The Canada Disability Benefit pays at most $2,400 a year where the federal budget office estimated $14,356 would close the widest provincial gap, and no identified instrument requires provinces to exempt it

What is happening?

Canada created a federal cash payment for working-age adults with disabilities and then set it at a level far below what its own costing office had measured, and it created the payment inside a division of powers where another government can take the money back.

The Canada Disability Benefit Act, introduced as Bill C-22, received royal assent on 2023-06-22 and came into force on 2024-06-22. Section 3 of the Act states the purpose as reducing poverty and supporting the financial security of working-age persons with disabilities. Section 12 carries a review provision. The Act names a purpose. It does not name an amount.

In 2023-11-16 the Office of the Parliamentary Budget Officer published a costing of the benefit. Measured against the Market Basket Measure poverty line, closing the largest gap among the provinces and territories would take $14,356 a year in 2024 terms, rising to $22,701 once the extra costs of living with a disability are counted. Three modelled scenarios were priced between $2.1 billion and $20.1 billion a year.

Budget 2024 allocated $6.1 billion over six years starting in 2024-25 and $1.4 billion a year after that, and Budget 2025 added $115.7 million over four years. The maximum benefit was set at $2,400 a year, or $200 a month. Applications opened 2025-06-20 and the first payments went out in 2025-07. As of 2025-11-09 more than 330,000 assessments had been processed and more than 218,000 people were receiving a payment, against the federal estimate of 600,000 eligible people. Payments over 2025-07 through 2025-10 totalled $154 million.

Social assistance in Canada is administered by the provinces and territories. The federal government asked each of them not to count the new payment as income when calculating provincial assistance. That request carries no legal force. As of 2025, 11 provinces and territories said they would not offset it, Alberta said it would offset it in full, and the Northwest Territories had not stated a position.

Alberta announced the full offset in 2025-03 and applied it from 2025-07. Its Assured Income for the Severely Handicapped program covered roughly 77,000 to 80,000 people in 2025 at $1,901 to $1,940 a month for a single recipient. The responsible provincial ministry stated in a published response of June 2025 that the Alberta payment level is the highest in Canada and already exceeds a federal benchmark it put at $1,811 a month. For those recipients the federal payment changes the source of the money and not the amount of it.

Whose problem is this?

The government that pays the money and the government that can cancel its effect are different governments, and the one that pays has no instrument to bind the other. That is the shape of this problem rather than a detail of it.

Where does this problem end?

The boundary here is not whether the payment is large enough on its own but whether one government can protect what it pays from being counted as income by another. Those two questions have different answers and different remedies, and this document holds both without merging them.

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

Now

Needs a new measurementthe target state: no source opened here names an official target for what the benefit amount should reach, a target for how many people it should lift above the poverty line beyond a published projection, or a date by which the question of provincial offsetting is to be settled. Section 3 of the Act states a purpose without a number, and the review provision at section 12 is a procedural step rather than a threshold. The costing office published what closing the gap would cost, which is a price and not a commitment. The one forward-looking figure attributed to the federal government is a projection that about 20,000 people would move above the poverty line in the first year and about 25,000 by the tenth, and a projection is a forecast of an outcome rather than a standard that anyone has undertaken to meet.

How big is it?

Between about 600,000 and about 911,000 people, depending on which end of the frame is counted.

The narrower count is the population the federal government itself said the payment would reach. The wider count is the population the stated purpose of the Act names, being working-age adults with disabilities living below the Market Basket Measure poverty line in 2022. The distance between the two is roughly 311,000 people, and most of that distance is the Disability Tax Credit approval that gates access to the payment.

Neither figure counts the people actually paid. More than 218,000 had received a payment as of 2025-11-09, which is about a third of the lower bound and about 36 percent of the federal eligibility estimate. In 2022 about 8 million Canadians aged 15 and over, or 27.0 percent of that population, reported at least one disability, which is the outer population from which every narrower count here is drawn.

Within the counted population sits a subset for whom the net effect is zero rather than small. Roughly 77,000 to 80,000 people were receiving Alberta assistance in 2025, and for them the federal payment is subtracted from the provincial one. About 46,000 of them had moved to the successor provincial program by 2026-07, with a transitional payment scheduled to end on 2027-12-31.

Under what conditions does it arise?

1. Social assistance is a provincial responsibility. A federal payment lands on top of programs whose income tests are written and changed elsewhere. The federal Act names a purpose and assigns no instrument that would keep the payment out of those tests.

2. The federal ask has no enforcement attached. Each jurisdiction decided on its own whether to exempt the payment, on its own timetable, and one decided the other way. Eleven agreed, one declined, one has not said. The count can move again in either direction without any federal step being taken.

3. The amount was set below the lowest priced scenario. The costing office published three scenarios and the amount adopted sits under all of them, so even where no offset applies the gap the Act named remains mostly open.

4. Access runs through a tax credit approval. Eligibility for the payment requires Disability Tax Credit approval, and one opened commentary reports that about 40 percent of working-age adults with a severe disability hold that approval, with about 25 percent completing every step from application through claiming the credit. The year of the underlying study is not stated in that commentary.

5. Provincial program design can move the total independently. Alberta replaced its assistance program with a successor from 2026-07-02, carrying a transitional payment that ends on 2027-12-31, a lower base rate for new applicants, a reduced earnings exemption, and a couple rate set at 88 percent of the individual maximum from 2026-08. Each of these changes the household total without any change to the federal payment.

What has been tried?

Two things were tried in sequence and neither reached the other. The federal side priced the gap, funded a payment, and asked for an exemption. The provincial side, in one jurisdiction, exercised a power the federal side had no way to condition.

What was found?

Why is it still unsolved?

Enforcement absent — the federal law states what it wants provinces to do and contains nothing that makes them do it.

A request that carries no instrument is not weaker enforcement but the absence of enforcement, and the difference shows up only where a jurisdiction decides the other way. For eleven jurisdictions the request and a requirement look identical, because the outcome is the same. In one they diverge completely, and that divergence is the whole measure of what the federal side actually holds. The eleven agreements are not evidence that the mechanism works. They are evidence that eleven governments happened to agree.

The second part is that the province is exercising a power it holds. Social assistance design falls to the provinces, and no source opened here identifies a federal provision that removes or conditions that power for this payment. The provincial ministry has stated its own position on the record, that its payment level is the highest in Canada and already above a federal benchmark it named. Whatever else is true of the outcome, the federal government did not lose an argument here. It never had the standing to make one.

The third part is that the shortfall and the offset are separate failures that read as one. Even in a jurisdiction that exempts the payment entirely, the recipient gains $200 a month against a gap the federal costing office measured at more than $1,100 a month. Fixing the offset would not close that. Raising the amount would not stop a jurisdiction from counting it. A remedy aimed at either one alone leaves the other exactly where it is, and the public argument tends to be about whichever one is in the news.

The fourth part is that the arithmetic keeps moving on the provincial side, on a schedule nobody at the federal level set. A transitional payment ends on 2027-12-31, a base rate drops, an earnings exemption falls, and a couple rate is cut to 88 percent of the individual maximum. Each of these changes what a household receives without changing a single federal figure, so the federal payment can stay constant while the outcome it was meant to produce moves.

What observation would mean it is solved?

Candidates — (a) every province and territory exempts the payment from social assistance income tests and that exemption is fixed in a form that does not depend on an annual policy choice (b) the benefit amount is raised toward the level the federal costing office measured, with the amount stated as of a year (c) the count of working-age adults with disabilities below the poverty line falls over several consecutive years.

(a) alone counts agreement, not protection. Eleven jurisdictions already agree, and that agreement is the current state rather than a solution to it. An exemption held as an administrative decision can be reversed by the same kind of decision that made it, and the sources opened here differ on whether the existing eleven are fixed in legislation or held as guidance. A count of twelve or thirteen would still be a count of choices.

(b) alone counts the federal side only. Raising the amount without changing how it is treated in an offsetting jurisdiction raises the size of the transfer to that provincial treasury rather than to the recipient. It also does not touch the tax credit approval that gates access, so a larger payment reaching the same share of the eligible population changes the depth of the remedy and not its reach.

(c) alone is the slowest and the most confounded. Poverty counts move with the labour market, with other transfers, with provincial program redesign and with the measure itself being rebased. A fall could reflect this benefit or none of it, and a fall recorded while a province is cutting a base rate would be measuring two policies against each other. The three have to be read together, and (a) has to be read as a question about durability rather than about the current tally.

What is it connected to?

Fills with researchfederal-provincial transfer conditionality in other social programs, the design of disability income support in other federal states, the tax credit approval process as an access gate in its own right, and the treatment of federal payments in provincial income tests more generally. Relation type and evidence grade were not confirmed in this round.

What these sources do not say

• Why the amount was set where it was. No source opened here carries an on-the-record federal explanation of the decision to adopt $2,400 a year after the costing office published $14,356. Sources record both figures and the gap between them; none records a stated reason.
• What the federal government says about an offsetting jurisdiction. No source opened here carries a federal statement of what follows when a province counts the payment as income in full, nor any federal assessment of how much of the allocated funding flows through to provincial treasuries as a result.
• Whether the eleven exemptions are legally fixed. Sources opened here describe the eleven decisions differently, some as legislated and some as administrative, and none sets out which form each jurisdiction used. That distinction determines whether the current tally is durable.
• The current position of the remaining jurisdiction. The tracker opened here shows the Northwest Territories as undecided and carries no visible date of last update, so no opened source establishes the position as of 2026-08.
• How many eligible people are stopped at the tax credit gate. No source opened here connects the tax credit approval share to the application and payment counts for this benefit. As a result no opened source establishes whether the 600,000 estimate counts only people already approved or also people who would be approved if they applied.
• The distribution of payments. The total paid and the number of recipients are published for different reference dates, and no source opened here gives an official distribution of how far individual payments were reduced by the income test.
• What the federal benchmark of $1,811 a month refers to. The provincial position rests on a federal figure of $1,811 a month, and no source opened here identifies what federal measure that figure comes from. It matches neither of the two annual amounts the costing office published, and no opened source restates it other than as the position of the province.
• Whether any legal challenge exists. No source opened here reports a court proceeding or a human rights complaint about the offset. This is recorded as an absence in the sources rather than as a finding about whether one could be brought.
• What the successor program changes in outcome terms. Sources opened here give the rates, the transition dates and the caseload moved, and none gives an estimate of the effect on household incomes after 2027-12-31.
• The 2026 figures. No source opened here gives a recipient count or a payment total for any month after 2025-10, so the most recent operating figures available are from late 2025.

See the evidence

No primary federal document beyond the statute text was read in full. The Act text, the costing publication and the two federal briefing notes were opened directly and carry the dates, the section references, the priced scenarios and the operating counts. The official federal benefit page, the tax credit statistics tables and the provincial fact sheet could not be retrieved, so the figures they would confirm are taken from sources that cite them, and those rows are left with an empty URL and a stated reason rather than filled with a substitute address. The original broadcaster report of the Alberta announcement returned HTTP 403 to automated retrieval. The same announcement, its dates, the caseload and the stated provincial position are carried by two other opened sources and by the published ministry response itself. Where sources overlap they agree: the maximum of $2,400 a year appears in two opened sources, the Alberta offset and its 2025-07 start appear in three, and the caseload range appears in two. Where they differ the difference is left visible rather than resolved — the count of jurisdictions declining to offset rests on a single tracker that carries no update date, the descriptions of whether those exemptions are legislated or administrative are inconsistent across sources, and the two figures on the tax credit gate come from one commentary that does not state the year of the study behind them. This is a research-based definition, so observation_refs is empty and provenance_mode: press-derived.

People affected

Estimated range 600,000–911,000 As of 2022-2024

Sensitivity The interval is bounded below by the count the paying government published and above by the count the stated purpose of the Act names. The width of about 311,000 is the distance between who the payment can reach and who it was written for, and most of that distance is the Disability Tax Credit approval that gates access. Neither end counts the people actually paid: more than 218,000 had received a payment as of 2025-11-09, about a third of the lower bound. Neither end isolates the subset for whom the net effect is zero rather than small, because a province counts the payment as income in full against its own assistance; that caseload was about 77,000 to 80,000 in 2025, of whom about 46,000 had moved to a successor provincial program by 2026-07. In the other direction the interval may be too wide: no opened source states whether the 600,000 figure counts only people already approved for the tax credit or also people who would be approved if they applied, no opened source restates either bound for a year later than 2024, and the poverty count is from 2022 while the benefit began paying in 2025-07. The count also says nothing about depth. Everyone inside it faces a maximum of 2400 dollars a year against the 14356 dollars a year the federal costing office measured as needed to close the largest provincial gap in 2024 terms, so the number of people affected and the size of the shortfall are separate quantities that this chain does not combine.

Regional breakdown No opened source gives a province-by-province count of people receiving the payment, or of people whose provincial assistance is reduced because of it. The only jurisdiction-level count available is a provincial assistance caseload, which counts recipients of a provincial program rather than recipients of the federal benefit living in that province, and the two are not the same set. Splitting the national total by population share would invent the distribution rather than measure it, and the distribution is exactly what is at issue here, because the treatment of the payment differs by jurisdiction.