Why aren't renters moving?
Canada's rental market has loosened: the national vacancy rate has doubled from its 2023 low and is back where it stood in 2020. Yet renters are staying put, and CMHC's own Rental Market Survey explains why. In the same building, the tenant who just signed a lease pays far more than the neighbour who stayed, and the gap has a different size in every city.
Two parts, as always: the story in plain language, then technical notes with definitions, the edition-stitching method, and CMHC's own reliability codes carried through every chart. Built on the same survey my Airbnb study uses as its rent benchmark.
Two rents for the same apartment, depending on who holds the lease
CMHC surveys every large rental building in the country each October and, since 2024, publishes a number that used to be invisible: how much more the units that changed tenants rent for, compared with the units in the same structure that did not.
How to read the first chart: each bar is that gap for two-bedroom apartments in October 2025. In Toronto a mover pays about a third more than a stayer in the same building; Halifax, Vancouver and Ottawa sit between one fifth and three tenths. In Calgary and Edmonton the gap is close to zero: prairie rents move with the market for everyone, not just for movers. Montréal and Winnipeg are shown as suppressed because CMHC withheld the estimate on sample-quality grounds, and Toronto's own 2025 figure carries the survey's "use with caution" code, both stated on the chart rather than dropped.
In the same building, the tenant who just moved in pays far more than the one who stayed
percent difference in two-bedroom rent between units that turned over in the past year and units that did not, same structure – dashed line = Canada, all centres 10,000+
CMHC Rental Market Survey, Canada data tables 2025, Table 6.2 · privately initiated apartments, three units and over
A big penalty freezes the market in place
If moving costs a third more, the rational move is not to move. The survey measures that too: the turnover rate, the share of apartments that changed tenants in the year before the survey.
How to read the dark chart: the same six markets, two bars each. Where the mover's premium is large, turnover has collapsed: fewer than one Toronto apartment in eleven changed hands in the 2025 survey year, against roughly one in four in Calgary and Edmonton, where moving costs almost nothing extra. The pattern matters for policy: a market can look calmer in its average rent while the price of actually moving, for a growing family or a new arrival, runs far ahead of it. Sitting tenants are protected; anyone who has to move pays the posted price.
Where moving costs the most, tenants have nearly stopped moving
same six markets, October 2025 – the mover's premium · turnover rate, the share of apartments that changed tenants that year
CMHC Rental Market Survey 2025, Tables 1.0 and 6.2 · turnover rate = units that changed tenants in the 12 months before the October survey
And the round trip that wasn't: the third chart is the national summary of the last six Octobers. Vacancy fell from 3.2% to 1.5% and climbed all the way back to 3.1%. Rent growth followed with a lag and only part way: 8.0% at the tightest point, still 5.1% in 2025, well above the roughly 3% pace of 2020 and 2021. The market loosened; the relief, so far, has mostly reached the people who move least.
Vacancy made a round trip; rent growth did not
Canada, October 2020 to October 2025 – vacancy rate · average two-bedroom rent growth, year over year
CMHC Rental Market Survey, Canada data tables 2021-2025, Table 1.0 · five editions stitched, each reporting two Octobers
The honest fine print
These are descriptive survey estimates, not a causal model: the premium and the turnover rate move together, and this page does not settle which drives which, or how much of each reflects rent control rules, tenant law, and market history that differ by province. The survey covers privately initiated apartment structures of three units and more, so basement suites, rented condos (surveyed separately), and single-family rentals sit outside these numbers. The premium series is young: CMHC first published Table 6.2 in the 2024 edition, so trend claims rest on two Octobers. And several estimates arrive with the survey's own quality flags, which this page shows instead of hiding: Toronto's 2025 premium is coded "use with caution", and Montréal and Winnipeg are suppressed outright.
Pre-flight catch: five separate annual workbooks had to be stitched into one panel, and the rent-level and rent-change columns share almost identical headers ("Average Rent Two Bedroom" and "Percentage Change of Average Rent Two Bedroom"). A naive header match silently reads growth rates as dollar levels; the parser matches the change column first and asserts the difference. Rents also arrive with thousands separators that break numeric parsing, a second quiet trap.
Technical terms, translated
- Turnover unit
- An apartment that changed tenants in the twelve months before the October survey. The premium compares its rent with non-turnover units in the same structure.
- Vacancy rate
- The share of rental apartments empty and available in October. Around 3% is the level economists usually call a balanced market.
- Turnover rate
- The share of apartments that changed tenants during the survey year: how much the market actually circulates.
- Reliability codes
- CMHC grades every estimate a through d and suppresses the worst. This page carries the codes through: a flagged number is shown flagged.
Technical notes
Sources, edition stitching, definitions, and reproducibility.
Data
CMHC Rental Market Survey, Canada data tables, editions 2021 through 2025 (each published in December, reporting the October survey), retrieved 2026-08-30 from CMHC's public data-tables site. Each edition's Table 1.0 reports vacancy rates, turnover rates, average two-bedroom rents, and year-over-year rent change for two Octobers; stitching the five editions yields a panel from October 2020 to October 2025, with the later edition's figure kept where editions overlap (revisions favoured). Table 6.2, first published in the 2024 edition, gives the percent difference in two-bedroom rent between turnover and non-turnover units within the same structure. The 2024 and 2025 editions use 2021 Census geographic definitions (per the source footnote); Ottawa means the Ontario part of the Ottawa-Gatineau CMA.
Definitions
Universe: privately initiated apartment structures of three units and over.
Turnover premium (CMHC footnote, quoted): the difference in rent "calculated
in the same structure between units that turned over and units that did not
turn over with respect to the survey reference year". Suppressed cells
(**) and changes flagged not statistically significant
(++) are read as missing, never as zero. Reliability codes
(a excellent, b very good, c good, d poor) are captured from the column
adjacent to each estimate and reported wherever a headline number is coded d.
Results table
Two-bedroom turnover premium, October 2025: Toronto +32.7% (code d; +43.6%
code b in October 2024), Halifax +28.7% (c), Vancouver +23.6% (b), Ottawa
+22.1% (c), Edmonton +3.6% (c), Calgary +3.0% (d); Canada +16.2% (b), down
from +19.9% in 2024. Turnover rates, 2025: Toronto 8.7%, Halifax 9.0%,
Montréal 9.4%, Vancouver 11.6%, Ottawa 15.4%, Winnipeg 22.2%, Calgary 24.9%,
Edmonton 28.8%. Canada vacancy by October: 3.2, 3.1, 1.9, 1.5, 2.2, 3.1;
Canada two-bedroom rent growth: 3.5, 3.0, 5.6, 8.0, 5.4, 5.1 (percent, 2020
to 2025). Full panel in panel.csv.
Threats to validity
Descriptive, not causal: premium and turnover are jointly determined, and provincial rent regulation is an obvious common cause this design cannot separate. The premium series has two observations. Suppression is not random (small or noisy samples are withheld), so the visible premiums are the better-measured markets. Average-rent changes mix composition effects (new, more expensive stock entering the universe) with same-unit increases; CMHC's Table 6.1 decomposition exists for readers who want that split.
Reproducibility
One script, run_analysis.py, parses the five workbooks from
data/ (header-anchored parsing with the change-versus-level
column guard described above) and writes every number on this page to
output/panel.csv and results.json, with the
suppression and reliability handling recorded in dq_notes.json.
Charts are generated from results.json.
References
- Canada Mortgage and Housing Corporation. (2021–2025). Rental Market Survey data tables, Canada [data sets, annual editions]. CMHC Housing Market Information. https://www.cmhc-schl.gc.ca/professionals/housing-markets-data-and-research/housing-data/data-tables/rental-market/rental-market-report-data-tables (retrieved August 30, 2026).
- Canada Mortgage and Housing Corporation. (2026). Rental Market Report, January 2026 edition. CMHC. (Context for the October 2025 survey results.)