uniteTOlove.ca

Encampment Clearance: What It Costs, and What's Still Unreviewed — Playbook

What it actually costs Toronto to clear an encampment, who has sued over how it's done, and what the law allows.

DRAFTThe playbookThe evidence file

What Toronto can do about an enforcement framework nobody has independently checked for fairness, and a set of encampment counts the City can't even agree with itself on.

The honest bottom line

Toronto already paid for the answers — the 2021 clearance operation alone cost roughly $2 million, itemized down to $792,668 in turf restoration. What the city hasn't paid for yet is finding out whether the enforcement framework replacing that era's practices actually fixed what its own Ombudsman found unfair, including the uncertainties. Toronto's Ombudsman found "significant unfairness" in 2021 clearance practices and issued 23 recommendations. Four years and two Council votes later — a 2024 strategy motion and a November 2025 vote tightening enforcement to a three-shelter-offer, 50-metre-buffer, 24-hour-notice model — no independent body has reviewed whether the new framework repeats or avoids the problems the Ombudsman found. That is not a rhetorical gap. It is a specific, nameable thing City Council has not yet commissioned, using a mechanism it has already used once before. Meanwhile the numbers the City points to as evidence of progress are real: people living in tents fell from 539 in November 2024 to 196 in December 2025. The City's own Chief Financial Officer says the reduction is real enough to "do the math" on. A community worker who has publicly disputed the City's framing before now credits the improvement too — while still flagging that the data available isn't enough to fully understand what's happening on the ground. Both things are true at once, and neither cancels the other out. A 2021 lawsuit alleging police used excessive force clearing the Lamport Stadium encampment is still unresolved. And a Charter ruling just handed down against a neighbouring region's clearance bylaw — struck down on two separate grounds, not one: that the bylaw offered no alternative site at all (a security-of-the-person violation), and, separately, that it discriminated against Indigenous residents, residents with disabilities, and women and gender-diverse residents, with the court recognizing homelessness itself as a Charter-protected ground of discrimination — is now under active appeal, with Ontario's Premier having publicly floated overriding it with the notwithstanding clause. Toronto's own three-shelter-offer model is structurally different from what got struck down elsewhere on the no-alternative-site point. It has also never been tested in court against either Charter floor. Two things are worth doing, at realistic cost. First: commission an independent fairness review of the current enforcement framework — not the 2021 practices, the 2025 ones — before Council votes to extend or tighten it again. The 2021 Ombudsman review is the model: it happened using existing municipal authority, no provincial or federal sign-off required, and it's the only reason anyone can name what was wrong with the old approach at all. Second: fix the count problem, because right now the city can't even agree with itself on the numbers. Toronto is currently running four different, non-comparable ways of counting encampments, and even a same-week check this review turned up a one-unit discrepancy between two City-sourced park counts that nobody has reconciled. Requiring a single reconciliation table each budget cycle costs low tens of thousands of dollars, because the underlying data already exists — this is a presentation fix, not new data collection. It would not settle the argument between "the numbers are down because things are better" and "the numbers are down because people moved into stairwells and ravines" — but it would let that argument happen against numbers both sides can actually agree describe the same thing. Neither recommendation resolves whether Toronto's own three-shelter-offer model would survive a Charter challenge like the one that just succeeded elsewhere in Ontario. Neither resolves the Lamport Stadium lawsuit. And neither pretends the transparency fix would end the underlying disagreement about what a declining tent count actually means — it would only make the disagreement legible against numbers that don't shift definition between budget cycles.

This is a CONTESTED topic. Each card's Confidence & uncertainties section states plainly which side's framing the card leans on and why, rather than presenting the recommendation as a neutral consensus position.

---

a recommendation card — Housing-Led Encampment Response Protocol (Shelter-Offer-Plus-Alternative-Site Model)

Card id: a recommendation card · Issue: encampment-clearance-policy-costs · Backgrounder: our research file for that page · Trust: New load-bearing findings (problem framing) / carried-forward (cost anchors)

Problem

Toronto's own November 2025 enforcement framework (MM34.4) permits enforcement action after three shelter offers, with a 50-metre buffer and 24-hour notice [CL-572][CL-713] — but the City's most recent formal fairness review (the Ombudsman's 2021 "significant unfairness" finding and 23 recommendations) predates this framework by four years [CL-099], and no subsequent fairness review of MM34.4 itself has been identified in this page’s sources. Separately, the Region of Waterloo's 2025 clearance bylaw was found unconstitutional in May 2026 (2026 ONSC 2971) on two independent Charter grounds: section 7, specifically because the Region offered no alternative site at all if the encampment was cleared, and section 15(1) equality rights — Gibson J. recognized homelessness itself as an analogous ground and found the bylaw's enforcement discriminatory in effect against Indigenous residents, residents with disabilities, and women/gender-diverse residents. Both are structural gaps Toronto's three-shelter-offer model does not obviously share on the same facts, but has also not been independently fairness- or Charter-reviewed against on either ground. This card addresses the specific gap between "an enforcement framework exists" and "an enforcement framework has been independently reviewed for fairness," not encampment policy generally.

Action

Commission an independent (non-Ombudsman-repeat, third-party) fairness review of the MM34.4 enforcement framework specifically — the three-shelter-offer/50-metre/24-hour model — before its next scheduled Council reauthorization, modeled on the scope and independence of the City's own 2021 Ombudsman review [CL-099], and require that any Council vote extending or tightening the framework further be preceded by a published review finding, not an unreviewed status quo.

Jurisdiction split

Cost

Order-of-magnitude: low hundreds of thousands of dollars CAD, anchored to the scale of a single Ombudsman-style investigation producing 23 recommendations [CL-099] — no per-review cost figure for the 2021 Ombudsman investigation itself was located, so this range is anchored to comparable municipal-ombudsman investigation costs generally, not a cited Toronto-specific figure. Flagged as an estimate, not a sourced figure.

Funding path

Existing City of Toronto Ombudsman office operating budget, or a one-time Council-directed allocation to the Auditor General's office if independence from the Ombudsman's own prior finding is judged preferable — both are existing municipal budget lines, no new funding mechanism required.

Who benefits, and how

Encampment residents subject to MM34.4 enforcement, via an independent check on whether the three-offer/50-metre model repeats or avoids the 2021 problems the Ombudsman found; City Council itself, via a defensible basis for any future vote tightening or loosening the framework, rather than voting on enforcement policy with no post-implementation fairness data, as EC13.8 and MM34.4 both did.

Who bears the cost, and how

City taxpayers city-wide, via the Ombudsman/Auditor General office's existing budget or a modest one-time allocation; no identified group bears a disproportionate cost.

Who benefits from the status quo

Garda Canada Security Corporation/GardaWorld (ENT-0009), per ACL-0009 (backing backgrounder's Cui Bono table) — holds an $11.9 million encampment-support contract under the current unreviewed enforcement framework; an independent fairness review is a process safeguard on that framework, not a review of the security contract itself, but the same status-quo enforcement activity this card's review would scrutinize is the activity the contract is priced against.

Financial ROI

Not separately modeled. Directionally: the 2021 clearance-cost itemization (~$2 million for one operation) [CL-354] suggests repeat unreviewed clearances carry real, recurring cost; a fairness review that identifies avoidable friction (e.g., repeat clearance-and-return cycling) could plausibly reduce recurring costs, but no source quantifies this — stated as a plausible mechanism, not a costed estimate.

Economic ROI

Not yet estimable for the review mechanism itself. A genuine, on-topic comparator exists for encampment response costs generally: a HUD/HHS-commissioned study (Abt Associates) of four U.S. cities found total encampment-response spending of roughly $3.4-8.6 million/year per city (FY2019), at $1,672-$6,208 per unsheltered person depending on the city. That figure describes the cost of encampment response generally (clearance, prevention, shelter, outreach combined), not the narrower economic impact of commissioning an independent fairness review, which is this card's actual Action — named here as context but not used as this card's own range. Confidence: low — the HUD comparator is real but answers a different question than this card asks.

Social ROI

Directional: the Ombudsman's 2021 finding of "significant unfairness" and its 23 recommendations [CL-099] establish that unreviewed clearance practice has previously produced real, documented harm; a review mechanism addresses that specific documented risk. No claim in this page quantifies a social-outcome improvement from review itself — this is a process safeguard, not an outcome intervention.

Environmental ROI

Genuinely environmentally neutral — commissioning a fairness review of an enforcement framework has no direct emissions, land-use, water, or waste mechanism; it does not itself alter how or whether clearances occur. (A live-discovery search this review on encampment-clearance environmental impact found real material on clearance operations themselves — waste/debris cleanup, displacement scattering — but that describes clearance activity, not a review of the enforcement policy governing it, so it is not used as a comparator here.) Confidence: high — no plausible environmental mechanism attaches to commissioning a review.

Evidence

Confidence & uncertainties

Medium confidence. This card leans toward the encampment-rights/oversight framing (an enforcement framework should be independently reviewed before being extended) rather than the clearance-advocate framing (enforcement tools, once Council-adopted, don't need a standing review requirement to function). The cost estimate is explicitly a rough anchor, not a sourced figure — flagged rather than presented with false precision, per the costing bar.

Status

DRAFT — blocked on: a real per-investigation cost comparator (current range is an estimate); fairness and legal review.

---

a recommendation card — Encampment Clearance-Cost Transparency (Single Reconciled Reporting Standard)

Card id: a recommendation card · Issue: encampment-clearance-policy-costs · Backgrounder: our research file for that page · Trust: New load-bearing findings (problem framing) / carried-forward (cost/count anchors)

Problem

Toronto's own reporting uses at least three or four genuinely different, non-comparable encampment metrics across different contexts [CL-470] — total sites (156→91), parks-with-tents (107→49 per one source, or "48... down from 107" per a separately fetched source this review, a one-unit variance not yet reconciled), street sites (49→42) [CL-204], City-property-specific counts (283→84) [CL-431], and people-in-tents (539→196) [CL-90684]. Direct clearance costs have been itemized in full detail only once (the 2021 operation, ~$2 million) [CL-354]; no source in this page shows a comparable itemized figure for any clearance since. This card addresses the reporting-transparency gap specifically, not the encampment-count trend itself.

Action

Require Toronto Shelter and Support Services and Parks, Forestry and Recreation to publish, in each annual budget cycle, a single reconciliation table naming every encampment-count metric in use that cycle (site count, parks-with-tents, street sites, people-in-tents, City-property-specific counts), the exact date each was measured, and — for any clearance operation costing more than a stated threshold — an itemized cost breakdown matching the level of detail the 2021 operation happened to receive [CL-354], rather than leaving that level of detail to whichever year a journalist happens to file a public-records request.

Jurisdiction split

Cost

Order-of-magnitude: low tens of thousands of dollars CAD for the reporting/reconciliation work itself, anchored to the scale of existing budget-note production (TSSS and PFR already produce detailed budget notes annually) — this is treating an existing practice as a template, not proposing new infrastructure. Not a per-clearance itemization cost, which varies with clearance scale (the 2021 operation itself cost roughly $2 million) [CL-354].

Funding path

Existing TSSS/PFR budget-note production process — no new funding mechanism, since the underlying data (counts, costs) is already collected internally; this is a presentation/reconciliation requirement, not new data collection.

Who benefits, and how

Journalists, advocates, and the public generally, via a single authoritative source rather than needing to cross-reference Global News, CBC, and City budget documents to determine which metric a given "encampments are up/down" claim refers to — a real, named transparency gap this page’s own sources document [CL-470]; Council itself, via a clearer basis for evaluating whether its own enforcement votes (EC13.8, MM34.4) are producing the outcomes claimed.

Who bears the cost, and how

City taxpayers city-wide, via the TSSS/PFR budget process; no disproportionate cost to any named group.

Who benefits from the status quo

Garda Canada Security Corporation/GardaWorld (ENT-0009), per ACL-0009 (backing backgrounder's Cui Bono table) — the current reporting patchwork this card addresses (multiple non-comparable encampment-count metrics) makes it harder for Council or the public to evaluate whether contracted encampment-security spending is tracking a genuine reduction in encampments or a redefinition of what counts as one; a single reconciled reporting standard would make that comparison more visible, without itself changing the contract.

Financial ROI

Not modeled — this is a transparency measure, not a cost-avoidance one. Indirectly, better-reconciled data could reduce the kind of interpretive dispute already visible between City officials and advocates (e.g., CFO Conforti's "you can do the math on the reduction" framing versus Diana Chan McNally's "without all the resources and without all the data available, it is hard... to understand," Global News, Jan. 14, 2026) — stated as a plausible benefit, not a quantified one.

Economic ROI

Not yet estimable. The same HUD/Abt Associates encampment-response cost study surfaced for a recommendation card documents real per-city response costs ($3.4-8.6M/year) and per-person costs ($1,672-$6,208), but that measures the economic scale of encampment response itself, not the economic effect of standardizing how that spending is reported, which is this card's actual Action. No source quantifies the economic impact of a reconciled-reporting-standard requirement specifically. Confidence: low — genuine gap; the adjacent HUD figures are named for context but not repurposed as this card's own range.

Social ROI

Directional: a single reconciled reporting standard would let the disputed interpretation already documented in this page [CL-431] — whether declining site counts reflect real reduction or displacement into stairwells, garages, transit, and ravines — be assessed against a stable metric set year over year, rather than against metrics that change definition between reporting cycles.

Environmental ROI

Genuinely environmentally neutral — requiring a single reconciled reporting table has no direct emissions, land-use, water, or waste mechanism; it changes how existing data is presented, not what clearance or encampment activity occurs. Confidence: high — no plausible environmental mechanism attaches to a reporting-standardization requirement.

Evidence

Confidence & uncertainties

Medium-high confidence on the transparency gap itself (well-documented across multiple independent sources); lower confidence on whether a single reconciliation table would resolve the underlying interpretive dispute (displacement vs. reduction) rather than just making the dispute easier to litigate in public — this card takes the position that clearer data helps regardless, which is a judgment call, not a proven outcome.

Status

DRAFT — blocked on: fairness and legal review; a second funding/cost comparator would strengthen the estimate.

---

Production record

Drafting record

Version: v1.1 · Date: 2026-07-13 · Status: DRAFT · Provenance: mixed — problem statements and cost anchors are carried-forward (a formally registered claim) or NEW (this review, inline source quote); the recommendations themselves are this document's own L-PLATFORM position, per the L6 template's firewall discipline.

Playbook conversion (2026-08-11, Lane L3a): opened with "The honest bottom line" adapted from archive/dayone/encampment-clearance-policy-costs.md (a recorded standing decision retired day-one memo, kept as history in archive/); ROI sections tightened, repeated "not yet estimable / genuine gap" boilerplate collapsed to one honest line each, matching that page's recommendation cards's playbook shape. All 9 a formally registered claim tokens (CL-099, CL-204, CL-354, CL-431, CL-470, CL-572, CL-713, CL-90684, and the CL-0009 substring within ACL-0009) preserved exactly, verified via bash token-set comparison before and after this edit.

About this file