Reform of the federal funding for First Nations child and family services now runs one region at a time — as of 2026-08-28 only Ontario has a final agreement approved by the Canadian Human Rights Tribunal

What is happening?

On 2016-01-26, in the decision cited as 2016 CHRT 2, the Canadian Human Rights Tribunal held that the way Canada funded child and family services for First Nations children ordinarily resident on reserve amounted to discrimination on the basis of race. The design the tribunal named was a split one — prevention and family support were funded at fixed amounts set in advance, while the cost of placing a child outside the home was reimbursed as actually incurred. The tribunal found that this arrangement pulled toward removal and away from the services that would have made removal unnecessary. The program itself is built in five streams: prevention, protection, post-majority support, representative services and capital.

Ten years later the case is still open. Since that ruling, 34 further orders, including non-compliance orders, have issued in the same proceeding, according to the running tally kept by the advocacy organization that brought the complaint.

Two tracks came out of the ruling, and they have moved at different speeds. On the compensation track, the tribunal ordered in 2019 that the statutory maximum of 40,000 Canadian dollars be paid to each child removed from home since 2006 and to parents and grandparents. On 2023-10-24 the Federal Court approved a settlement of 23.34 billion Canadian dollars built on that order, covering more than 300,000 people. As of 2026-08-17, more than 1.2 billion Canadian dollars had been distributed to more than 30,000 people, and claims remain open until 2028-03-10.

On the reform track, the national route closed. An agreement in principle on long-term reform was signed on 2021-12-30, and program spending rose from 680 million Canadian dollars in 2016 to 3.571 billion in fiscal 2023-24 — but the structural design was left to a final agreement still to come. A national final settlement agreement valued at 47.8 billion Canadian dollars was assembled in 2024 and put to a special assembly of First Nations chiefs on 2024-10-21, where 267 of the 414 chiefs present voted against ratifying it. The reasons recorded for that vote were the reliance of the package on annual parliamentary appropriations and uncertainty about how it would be implemented. On 2025-01-06 the federal justice department wrote to the national First Nations representative organization stating that Canada no longer held a mandate to continue national negotiations, and that it would negotiate instead with the two Ontario First Nations political organizations that had supported the agreement.

One region finished. That Ontario negotiation concluded in February 2025 and the tribunal approved the resulting final agreement on 2026-03-30: 8.5 billion Canadian dollars over eight years, more than 900 million a year. Two First Nations stayed outside it. On 2026-05-06 the national representative organization issued a statement calling for the agreement to be implemented.

For everywhere else, what exists is an announcement and a filing. On 2025-12-22 Indigenous Services Canada announced a funding framework of 35.5 billion Canadian dollars to fiscal 2033-34 and 4.4 billion a year after that, to be delivered through regional agreements that the department describes as still under negotiation. On the same day the complainant advocacy organization filed its own national plan, the Loving Justice National Plan, directly with the tribunal as an alternative to a negotiated settlement for the regions outside Ontario. As of 2026-08-28 no region other than Ontario has a final agreement the tribunal has approved.

Whose problem is this?

The body that must deliver the remedy is also the body that must be persuaded to agree to it, and on the other side of the table sits a set of organizations that can decline. Nothing in the arrangement obliges either side to arrive at a date.

Where does this problem end?

The boundary here separates the design of a funding stream from the question of who ought to hold jurisdiction over child welfare in the first place. Those two questions are argued in different forums and by different parties, and a remedy in one of them does not settle the other.

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

Now

Needs a new measurementthe target state: no source opened here names a date by which the regions outside Ontario are meant to have agreements in force, nor a level to which the over-representation of First Nations children in care is meant to fall. The announced framework carries dollar figures and fiscal years for the money. It does not carry a completion date for the agreements that would spend it, and no opened source supplies one from anywhere else.

How big is it?

The counted quantities are money and claimants, not the population standing outside a remedy. The compensation track has a firm denominator: more than 300,000 people were found eligible under the settlement approved on 2023-10-24, and more than 30,000 of them had been paid as of 2026-08-17. The reform track has budget figures — 8.5 billion Canadian dollars for one region over eight years, 35.5 billion announced for the rest to fiscal 2033-34 — and no headcount attached to either.

The rate figures come from the census and describe a wider group. In 2021, 53.7 percent of children in foster care in private households were Indigenous, while Indigenous children were 7.6 percent of all children aged 14 and under. For First Nations children specifically the foster care rate was 41.8 per 1,000, against 2.3 per 1,000 for non-Indigenous children. These are census counts of children in care, not counts of children served by the federal program, and the two populations are not the same.

A headcount of the affected population is therefore recorded as not derivable. No source opened here gives the number of children and families the program serves. The national count of First Nations communities that is published federally is more than 630, and the page carrying it gives no date for that number. The count of First Nations covered by the one approved agreement is unsettled between two figures, 128 and 133, neither of which was confirmed here, so the remainder cannot be obtained by subtraction either.

Under what conditions does it arise?

1. The finding is binding and the remedy is negotiated. A tribunal can hold that a funding design discriminates and can order that it be reformed. It cannot write the appropriation. The instrument that would actually change the design is a funding agreement, and a funding agreement requires a federal mandate to negotiate on one side and ratification on the other.

2. The money is voted one year at a time. A settlement that describes a decade of spending rests on appropriations Parliament passes annually, and that reliance was among the reasons recorded for the vote of 2024-10-21 against ratifying the national package. A remedy that cannot guarantee its own durability is harder to ratify, and for as long as no agreement is in force the original design stays in place.

3. Delivery moved from one negotiation to many. After the national mandate ended on 2025-01-06, the route became region by region. Each region has its own parties and its own pace, and the one that finished took from February 2025 to a tribunal approval on 2026-03-30.

4. Only one of the two tracks has a clock. Compensation claims close on 2028-03-10. The reform track has no equivalent date in any source opened here, and a missing deadline produces no event that anyone reports.

5. Total spending can rise while the design stays. Program spending grew from 680 million Canadian dollars in 2016 to 3.571 billion in fiscal 2023-24. The finding of 2016 was about the relation between two streams rather than about the size of the whole, so a larger budget does not by itself answer it.

What has been tried?

Three routes have been tried in sequence and all three are still live at once. Litigation produced the finding and keeps producing orders. Negotiation produced one regional agreement and one rejected national package. Announcement produced a framework whose agreements do not yet exist.

What was found?

Why is it still unsolved?

Enforcement absent — the finding binds, the orders keep issuing, and the instrument that would actually change the funding is an agreement that neither side is obliged to reach by any particular day.

A tribunal can declare a funding design discriminatory. It cannot pass an appropriation. So the remedy has to pass through a negotiation, and a negotiation has two conditions that can each fail without anyone breaching an order: a mandate on the federal side to be at the table, and ratification on the other. The first ended on 2025-01-06. The second failed on 2024-10-21. Neither of those was a refusal to obey the tribunal, which is exactly why neither produced a consequence that the tribunal could impose.

The second part of the gap is the annual appropriation. A package describing a decade of spending is built on money Parliament votes one year at a time, and the reliance of the 2024 package on those annual votes was among the reasons recorded when the assembly declined to ratify it. That is a stable loop rather than a one-off: a system that votes the money one year at a time cannot write into the document the durability that ratification requires, and for as long as that remains unresolved the design of 2016 keeps running.

The third part is that moving from one national settlement to a set of regional ones converts a single deadline problem into many. Each region negotiates separately, the one completed took roughly a year from conclusion to tribunal approval, and no source opened here carries a schedule or a target for the rest. Compensation, by contrast, has a claims deadline of 2028-03-10 and a running total that gets reported. A track with a date generates news when the date approaches. A track without one generates nothing to report, and the absence of reporting reads from outside exactly like progress.

The last part is the reason a larger budget does not settle the question. Spending on the program rose more than fivefold between 2016 and fiscal 2023-24, and the 2016 finding was not about the size of the total. It was about the relation between a stream funded at fixed amounts in advance and a stream reimbursed at whatever it costs. Until an agreement changes that relation in a region, that region keeps the arrangement the tribunal ruled on, no matter what the annual figure says.

What observation would mean it is solved?

Candidates — (a) every region has a final agreement in force that the tribunal has approved, and the tribunal closes the compliance file (b) the gap between the foster care rate for First Nations children and the rate for non-Indigenous children narrows across successive census cycles (c) the approved compensation fund is fully distributed to eligible claimants before the claims deadline.

(a) alone counts agreements, not funding. An agreement can be approved and still leave the balance between prevention money and care money close to where it was, and the amounts inside it are still voted annually. Coverage is also not the same as participation: two First Nations stayed outside the one agreement that exists, so a region can be reported as settled while parts of it are not.

(b) alone moves for reasons that have nothing to do with this. Census foster care rates respond to provincial child protection practice, to how children are placed with relatives rather than in formal care, and to how identity is reported on the census form. A fall could mean fewer children removed or fewer children counted, and no source opened here separates those.

(c) alone looks backward. The compensation track settles harm already done to people removed since 2006. Paying it in full is owed, and it changes nothing about how the program funds prevention tomorrow. The three have to be read together, and (a) has to be read against what the agreements actually do to the two streams.

What is it connected to?

Fills with researchrecognition of First Nations jurisdiction over child and family services, provincial child protection systems and their interaction with the federal program, the wider federal fiscal relationship with First Nations, the handling of individual service requests under Jordan's Principle, and comparable over-representation of Indigenous children in care in other countries. Relation type and evidence grade were not confirmed in this round.

What these sources do not say

• When the remaining regions are meant to have agreements. No source opened here carries a schedule, a progress figure or a completion target for the regional negotiations. What the opened sources show is fragments at the level of individual meetings, and the federal page that describes the negotiations gives money and fiscal years without a date for the agreements themselves.
• How large the population outside an approved agreement is. No source opened here gives the number of children and families served by the program, and no source opened here states how many First Nations the Ontario agreement covers. Two counts circulate for the number of First Nations in Ontario, 128 and 133, and the page that would settle it returned only truncated text.
• What the published federal position adds beyond the framework. The position that is published is the departmental page describing regional negotiations and the framework announced on 2025-12-22. No source opened here carries a federal explanation on the record of the decision to end national negotiations beyond the statement that the mandate had ended.
• The date the 2024 national package was reached. No source opened here establishes it. The federal news release that would carry it refused automated retrieval, so the year alone is used in this document.
• How the compensation total relates to the order it rests on. No source opened here shows how a per person maximum of 40,000 Canadian dollars ordered in 2019 becomes a settlement total of 23.34 billion Canadian dollars approved in 2023.
• What has become of the plan filed on 2025-12-22. No source opened here reports any ruling, schedule or response from the tribunal on that filing.
• Why two First Nations stayed outside the Ontario agreement. No source opened here gives their reasons, the terms they objected to, or what arrangement applies to them now.
• How the 34 orders break down. No source opened here separates non-compliance orders from procedural and remedial ones, or reports which of them the federal government contested.

See the evidence

No primary tribunal or court document was read in full. The five federal department pages listed above were opened directly and carry the program structure, the Ontario agreement terms, the announced framework, the spending figures and the national count of First Nations communities. The national statistical agency release was also opened directly and carries the census rates. The discrimination finding of 2016, the order count, the compensation settlement, the ratification vote and the end of the national mandate all come from summaries, trade publications, news outlets and organizational bulletins that cite the underlying documents, and each is graded accordingly above. Where the opened sources overlap they agree: the Ontario agreement value and its approval date appear in both a federal page and a law firm note, and the framework of 2025-12-22 appears in the departmental page and in the filing made the same day. Where they disagree the disagreement is left visible rather than resolved — the number of First Nations in Ontario is given as 128 by one count and 133 by another, and the page that would settle it did not return its full text. This is a Path A output, research-based definition, so observation_refs is empty and provenance_mode: press-derived.

People affected

Estimated range Not derivable

The reason and what is missing are listed under “What is missing” below