And just like that, Canada’s shrinking population went the way of its technical recession—revised away. Statistics Canada (StatCan) data shows that the country’s population grew in Q3 2026. More importantly, the data includes the upward revisions we’ve been discussing for months. It turns out that the much-discussed population contraction plaguing the country didn’t exist. The revisions reveal a bigger population, but more importantly, a much worse economy.
Canada’s “Shrinking” Population Has Been Growing This Whole Time
Canadian quarterly population estimates: Reported vs latest revisions.
Source: StatCan; Better Dwelling.
Canada’s population climbed 0.2% (+80.3k people) to 41.8 million in Q3 2026, rising 0.5% (+189.4k) since last year. Annual growth is the weakest since Q2 2021, roughly the end of the pandemic’s lockdowns. The strength is likely to surprise most people, as the country was supposed to be shrinking. Our first hint that this wasn’t true should have been policymakers promising to do just that.
Canada’s Temporary Resident Estimates: Under On Decline, Inflated At Peak
StatCan made sharp revisions to its population estimates, as we previously warned. They originally reported a 0.5% y/y drop for Q2 2026, marking a rare annual decline. This morning’s update shows they missed roughly 300,000 people. Whoops!
The entire revision is due to methodology changes for estimating non-permanent residents (NPRs). There were 2.78 million NPRs in Q3 2026, down 0.7% (-18.9k) from last quarter and 5.3% (-154.6k) below last year’s estimates. However, Q2 2026 also saw an upward revision of 2.8 million, 9.3% (+239k people) higher than originally reported. The NPR peak in Q4 2024 was also revised lower from 3.15 million to 2.98 million, 5.4% (-170k) lower than reported. In short, even when policymakers launch half-baked plans, the numbers show stability. Cute.
Since these residents are in the country, it doesn’t change reality for consumption. Even if a bureaucrat forgets to count them, they still consume food and shelter. Demand-side pressures are the same. Policymakers may be delighted, as more human capital stock means higher credit capacity, however, reality doesn’t change, but the country’s economic narrative will.
Canada’s Upward Revisions Reveal A Much Worse Economy
Canada’s economy was doing really well for a country with a shrinking population. Now that it’s not shrinking, the economic reality is a disaster, validating how many feel. GDP growing 0.8% in Q2 2026 at the seasonally adjusted annual rate is impressive with a 0.5% population drop. Not so much if the population is growing 0.5% over that same period.
Breaking down that growth reveals how problematic those GDP numbers really were. Exports of goods represented 1.23 percentage points of that 0.8% growth, due mostly to higher oil prices. That means Iran’s closure of the Strait of Hormuz did more for Canada’s GDP than its policymakers. A war between two other countries was a bigger boost than near-pandemic levels of new debt. The math is even more disturbing when it comes to the job market.
Canada’s employment data relies heavily on population estimates. The job data would have collected the number of NPRs with a job, but the population estimates wouldn’t have. This inflates the number of workers while underestimating the total number of people. We previously estimated that the undercount means the 6.7% unemployment rate is closer to 7.6%. That would make it the third-highest report since the pandemic, and it was a conservative call. We used the national average for the share in the labour force, but it’s typically much higher for NPRs.
StatCan confirmed it will revise the employment data in the coming months. However, the methodology means the revisions will be slowly introduced. Funny how reality can’t hit all at once when it impacts national optics negatively, eh?
The population revisions are surprising for a few reasons, and not just the upward ones. Upward revisions reveal a fragile economy, masked by ignoring hundreds of thousands of workers. They also make slow home sales and a sharp climb in rental vacancies even more troubling.
Downward revisions at the peak reveal the demand-side narrative was softer than presented. But what’s really surprising here is that they appeared at all. Our previous estimate shows StatCan overestimated the population for ~30 years. An agency study followed and found the overcount was even larger than our estimate. It appears revisions that expand the country’s borrowing capacity make it into official data faster than those that can slow it.
