clause actually requires. Nothing opened here describes what developing strategic partnerships obliges the department to do, how compliance with it would be judged, or by whom.
• The size of the next generation behind the current one. One report gives shares rather than counts — about 70 percent of Māori parents with care experience have children who have had contact with Oranga Tamariki, and about one in eight have children who entered care — and no source opened here gives the denominator, so the absolute size of that cohort cannot be stated.
• Whether the Crown answered the finding. No source opened here records what the Crown said in response to the 2024 Tribunal report, or whether any formal answer was given before the third reading.
• Why the disparity did not move while section 7AA was in force. The section stood from 2019 to 2025 and the over-representation is recorded as unchanged. No source opened here says what targets were set under the duty, what the annual reports concluded, or what the department did differently because of it.

See the evidence

No primary legal document was read in full. The text of the repealing act and the Tribunal report were both unreachable through automated retrieval, so every statement here about what section 7AA required, what replaced it and what the Tribunal held rests on secondary description, which is why evidence_tier is secondary and why no recommendation wording appears anywhere above. Where the secondary descriptions overlap they agree: the royal assent date of 2025-04-07 appears in two opened sources, the non-binding character of Tribunal recommendations appears in two, and the duties of section 7AA are described consistently by the departmental page and by the account of the repeal. The care figures are the strongest evidence in this document because they come from the independent monitor directly. The outcome figures for young people with care experience come from reporting on that same monitor rather than from the monitoring report itself, which is why they carry a lower grade, and they are given as aggregate rates without further detail. Five rows above carry no URL because the sources behind them refused automated retrieval or returned a document that could not be parsed; those rows are kept rather than deleted, because a table that looks complete by being shorter is the worse outcome. This is a research-based definition, so observation_refs is empty and provenance_mode is press-derived.

People affected

Estimated range 2,889–2,931 As of 2025-06-30

Sensitivity The interval is not a confidence interval. Its width is the rounding of one published percentage and nothing else, so it says how precisely the share was printed rather than how uncertain the count is. Three corrections run outward from it and none of them can be computed from the sources opened here. The first is the difference between a day and a year. This interval counts children in care on 30 June 2025, and 5,640 children and young people spent time in care at some point across the year from July 2024 to June 2025, about a third more than the stock. The Māori share of that annual flow is not published anywhere opened here, and it cannot be assumed to equal the share on one date, because the only evidence on entries states that a majority of children newly entering care over the past decade were Māori, and a majority is a different quantity from 69 percent. The direction of that difference is unknown until the flow share is published, so the annual number is larger than this interval by an amount nobody here has measured. The second correction is that children at risk of entering care are excluded by construction, although the repealed duty was written to reach exactly that group through policies and practices meant to reduce disparity before a removal happens, and no source opened here counts them. The third is that whānau are excluded, and the report that touches the next generation gives only shares, about 70 percent of Māori parents with care experience having children who have had contact with the agency and about one in eight having children who entered care, with no denominator anywhere opened, so that cohort cannot be sized at all. Against all of that, this count measures exposure to the care system rather than harm, and this document does not attempt to quantify harm.

Regional breakdown No source opened here gives a regional split of Māori children and young people in Oranga Tamariki care. The independent monitor publishes the national count, the national Māori share and the national comparison against the population under 19, and the older disparity report publishes rates per 1,000 at national level as well. Splitting the national count by regional population would be proportional allocation, and it would be wrong here rather than merely approximate, because entry into care and the Māori share of the child population vary across regions and iwi rohe on different patterns, and no opened source states either pattern.