Ending Homelessness in Toronto · Chapter 8

Who Has Done It

A chapter of Ending Homelessness in Toronto — this project’s receipt-backed book on ending homelessness for less than the crisis costs now. Where a figure could not be confirmed against a primary source, the text says so on its face. How the checking works: how we verify · corrections.

No city has a perfect record on homelessness, and this chapter does not pretend otherwise. What follows are the best-documented cases anywhere in the world of a city or country moving the number down — reported honestly, including where the story reversed, stalled, or turned out to rest on shakier ground than its own headline suggested. Toronto does not need a perfect model to copy. It needs an honest one.

Finland — the strongest policy architecture, honestly reported

Finland remains the best-documented national success at reducing homelessness, and the most important thing to get right about it is that the story is not finished. Homelessness fell from over 18,000 people in 1987 to under 5,000 by 2019 — Helsinki's shelter beds fell from 2,121 in 1985 to 52 in 2016, replaced by supported and independent housing — through a national shift from a conditional "staircase" model to unconditional Housing First, formalized in a 2007 national report and delivered through the Y-Foundation (Y-Säätiö), which became Finland's fourth-largest landlord and now manages over 19,000 apartments, funded through the state housing finance agency via construction subsidies and interest guarantees (Marybeth Shinn & Jill Khadduri, "How Finland Ended Homelessness," Cityscape, Vol. 22, No. 2, U.S. Department of Housing and Urban Development, 2020; Y-Säätiö, "History").

But after eleven consecutive years of decline, Finnish homelessness rose in 2024 and rose sharply again in 2025 — up 20% year-over-year, with the sharpest increases among people under 25, women, and people of foreign background — which the Y-Foundation's own chief executive attributes directly to cuts in housing allowance and social assistance (Y-Säätiö, "The Rise in Homelessness Is the Largest in Recorded History," March 3, 2026). "Cuts to housing allowance and social assistance in a societal situation where many individuals and families are already financially stretched create situations where people are unable to hold on to their homes," in the words of the Foundation's chief executive, Teija Ojankoski. Long-term homelessness rose even faster than the headline number: 1,306 people were long-term homeless in the 2025 count, up 29% in a single year. This is the first consecutive-year rise in Finnish homelessness since 2009 — since before the national Housing First programme existed at all.

The lesson for Toronto is precise, not vague: Housing First infrastructure — a dedicated capital agency, a large non-profit landlord, unconditional access — is necessary but not sufficient. It has to be paired with income supports that keep pace with the cost of living, or the gains erode within a few budget cycles. That is the same warning this book's own analysis of Ontario's frozen income-support rates makes for Toronto directly, and Finland's 2025 reversal is not a footnote to be minimized — it belongs at the center of how this book argues for permanent, not one-term, funding.

Houston, Texas — regional coordination and a public-private funding base

Houston's "The Way Home" collaborative, spanning Harris, Fort Bend, and Montgomery counties, reported a 57% reduction in homelessness since 2011 by 2016, and more recent figures cited by its leadership put cumulative reductions above 60% against that same 2011 baseline, and more than 28,000 people housed since 2012, the year the collaborative's Continuum of Care formally launched (Marc Eichenbaum & Michael Nichols, "Opinion: How Houston's homelessness breakthrough could be a national game-changer," CNN, June 8, 2023). Those two headline figures use two different base years — 2011 for the reduction percentage, 2012 for the housed count — and should never be spliced into a single "60%-plus since 2012" claim; each is real on its own terms, but not interchangeable.

The reported cost is roughly $18,000 per person housed per year, against $30,000 to $50,000-plus for continued jail or emergency-service use. The mechanism is a single regional coordinated-access system spanning roughly 100 agencies, combining federal Continuum of Care funding, city and county government, and private philanthropy — a model directly relevant to the fractured, three-government coordination problem this book documents for Toronto. Houston's precise percentages have varied across press cycles and across which year's baseline is used, and should be read with that caution rather than treated as a single fixed figure.

Vienna — social housing at a scale North America has never attempted

Vienna's municipal housing company houses roughly 220,000 units directly — about 22% of the city's total housing stock in the city-owned portion specifically — and, combined with the city's limited-profit co-operative sector, social and subsidized housing overall accounts for an estimated 45% of Vienna's total housing stock. The City of Vienna's own official figure states that around 50% of Viennese residents live in subsidized dwellings (Socialhousing.wien, "The Vienna Model"). A commonly circulated figure of 60% or more of Vienna's population living in subsidized housing traces to a broader definition used by an outside source, not the City of Vienna's own accounting, and this book does not repeat it.

Vienna also runs proactive eviction-prevention services triggered automatically when an eviction notice is filed. Vienna is not a template Toronto can replicate at speed — its housing stock was built over a century — but it is the clearest existing proof that a large Western city can treat social housing as permanent public infrastructure rather than a residual, means-tested program, which is the underlying premise of Canada's own National Housing Strategy.

Scotland — a live, national-scale rollout

Scotland's Housing First Pathfinder — a 2018–2021 program spanning Aberdeen City, Aberdeenshire, Dundee, Edinburgh, Glasgow, and Stirling, funded with over £5.5 million from the Scottish Government, £2 million from the charity Social Bite, and £150,000 from the Merchants House of Glasgow — housed and supported 579 people, with tenancy sustainment rates of 88% at twelve months and 80% at twenty-four months (Heriot-Watt University / I-SPHERE, "Benefits of scaling 'housing first' approach to address homelessness," November 17, 2022).

The Pathfinder was folded into a national rollout: as of March 31, 2025, an estimated 2,154 Housing First tenancies had started across Scotland — the Pathfinder's roughly 834 tenancies are a subset of that total, not double-counted — with a twelve-month sustainment rate of 83% (Scottish Government, "Housing First: monitoring reports — 1 October 2024 to 31 March 2025"). The model is scatter-site housing, dispersed among ordinary neighbourhoods rather than concentrated in dedicated buildings, with a recommended staff-to-tenant ratio of one worker to seven tenants, though some areas have operated above that ratio under resource constraints. For Toronto, Scotland is the closest thing anywhere to a live, national-scale Housing First rollout with clean tenancy-sustainment data across multiple large cities running at once — a direct comparator to the three-government coordination problem this book documents at home.

Utah — a cautionary tale about measurement, not a model to copy

Utah's widely repeated claim of a 91% reduction in chronic homelessness between 2005 and 2015 does not hold up. American Enterprise Institute researcher Kevin Corinth found the underlying counts used inconsistent annualization methods from year to year — "the 2009 count was doubled, the 2011 count was less than doubled, and the 2015 count was not adjusted at all" — and that the real, methodologically consistent reduction was closer to 71%, concentrated in two anomalous years rather than a steady decline (Kevin Corinth, "On Utah’s 91 Percent Decrease in Chronic Homelessness," American Enterprise Institute, March 2016).

A 2025 follow-up documents that Utah subsequently changed its chronic-homelessness definition twice, both times coinciding with its largest reported single-year drops; that chronic homelessness actually rose 84% between 2010 and 2014, a period of stable methodology; and that it has risen roughly 400% since 2015 — with Utah having since abandoned Housing First altogether (Christopher J. Calton, "The Rise and Fall of ‘Housing First’ in Utah," Independent Institute, September 2, 2025). Toronto should treat Utah as a warning about two specific risks this book takes seriously: declaring victory on a metric that can be redefined, and treating a program's early success as evidence it no longer needs sustained funding.

England, Glasgow, and Toronto's closest domestic peers

England's three regional Housing First pilots — West Midlands, Liverpool City Region, and Greater Manchester — funded with £28 million from 2019, housed 1,061 people with complex needs, with measured retention rising from 84% to 92% at twelve months across the evaluation's waves, and roughly 84% of those still tracked (69% on the strictest denominator) in stable accommodation around three years in (UK Department for Levelling Up, Housing and Communities, "Evaluation of the Housing First Pilots — Final Synthesis Report," October 2024). Glasgow's programme grew from 31 tenancies in 2018 to 318 by mid-2024, with 45 of 61 evaluated tenancies (roughly 74%) still held at follow-up — a multi-year figure, not a one-year rate — among a cohort that had, between them, 220 prior homelessness applications dating back to 1994 — direct evidence of the revolving-door pattern Housing First is designed to break (Homeless Network Scotland, "Housing First in Focus: Glasgow," July 2024).

Closest to home, Medicine Hat, Alberta became the first Canadian city to reach functional zero for chronic homelessness under the Built for Zero Canada methodology, confirmed June 2, 2021 — the most directly transferable evidence available to Toronto, because it runs on Canadian funding structures and Canadian data systems, through the same national alliance already active here. But Medicine Hat is also an honest caution, not just a model, and its record deserves to be told in full: its 2021 functional-zero status lapsed within months, with roughly 40 people found sleeping rough in a November 2021 count. That was not the city's first such reversal — Medicine Hat had separately declared it had "ended homelessness" in 2015, the target year of its original 2008/2009 ten-year plan, and that earlier declaration also lapsed, within roughly fifteen months. Two declarations, two lapses, in the smallest and most mature coordinated-entry system in the country. The lesson is not that Housing First failed in Medicine Hat — it is that functional zero has to be maintained through permanent, indexed funding for both housing supply and support services, not declared once and left to erode. London, Ontario has separately become the first Canadian city to functionally end veteran homelessness, using the same methodology.

Every case in this chapter argues the same two things at once: the tools to end homelessness at scale already exist and are proven, and every one of them has failed somewhere the moment funding stopped being permanent. Toronto's own plan, set out in the chapters before this one, is built to answer the second half of that lesson, not just the first. Readers who want the full underlying research base behind every case in this chapter — books, primary reports, and the strongest critical literature alongside it — will find it gathered at the end of this book.