Canada set a goal to halve chronic homelessness by 2027-28, yet its housing law obliges the Minister only to respond to its oversight office within 120 days — in 2024 the budget officer estimated chronic homelessness in the 2020-22 counts at 38 percent above 2018

What is happening?

Canada has a federal housing law with a built-in watchdog, and the watchdog has no power to make anything happen. The National Housing Strategy Act, in force since 2019, declares in section 4 that adequate housing is a fundamental human right affirmed in international law, and in section 5 requires the Minister to maintain a national housing strategy with national goals, timelines and desired outcomes. The strategy and its homelessness program, Reaching Home, carry a stated goal: reduce chronic homelessness by 50 percent by 2027-28.

The same Act creates a Federal Housing Advocate. Section 13 lists what the Advocate may do — monitor the housing policy and progress against the strategy goals, research, consult, receive submissions, advise the Minister, and submit reports with recommendations. It lists no power to order or direct anything. Sections 17.1 and 17.2(1) require the Minister to respond to Advocate reports and to review panel reports within 120 days of receiving them. No section of the Act requires the Minister to accept or implement what those reports recommend.

Two independent federal offices have since looked at the goal. The Auditor General, in a November 2022 audit of chronic homelessness, found according to the committee and advocacy sources that quote it that federal accountability for the goal was minimal and that without better alignment and clear accountability at the federal level the goal was unlikely to be met by 2028. The Parliamentary Budget Officer, in a report dated 2024-05-22, estimated that the number of chronically homeless people in the point-in-time counts taken between March 2020 and December 2022 was 38 percent higher than in 2018, and concluded that Reaching Home was not on track to meet its homelessness targets.

Whose problem is this?

The office that watches the goal and the office that could act on the watching are separate by design. The first can only report, and the second is obliged only to answer, so a report that finds the goal slipping ends in a written reply rather than in any required change.

Where does this problem end?

The document is about a design choice in federal law and what independent federal offices found about the goal that design was meant to support. It does not argue for any particular level of spending or any particular remedy.

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

Now

What it should be

The target state is set by the government itself: chronic homelessness down by half by the 2027-28 fiscal year. The goal text found in these sources does not state the year it is measured from, and none of these sources confirms whether the goal is still the official one as of 2026-09-30.

How big is it?

The only national headcount in these sources is 34,270 people found in shelters or unsheltered locations in the point-in-time counts taken between March 2020 and December 2022, as reported by the Parliamentary Budget Officer on 2024-05-22. That number covers all homelessness counted on a given night, not chronic homelessness alone, and point-in-time counts miss people who are not visible on the count night.

For chronic homelessness these sources give a direction and a rate, not a count. The budget officer estimated a rise of 38 percent against 2018 for the same period of counts. A later national count, Everyone Counts 2024, exists, and the Advocate reports from it that the unsheltered population rose 107 percent between the 2020-2022 counts and 2024. That is a different indicator from chronic homelessness and the two rates are not combined here.

Because no chronic homelessness headcount appears in these sources, the affected population is not derived. The figures above describe the direction of travel against the goal, not the number of people inside it.

Under what conditions does it arise?

1. A numeric goal without an owner of the result. The Act requires a strategy with goals and timelines, and the strategy names a number and a year. The 2022 audit, as quoted by the Public Accounts committee, tied the risk of missing that number to the absence of aligned efforts and clear accountability at the federal level between Infrastructure Canada and the federal Crown housing corporation.

2. An oversight office whose output is advice. Section 13 gives the Advocate the tools of a watchdog and none of the tools of a regulator. The Advocate can track progress against the goal year after year and put the result on the record, and the law treats that record as input to the Minister rather than as a trigger for any required step.

3. A response duty that measures the reply, not the outcome. The 120-day clock in sections 17.1 and 17.2(1) runs on whether the Minister answers. A reply that declines a recommendation satisfies it as fully as a reply that adopts one.

4. A federal instrument in a shared field. Much of housing policy sits with provinces, territories and municipalities, and the procedures of the Act reach only federal matters. The federal goal therefore depends on actors the federal accountability chain does not bind.

What has been tried?

Every step in this table produced a report or a reply. The Act worked as written at each stage, and the goal it was meant to support moved the other way in the budget officer estimate.

What was found?

Why is it still unsolved?

Institutional gap — the law builds a full oversight loop up to the point of decision and stops there.

Everything the Act asks for can be done on time while the goal it serves moves further away. The Advocate monitors, the Advocate reports, the Minister responds within 120 days, and each of those steps is complete in itself. The chain has no link that turns a finding of slippage into a required change in funding, program design or accountability. That is the whole of the gap: a response is owed, an outcome is not.

The 2022 audit, as quoted by the committee report, located the missing link inside the federal government itself, between the department that runs the homelessness program and the Crown corporation that runs housing programs. Its recommendation 4 asked the two to say jointly who is accountable for what. These sources could not establish how the government answered that recommendation.

The one published government position in these sources shows how the design behaves in practice. Asked by the Advocate for a single national encampments plan by a fixed date, the Minister replied with a two-year funding program for local plans and gave reasons. That reply is a lawful and complete answer under the Act. Nothing in the Act measures whether it moves the goal.

No source here says the response-only design caused the rise in chronic homelessness. The budget officer describes the rise as occurring in spite of Reaching Home, and the causes of homelessness lie largely outside this document. What the sources support is narrower: the numbers moved away from the goal, the offices that found this can only report it, and the office that receives the reports is obliged only to answer.

What observation would mean it is solved?

Candidates — (a) a national count shows chronic homelessness at or below half of the level the goal is measured from by the 2027-28 fiscal year (b) the Act or the strategy assigns a named federal body accountability for reaching the goal, with a required step when progress falls short (c) Advocate and review panel recommendations carry a duty beyond a response, such as a duty to implement or to report publicly on implementation.

(a) alone is incomplete. The goal text found here names no baseline year, so the same headcount could count as success or failure depending on the year chosen. Point-in-time counts also change method and timing between rounds, as the gap between the 2020-2022 counts and the 2024 count shows.

(b) alone is paper. An accountable body with no resources or levers would move the gap without closing it. The 2025 site-visit finding that a two-year program constrained progress shows how an assigned responsibility can still run out.

(c) alone can freeze the system. A duty to implement every recommendation of an advisory office would shift decisions away from Parliament, and no source here proposes it. The three have to be read together, with (a) as the observation that the goal itself was met.

What is it connected to?

This case sits beside other documents in this collection that describe recommendations without a duty to act, and differs from each in a specific way noted by the research for this document.

• Canada, calls for justice on missing and murdered Indigenous women and girls — the national action plan and tracking tools in that document are run by the executive and carry no response deadline. The housing case has a 120-day response deadline written into statute and still shows the goal moving the wrong way.
• New Zealand, emergency housing target — there the government reported meeting a reduction target early and the question is what the measure counts. Here no one claims the goal has been met, and an independent federal office estimates movement away from it.
• Australia, family violence recommendations — that document describes a standing commission that tracks and reports annually and records no statutory response deadline; its spread of responsibility runs between the Commonwealth and the states. The housing audit located the gap inside the federal government, between two federal bodies.
• Korea, anti-discrimination remedies — that system has binding correction orders that are rarely enforced. The Canadian Advocate was never given a binding order power, so the gap here is one of design rather than of enforcement.

What these sources do not say

• How the government answered audit recommendation 4. A government response to Public Accounts committee report 31 exists, but its text could not be retrieved in this round, so whether the government agreed, partly agreed or disagreed with the recommendation on alignment and accountability, and what it did afterward, is not established here.
• What the government says about the budget officer estimate. No government statement responding to the 2024-05-22 estimate appears in these sources.
• Whether the goal is still current. The Advocate annual report for 2025-2026 mentions a renewed housing strategy and a new body called Build Canada Homes. Whether that renewal keeps, changes or drops the 50 percent goal for 2027-28 is not established here.
• A newer chronic homelessness figure. Everyone Counts 2024 was conducted in October and November 2024, and these sources do not give its chronic homelessness figure. The 38 percent estimate therefore rests on counts from 2020 to 2022, and the 107 percent figure the Advocate cites from the 2024 count is about the unsheltered population, which is a different measure.
• The baseline year of the goal. The goal text found here names a target year but not the year from which the 50 percent is measured.
• Whether the review panel response met the 120-day duty. The panel report was submitted on 2025-11-24 and the response tabled on 2026-04-17. These sources do not establish whether section 17.2(1) runs to the response itself or to its tabling, or whether a separate tabling rule applies, so no finding of compliance or breach is made.
• Whether the Minister reply rejected the 2024-08-31 date. The reply of 2024-05-30 does not mention that date; the date itself is known from a national news report quoting the Advocate report.
• Whether individuals can go to court. The Act contains no clause that expressly excludes judicial review. One advocacy organization writes that individuals have no concrete route to challenge outcomes in court. Whether that follows from the silence of the statute or from case law is not established in these sources.
• Any causal link between the response-only design and the rise. No source attributes the increase in chronic homelessness to the design of the Act. The budget officer describes the rise as occurring in spite of the program.
• Newer independent reviews. No audit or budget officer review of the goal after 2024 appears in these sources.

See the evidence

The Act and the budget officer report were read directly and carry the title facts. The Reaching Home overview was also opened directly and supplies the goal wording. The 2022 audit is the weakest link among the independent findings: its PDF yielded only its cover and metadata, so its conclusions appear here through a House of Commons committee report and an advocacy organization that quote it, and both carry the same sentences. The Advocate annual report, the ministerial reply and the site-visit release were opened directly; the 107 percent unsheltered figure in the annual report rests on Everyone Counts 2024, which was not opened. 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