StatCan's Population Revisions Could Erase Canada's Decline and Rewrite Housing Assumptions
Benjamin Tal's team at CIBC Capital Markets expects Statistics Canada to add hundreds of thousands of people back into the national population count when the agency releases its next major revision. The revision matters because nearly every major policy calculation in Canada, GDP per capita, housing targets, healthcare capacity, infrastructure spending, depends on knowing how many people actually live here. Right now, we don't.
The "decline" observed in late 2025 and early 2026 was never a mass exodus. It was a data problem. Administrative systems that track non-permanent residents, international students, temporary foreign workers, people on expired work permits who remained legally under "implied status" while renewals processed, flagged departures that never happened. The people stayed. The count dropped.
Why the numbers diverged
Canada's population tracking relies on a patchwork of data sources: census surveys, tax filings, provincial health card issuance, border records, and work permit databases. These systems do not speak to each other in real time. When someone's work permit expires on paper but they remain in the country legally while Immigration, Refugees and Citizenship Canada processes their renewal, older tracking models treat them as departed. The individual keeps paying rent, filing taxes, using transit. The official count drops by one.
This gap widened sharply after 2024, when the federal government introduced new caps on non-permanent resident intake. The policy change introduced confusion into the tracking layer. People who would have been counted under the old system were suddenly invisible in administrative records, even as they continued to occupy housing units and participate in the labor market.
The tell was in the economic data. GDP growth remained positive. Labor force participation held steady. Housing vacancy rates stayed tight in major metros. The rental market in Toronto and Vancouver never softened the way a genuine population decline would produce. Investors betting on weaker demand for apartments found the opposite: supply constraints persisted, rents climbed, and the expected relief never arrived.
The denominator problem
Population is not just a count. It is the denominator in nearly every calculation of national well-being. GDP per capita, the most cited measure of living standards, divides total economic output by the number of people sharing it. If the population estimate was too low, then Canada's GDP per capita in 2025 and early 2026 was overstated. The real number, adjusted for the uncounted residents, is worse.
This creates a peculiar inversion. The revision will make the population number go up. It will make the standard-of-living number go down. A country with more people than it thought, producing the same output, is poorer per person than the data suggested.
CIBC's analysis suggests the economic drag from undercounting could be in the range of $7,000 to $10,000 per capita. That figure is speculative, but the direction is certain. The revision does not add wealth. It redistributes the existing economic pie across a larger headcount.
What this does to housing targets
The Canada Mortgage and Housing Corporation bases its supply projections on population growth forecasts. If the population was undercounted by several hundred thousand people, then the housing deficit is larger than the 3.5 million units CMHC currently projects as necessary by 2030. Every unit of missing supply tightens the rental market further, pushes prices higher, and extends the timeline for affordability stabilization.
StatCan's shift toward real-time administrative data, tax returns, health card records, permit renewals processed through IRCC, should close the gap going forward. But the lag still exists. Administrative data from 2026 reflects conditions from late 2024. The revisions arriving now are not current. They are a look-back.
For developers and lenders, this means the market never softened as much as the population data implied. Demand remained higher than official estimates suggested. Projects greenlighted based on conservative population assumptions may find themselves undersupplied. Conversely, any investor who paused acquisitions in 2025 expecting weaker fundamentals missed the actual trajectory.
The Bank of Canada watches population growth as a dual signal: labor supply (disinflationary, because more workers ease wage pressure) and housing demand (inflationary, because more people compete for fixed supply). A sudden upward revision complicates the rate-setting calculus. If the population was larger than thought, then demand-side inflation in housing has more structural support than the central bank modeled. That argues for higher rates, or at least slower cuts, even as other inflation components cool.
The invisible tenant
The rental market is where the revision lands hardest. A landlord in Mississauga with a 400-unit building does not care about the national population estimate. They care about vacancy rates, turnover, and demand at the current rent level. If the official count said population declined by 0.8% in a given quarter but vacancies stayed near zero and rents climbed, the landlord had better information than StatCan.
That informational advantage is now being corrected. The people who were always there are being counted. But the correction arrives after policy decisions have been made. Immigration caps were set partly in response to perceived softening in housing demand. If that softening was a mirage, the caps may have been tighter than conditions justified.
Natural population growth, births minus deaths, is at historic lows in Canada. The country adds people almost entirely through migration. A revision that adds hundreds of thousands of non-permanent residents back into the count does not change that dependency. It just makes it more visible. Every major demographic projection for the next two decades assumes continued net inflows in the range of 400,000 to 500,000 people per year. If the tracking systems cannot accurately count who is already here, they cannot reliably forecast who will be here.
The revision will not resolve the housing crisis. It will clarify the scale. The shortage was always worse than the official numbers admitted. The people were always here, competing for the same units, bidding up the same rents. The only thing that changes is the acknowledgment.
Benjamin Tal's team at CIBC Capital Markets expects Statistics Canada to add hundreds of thousands of people back into the national population count when the agency releases its next major revision. The revision matters because nearly every major policy calculation in Canada, GDP per capita, housing targets, healthcare capacity, infrastructure spending, depends on knowing how many people actually live here. Right now, we don't.
The "decline" observed in late 2025 and early 2026 was never a mass exodus. It was a data problem. Administrative systems that track non-permanent residents, international students, temporary foreign workers, people on expired work permits who remained legally under "implied status" while renewals processed, flagged departures that never happened. The people stayed. The count dropped.
Why the numbers diverged
Canada's population tracking relies on a patchwork of data sources: census surveys, tax filings, provincial health card issuance, border records, and work permit databases. These systems do not speak to each other in real time. When someone's work permit expires on paper but they remain in the country legally while Immigration, Refugees and Citizenship Canada processes their renewal, older tracking models treat them as departed. The individual keeps paying rent, filing taxes, using transit. The official count drops by one.
This gap widened sharply after 2024, when the federal government introduced new caps on non-permanent resident intake. The policy change introduced confusion into the tracking layer. People who would have been counted under the old system were suddenly invisible in administrative records, even as they continued to occupy housing units and participate in the labor market.
The tell was in the economic data. GDP growth remained positive. Labor force participation held steady. Housing vacancy rates stayed tight in major metros. The rental market in Toronto and Vancouver never softened the way a genuine population decline would produce. Investors betting on weaker demand for apartments found the opposite: supply constraints persisted, rents climbed, and the expected relief never arrived.
The denominator problem
Population is not just a count. It is the denominator in nearly every calculation of national well-being. GDP per capita, the most cited measure of living standards, divides total economic output by the number of people sharing it. If the population estimate was too low, then Canada's GDP per capita in 2025 and early 2026 was overstated. The real number, adjusted for the uncounted residents, is worse.
This creates a peculiar inversion. The revision will make the population number go up. It will make the standard-of-living number go down. A country with more people than it thought, producing the same output, is poorer per person than the data suggested.
CIBC's analysis suggests the economic drag from undercounting could be in the range of $7,000 to $10,000 per capita. That figure is speculative, but the direction is certain. The revision does not add wealth. It redistributes the existing economic pie across a larger headcount.
What this does to housing targets
The Canada Mortgage and Housing Corporation bases its supply projections on population growth forecasts. If the population was undercounted by several hundred thousand people, then the housing deficit is larger than the 3.5 million units CMHC currently projects as necessary by 2030. Every unit of missing supply tightens the rental market further, pushes prices higher, and extends the timeline for affordability stabilization.
StatCan's shift toward real-time administrative data, tax returns, health card records, permit renewals processed through IRCC, should close the gap going forward. But the lag still exists. Administrative data from 2026 reflects conditions from late 2024. The revisions arriving now are not current. They are a look-back.
For developers and lenders, this means the market never softened as much as the population data implied. Demand remained higher than official estimates suggested. Projects greenlighted based on conservative population assumptions may find themselves undersupplied. Conversely, any investor who paused acquisitions in 2025 expecting weaker fundamentals missed the actual trajectory.
The Bank of Canada watches population growth as a dual signal: labor supply (disinflationary, because more workers ease wage pressure) and housing demand (inflationary, because more people compete for fixed supply). A sudden upward revision complicates the rate-setting calculus. If the population was larger than thought, then demand-side inflation in housing has more structural support than the central bank modeled. That argues for higher rates, or at least slower cuts, even as other inflation components cool.
The invisible tenant
The rental market is where the revision lands hardest. A landlord in Mississauga with a 400-unit building does not care about the national population estimate. They care about vacancy rates, turnover, and demand at the current rent level. If the official count said population declined by 0.8% in a given quarter but vacancies stayed near zero and rents climbed, the landlord had better information than StatCan.
That informational advantage is now being corrected. The people who were always there are being counted. But the correction arrives after policy decisions have been made. Immigration caps were set partly in response to perceived softening in housing demand. If that softening was a mirage, the caps may have been tighter than conditions justified.
Natural population growth, births minus deaths, is at historic lows in Canada. The country adds people almost entirely through migration. A revision that adds hundreds of thousands of non-permanent residents back into the count does not change that dependency. It just makes it more visible. Every major demographic projection for the next two decades assumes continued net inflows in the range of 400,000 to 500,000 people per year. If the tracking systems cannot accurately count who is already here, they cannot reliably forecast who will be here.
The revision will not resolve the housing crisis. It will clarify the scale. The shortage was always worse than the official numbers admitted. The people were always here, competing for the same units, bidding up the same rents. The only thing that changes is the acknowledgment.
Read Next
MCAN's 19% earnings jump proves mortgage impairments aren't the risk signal investors think they are
Toronto Buyers Can Stop Writing Love Letters, The Market Already Did the Work
25 States Sue Trump Over Tariffs, Claiming Presidential Overreach on Import Taxes
Carney's Alberta housing pitch meets canola fields and separatist flags