FROZEN EDITION — July 2026. This page is part of the v1 Toronto homelessness research library, a complete, sealed project frozen at its July 2026 figures (92% of its 770 claims individually verified against primary sources). It is a product of its July-2026 campaign context, published as a citable historical artifact. Where our newer research disagrees, a dated margin note points at the living page — the frozen text is never silently rewritten. Why there are two editions.

Declining Shelter-Cost Transparency

THE CORE FINDING, STATED PLAINLY

Toronto used to publish exactly how much it pays each shelter operator per bed, per night. It stopped, and has not resumed, for over a decade — even as the system's cost has grown roughly tenfold and its scale has roughly doubled.

From 2010 through at least 2012, the City published a detailed, shelter-by-shelter per diem schedule: dozens of named operators, each with a specific dollar figure, in a public staff report anyone could read. This document has independently reconstructed that entire 2010 schedule — every sector, every named shelter, rates ranging from $20.25 (Fife House, no food) to $75.75 (St. Vincent de Paul – Amelie). It is real, it is detailed, and it is exactly the kind of granular public accountability record a functioning oversight system should produce every year.

It does not exist in that form anymore. The most recent equivalent this library could find is from 2012. Every current TSSS budget document — 2024, 2025, 2026, all read directly and in full — references "inflationary adjustments to per diems, grants, room rent, utilities and food costs" as a line item, without ever stating what the resulting rate actually is, for the system as a whole or for any individual operator.


THE EVIDENCE, LAYERED

Layer 1 — direct confirmation from multiple independent sources. Multiple separately-conducted research passes, working independently of each other, arrived at the identical conclusion. One: "The City has not published a detailed shelter-by-shelter per diem schedule for the 2020s in the same way it did for 2010-2012... the exact current per diem for each POS operator is treated as contract information and is not publicly itemized in budget documents." A separate, independent pass: "No Central Database... the most detailed public per diem data is from 2011... For the most current rates, you would likely need to file a formal access-to-information request." Neither source was working from the other's output — this is convergent, not repeated, evidence.

Layer 2 — a dedicated direct search of this library's own, confirming the negative result. Rather than accept the above secondhand, this library independently searched the 2024, 2025, and 2026 TSSS Budget Notes and Program Summary documents directly, in full, specifically looking for a current average or per-operator per diem figure. None was found. The budget documents are, in every other respect, detailed — precise bed-night counts, precise revenue splits by funding source, a tracked "sleeping accommodation turnover" metric down to two decimal places. The absence of a per diem rate in documents this granular in every other respect is conspicuous, not incidental.

Layer 3 — a data point that sharpens the finding rather than softens it. The campaign researcher separately located a reported 2025 figure ($146/person/night) inside a .docx file accessed through the City's council meeting data portal — not the headline budget documents, a specific committee attachment. This library's own search tools, despite a dedicated attempt, could not independently locate this figure through normal web search. Taken together, this suggests the data likely still exists somewhere in Toronto's records — it is not that the City has stopped calculating this number internally — but that it is no longer surfaced anywhere a member of the public, a journalist, or an opposing candidate would naturally encounter it. That is arguably a more serious accountability problem than the number simply not existing: it means the transparency exists on paper, in a technical sense, while being functionally inaccessible in practice.

Layer 4 — the scale this has happened at. In 2010, when this data was last published in full, the entire Purchase-of-Service shelter system's operating budget was $60.7 million. TSSS's 2025 gross budget is $897.957 million. The system that stopped publishing granular per-operator costs is the same system whose costs grew roughly fifteenfold over the same window (with the important caveat, established in HMK-038, that this fifteenfold figure spans genuine mandate expansion as well as pure cost growth — see HMK-038 Part 3 for why that comparison needs care). The point stands regardless of exactly how that growth decomposes: oversight granularity moved in the opposite direction from cost and scale.


WHY THIS MATTERS AS ITS OWN ARGUMENT, SEPARATE FROM THE COST NUMBERS

A hostile reader can argue with almost any specific cost figure in this library — dispute the inflation adjustment, dispute the scope comparison, dispute whether service intensity explains the growth. It is much harder to argue with an absence. Either the City publishes a current per-operator or system-wide per diem rate, or it doesn't. This library checked, directly, in the primary documents, and it doesn't. That is a fact about process and accountability, not a modelled estimate or an inference — and it converts cleanly into a specific, answerable demand: publish it, or explain in public why not.

This also gives the campaign a concrete, low-risk FOI target that doesn't depend on any of the more contestable cost-escalation arguments being right. A request for "the current per-operator or system-average per diem rate, in the same format the City published from 2010-2012" is specific, has clear precedent (the City did exactly this before), and has a real chance of either producing a genuine transparency win (the data comes back) or a genuine accountability story (the City can't or won't produce it).


HOW TO USE THIS SAFELY


OPEN ITEM

Obtain the exact source of the $146/night 2025 figure — the specific council meeting, item number, and document URL/filename — from the campaign researcher, and independently verify it against that primary source directly. Until then, it remains a reported, not confirmed, figure (CL-564, marked ⚠️ VERIFY).


REF_002_Sources


Legal review not required for the core process finding (a documented absence of publication is a low-risk factual claim); recommended before public use of any specific dollar-figure comparison drawn from it.

Frozen v1 edition (July 2026) · published 2026-08-17 · corrections to the living library are welcome — tell us where we’re wrong.