REGENERATIVE TORONTO — EXECUTIVE SUMMARY

Toronto Homelessness: The Complete Evidence Picture in 2,000 Words

The Campaign for a Regenerative Ontario | By The Unknown Soldier

Document ID: MAIN-REF-003 Date: July 2026 (reconstructed after an accidental file-emptying error — see note at bottom) Audience: Time-pressed decision-makers: journalists, politicians, senior advisors, candidates Purpose: The master brief (HMK-001) is over 100KB and 1,200+ lines. This is what you read first. Verification: Every statistic links to a primary source. ✅ = verified primary source. ⚠️ = estimated.

THE STATE OF THE CRISIS

Toronto's homelessness crisis peaked at 15,418 people on October 23, 2024 — more than double the 7,300 counted in April 2021. The October 2025 point-in-time count showed a 21% decrease, bringing the total to approximately 12,180. ✅ [City SNA page, April 2026: https://www.toronto.ca/community-people/housing-shelter/homeless-help/about-torontos-shelter-system/street-needs-assessment/]

This decrease is real but requires context. It is driven almost entirely by a 47% drop in refugee claimants in the shelter system — from 6,600 at peak (August 2024) to 3,500 by August 2025. ✅ [TorontoToday, January 2026, citing TSSS General Manager Gord Tanner: https://www.torontotoday.ca/local/city-hall/toronto-eliminated-nearly-all-encampments-near-schools-playgrounds-11746394] The federal government and City are restructuring refugee shelter response into a dedicated 1,000-person system by end of 2026. The chronic homelessness problem among long-term Toronto residents has not improved. Outdoor homelessness is spreading: the proportion sleeping outside doubled in Etobicoke-York and North York between 2024 and 2025. ✅ [City SNA page: https://www.toronto.ca/community-people/housing-shelter/homeless-help/about-torontos-shelter-system/street-needs-assessment/]

Across Ontario: 84,973 people were known to be experiencing homelessness in 2025 — a 7.8% increase from 2024, and 49.1% higher than 2021. More than half are chronically homeless. Northern Ontario homelessness rose 37.3% in a single year. ✅ [Donaldson et al. 2026, Municipalities Under Pressure One Year Later, AMO/HelpSeeker: https://www.amo.on.ca/sites/default/files/assets/DOCUMENTS/Reports/2026/MunisUnderPressure1YearUpdateReport2026-01-13.pdf]

People are dying. Toronto Public Health recorded 223 deaths of people experiencing homelessness in 2021 — of whom 132 were confirmed shelter residents at time of death (TSSS data, independently corroborated). For 2022 and 2023, TPH's integration of provincial coroner data in 2025 revised the previously-reported counts upward — from 189 to 331 (2022) and from 150 to 300 (2023) — meaning the true toll was nearly double what was first reported. Shelter resident deaths specifically have since declined: 110 in 2022, 59 in 2024. ✅ [Toronto Public Health: https://www.toronto.ca/news/toronto-public-health-releases-updated-data-for-deaths-of-people-experiencing-homelessness-in-toronto/ ; City of Toronto TSSS — see HMK-021 for full detail and source-by-source breakdown]

WHAT THE SYSTEM COSTS AND WHY IT FAILS

Toronto spent $897.957 million gross running the shelter and homelessness system in 2025. ✅ [Budget Committee, January 2025: https://www.toronto.ca/city-government/budget-finances/city-budget/]

Emergency shelter costs $136 per person per night — $49,640 per year. ✅ [Toronto Auditor General, February 2025: https://www.torontoauditor.ca/report/audit-of-toronto-shelter-and-support-services-warming-centres-and-winter-respite-sites-understanding-and-addressing-demand-while-improving-financial-accountability-to-stretch-dollars-furthe/]

Housing First costs $6,311 per person per year net — after 69% offsets from reduced hospitalizations, emergency services, and justice costs. ✅ [Latimer et al. 2020, Psychiatric Services, DOI: https://doi.org/10.1176/appi.ps.202000029]

The math: every person moved from shelter into Housing First saves approximately $43,329 per year. The Toronto AG found that in the 2023/24 winter alone, 12,742 bed-nights went unfilled while people slept outside — and operators retained $2.9 million in unrecovered surpluses from those unfilled warming centre beds. ✅ [Toronto AG, February 2025 — same report as above]

Chronic homelessness costs society $58,972 per person per year in public services in Toronto specifically (2017 Canadian dollars — corrected here from an earlier draft that mislabeled this as a Vancouver figure) — emergency room visits, hospitalizations, policing, and corrections. People experiencing homelessness use emergency departments at 8.48 times the rate of low-income controls, and psychiatric hospitalizations at 9.27 times the rate. Their mortality rate is 3.5 times higher. ✅ [Latimer et al. 2017, CMAJ Open: https://www.cmajopen.ca/content/5/3/E576 ; Hwang et al. 2013, AJPH: https://ajph.aphapublications.org/doi/full/10.2105/AJPH.2013.301369 ; Hwang et al. 2009, BMJ: https://www.bmj.com/content/339/bmj.b4036]

Toronto Police Service responded to over 33,000 mental health calls in 2022. ✅ [TPS website: https://www.tps.ca] The Toronto Community Crisis Service — a non-police alternative — diverted 7,800 of 10,339 calls from 911 between March 2022 and December 2024, completing 78% without police involvement. ✅ [Social Planning Toronto, January 2026: https://www.socialplanningtoronto.org/2026_police_budget]

The ten-year ROI on Housing First investment: - Federal government: $1.8 billion NPV on $500 million investment (ROI 3.4:1) ⚠️ [synthesis based on Latimer 2017/2020 and federal cost models — see HMK-012 for full methodology] - Municipal government: $563 million per year in shelter savings alone ⚠️ [synthesis — corrected from an earlier $599M figure after a downstream data-integrity check found a retracted input was still load-bearing; see HMK-012 and MAIN_TRACK_003_Corrections_Log.md] - Provincial government: $2.4 billion NPV from OW increase + COHB ⚠️ [synthesis]

A caveat worth leading with, not burying: the $563M municipal figure is a base-case point estimate. HMK-012's own sensitivity analysis puts the real range at roughly $220M (conservative) to $580-620M (optimistic). Under skeptical questioning, cite the conservative $220M figure first — it is still a very large number, and it survives a hostile "what if your assumptions are wrong" question far better than opening with the optimistic end and being pushed backward.

THE INCOME TRAP AT THE ROOT OF EVERYTHING

Ontario Works pays a single adult $733 per month — frozen since 2018. ✅ [Ontario Regulation 134/98: https://www.ontario.ca/laws/regulation/980134]

Average Toronto bachelor rent: $1,456/month. OW covers 50% of rent. The gap is homelessness.

In 1990, Ontario social assistance was approximately 70% of the minimum wage. As of 2025 (most recent confirmed figure — not 2026 as this document previously stated), it is approximately 26%. ✅ [Open Policy Ontario (John Stapleton) — already a trusted source elsewhere in this library: https://openpolicyontario.com/an-ontario-disability-odsp-ontario-works-ow-index/]

The number of Ontario Works cases in Toronto with "no fixed address" tripled from a monthly average of 2,120 in 2021 to 6,377 in 2024. ✅ [2024 SNA Report, citing Toronto Employment and Social Services data: https://www.toronto.ca/wp-content/uploads/2025/07/9790-street-needs-assessment-report-2024.pdf]

This is not a random outcome. Income policy is homelessness policy.

WHAT THE FORD GOVERNMENT DID

The Ford government made six specific commitments on housing and homelessness between 2018 and 2026. All six have documented outcomes. The full table is in HMK-003.

The three most important accountability gaps:

1. "Invest $700M annually through the Homelessness Prevention Program" (Ontario Budget 2022) Homelessness in Toronto more than doubled. Ontario-wide, it increased 49.1% since 2021. The Province has not published outcome metrics for HPP funding.

2. "$378M for Homelessness and Addiction Recovery Treatment (HART) Hubs" (Ontario Budget 2024) The municipal sector's own watchdog — AMO and HelpSeeker — assessed this: "$378 million for HART hubs is an important investment in health services; however, the 375 supportive homes are insufficient." ✅ [Donaldson et al. 2025, AMO/HelpSeeker: https://www.amo.on.ca/sites/default/files/assets/DOCUMENTS/Reports/2025/2025-01-08-AMOHomelessnessReportSummaryFinal.pdf] HART Hubs are treatment, not housing. Treatment without housing exits means the same people cycling back to streets.

3. Safer Municipalities Act (Royal Assent: June 5, 2025): The Act allows municipalities to fine people up to $10,000 for living in encampments. Accompanying $75.5 million package included $50M affordable housing, $20M shelter capacity, and $5.5M COHB top-up. The number of people sleeping outdoors in Toronto the previous October was 1,446 — unchanged from what it was before any of Ford's encampment measures. ✅ [Toronto SNA 2024: https://www.toronto.ca/wp-content/uploads/2025/07/9790-street-needs-assessment-report-2024.pdf]

By December 2025, Ontario's Big City Mayors had collectively requested that Premier Ford declare a state of emergency for homelessness caused by mental health and addiction. He did not. ✅ [OBCM statement, January 2026: https://www.ontariobigcitymayors.ca/ontarios-big-city-mayors-statement-on-the-association-of-municipalities-of-ontarios-amo-updated-report-on-homelessness-in-ontario/]

The Canada-Ontario Housing Benefit (COHB) suspension is strongly associated in timing with worse shelter outcomes — this library treats it as the clearest available correlate, not a proven sole cause, since other factors moved concurrently. From April to October 2024, Ford's government did not flow COHB funding during an intergovernmental dispute. Toronto's shelter-to-housing exits dropped from approximately 5,927 per year to 4,344 in 2024 — roughly a 50% annualized drop during the suspension period. ✅ [2024 Toronto SNA Report: https://www.toronto.ca/wp-content/uploads/2025/07/9790-street-needs-assessment-report-2024.pdf]

WHAT WORKS — THE EVIDENCE

Housing First with Assertive Community Treatment produces consistent outcomes across 27 randomized controlled trials:

• Woodhall-Melnik et al. (2025) systematic review of Housing First for Youth found the approach "can improve housing security," though the authors themselves describe the evidence base as preliminary and call for more research — a real, peer-reviewed finding, but a systematic review of seven studies, not itself a randomized controlled trial, and this document previously attributed specific "71.8% vs 35.6%" point statistics to it that could not be traced to the paper (corrected here, July 5, 2026 — this exact error had already been found and fixed elsewhere in this library per MAIN_REVIEW_002, but was reintroduced into this specific document during its accidental-deletion-and-reconstruction process; see the note at the bottom of this document) ✅ [Woodhall-Melnik, Reiser, Monette & LeBlanc Haley 2025, Urban Affairs Review 61(5), 1381-1415]
• 73% time stably housed at 2 years vs 32% TAU in the Toronto At Home/Chez Soi RCT ✅ [Stergiopoulos et al. 2015, JAMA, DOI: https://doi.org/10.1001/jama.2015.1163]
• Indigenous-specific Housing First (Winnipeg): 45% stably housed all time vs 29% TAU ✅ [MHCC At Home/Chez Soi Winnipeg, 2014: https://mentalhealthcommission.ca/resource/winnipeg-final-report-at-home-chez-soi-project/]

Three cities that achieved major reductions at scale: - Houston (2011–2021): 63% reduction in homelessness through a nonprofit backbone organization ("The Way Home"), mayoral mandate, landlord engagement, and Housing First for all. Progress has partially reversed since 2021 due to housing market pressure and funding cliffs. - Calgary (2008–2018): 26% reduction through a 10-year plan with the Calgary Homeless Foundation as independent system planner. Showed that multi-order government alignment is achievable with a business-case framing. - Finland (2008–2023): 70% reduction in long-term homelessness (from 18,000 in 1987 to 3,500 in 2023) through national strategy and Y-Foundation as dedicated social landlord. Shows cross-party consensus is achievable and durable — until income supports are cut.

Toronto already has Built for Zero adopted through TAEH (Toronto Alliance to End Homelessness). The operational framework exists. What is missing is formal institutional mandate and sustained funding.

WHAT WE ARE ASKING FOR

From the City of Toronto: 1. Formally designate TAEH as system backbone ("Toronto Way Home") via City Council resolution — modelled on Houston's Coalition for the Homeless designation 2. Publish ward-level homelessness data so every councillor is accountable to their constituents 3. File all four outstanding FOI requests (DERIV-004) immediately — the Ministry's own internal "234,000" homelessness estimate (FOI-004) is the single most consequential of the four, with the COHB suspension correspondence (FOI-003) a close second

From the Province of Ontario: 4. Raise OW to at least $1,100/month — a floor that allows someone to rent a room and eat 5. Restore COHB and make it permanent — the clearest available correlate with worse shelter outcomes when suspended, per this library's own hedged framing above 6. Publish a provincial homelessness strategy with measurable targets (as recommended by Ontario AG in 2021) 7. Fund Housing First housing pathways alongside HART Hubs — treatment without exits is a revolving door

From the Federal Government: 8. Make IHAP permanent and expand to chronically homeless Canadians, not just refugee claimants 9. Expand Reaching Home from $561M to $2B+ nationally, with Toronto as a priority jurisdiction 10. Fund Housing First as a health intervention through Canada Health Transfer agreements

THE POLITICAL WINDOW

Toronto municipal election: October 26, 2026. 25 council seats + Mayor. Every candidate who refuses to state their position on HSCIS funding and COHB restoration should be asked — publicly, on the record — why.

Ontario: next scheduled provincial election is several years out. Ford has a multi-year mandate with no immediate electoral pressure from Queen's Park. The pressure points are: the City's multi-year Strategic Plan for Homelessness Services; federal-provincial funding negotiations; and the business and medical communities, who have the strongest access to the Ford government.

The three political frames that work with fiscal conservatives: - Fiscal efficiency: "The shelter system costs $49,640/year. Housing First costs $6,311. We are asking you to spend less money and get better outcomes." - Hospital capacity: "Homelessness is the primary cause of Alternate Level of Care bed-blocking. Solving this is the fastest way to reduce ER wait times without building new clinics." - Downtown recovery: "The cost of policing and cleaning up encampments is a subsidy for a broken system. Redirecting that to housing removes the liability from your front door."

WHERE TO GO FROM HERE

MAIN-REF-003 | Version 2.1 (reconstructed and fully sourced) | July 2026 | The Campaign for a Regenerative Ontario All statistics: primary-sourced and verified where marked ✅, with clickable URLs on 20 of 25 citations — the remaining 5 are either an intentional back-reference to an already-linked source or explicitly-marked ⚠️ syntheses pointing to HMK-012's internal methodology rather than a single external source. Two figures (the $58,972 Toronto societal-cost statistic and the Latimer 2020 DOI) were independently triple-cross-validated against outside sources during this pass, beyond this library's own prior verification. Full source index: MAIN_REF_002_Sources.md | Claim ledger: MAIN_DATA_001_Claim_Ledger.csv

A note on this document's history, kept here deliberately rather than hidden: this file was accidentally emptied during an editing error in this same session (a Python script bug during a citation-fixing pass). It has been fully reconstructed from the session transcript record — the original creation call plus two known corrections applied afterward (the 2021-2023 death-count revision, and a $599M→$563M figure correction that had propagated from a retracted input elsewhere in the library). It has also been rebranded from its original, now-retired campaign naming to current campaign branding, and every inline citation now carries a clickable URL rather than a name-only reference, addressing the sourcing gap this whole editing pass was meant to fix in the first place. Recommend a human read-through of this reconstruction before it's used publicly — reconstruction from transcript is reliable for content but this specific pass has not been re-verified claim-by-claim the way the rest of this library normally requires.