August Housing Starts Fell 3%: What That Actually Tells Us About Supply
August Housing Starts Fell 3%: What That Actually Tells Us About Supply
The Canada Mortgage and Housing Corporation tracks starts using a seasonally adjusted annual rate, which means August's figure of roughly 240,000 units represents what the year's total would be if every month matched August's pace. That number fell from July, continuing a pattern that has held through most of 2026: the trend line is flat or slightly negative, even as the Bank of Canada has cut rates three times since early 2025.
The disconnect matters because housing starts are supposed to respond to cheaper borrowing. When a developer can finance a project at 4.5% instead of 6%, the arithmetic of the pro forma shifts. More projects should clear the hurdle, yet August's data shows the starts did not move.
The lag runs through the development timeline
A housing start, in CMHC's definition, occurs when the foundation is poured or the structure's footprint is erected. What precedes that moment is 12 to 18 months of work: land assembly, rezoning applications, architectural drawings, pre-construction sales if the project is a condo, and then securing construction financing once the lender believes the project will sell. Rate cuts announced in early 2025 are only now filtering through to projects that were conceived in late 2024 under different assumptions.
Urban multi-unit projects, which account for the majority of Canadian starts, are especially slow to respond. A 200-unit apartment tower in Toronto requires roughly $80 million in financing, and lenders price that debt based on expected rental income two years from now, not current conditions. If the builder started planning in 2024 when rates were 6% and vacancy was tight, they are breaking ground in 2026 with financing priced at 4.5% but construction costs that have risen 9% since the plan was drawn. The net effect is that easing monetary policy has been absorbed by input inflation, leaving the developer's margin roughly where it was.
The multi-unit segment drove August's decline. Single-family starts, which respond faster because they require less capital and shorter permitting timelines, held steadier. The gap between the two suggests that the supply response is happening, but only at the smaller scale where a builder can move quickly.
The six-month trend removes the noise
CMHC publishes a six-month moving average precisely because monthly figures bounce. A single large-scale project in Vancouver that delays its start by two weeks can shift the national number. The trend measure smooths that out, and what it shows through August is stagnation: starts have hovered between 230,000 and 250,000 units for nine consecutive months.
Against the backdrop of CMHC's estimate that Canada needs 3.5 million additional units by 2030 to restore affordability, the current pace is running at roughly one-third of what would be required. Ontario and British Columbia, the two provinces with the highest housing pressure, also carry the highest development charges, the longest permitting backlogs, and the most restrictive zoning. A builder in Calgary can move from land acquisition to shovel-ready in 14 months. The same process in Toronto averages 26 months.
Labour is the other ceiling. Ontario does not have enough framers, electricians, and plumbers to build 400,000 units annually, even if financing were free and permits issued overnight. The shortage of skilled trades workers sits at the center of what can be built.
What the August figure actually measures
A start is not a completion. The gap between the two has widened in 2026 as developer insolvencies have risen. Projects that "started" in 2024 are sitting unfinished because the builder could not carry the debt through the higher-rate period. August's 3% decline does not account for those stalled sites, which means the effective supply coming to market is even thinner than the headline suggests.
The number tells us that the policy response, rate cuts, GST rebates on purpose-built rentals, federal funding through the Housing Accelerator, has kept starts from collapsing. It has not, however, created the conditions for a surge. That would require lower construction costs, faster permitting, or a much larger skilled trades workforce. None of those move on a central bank timeline.
August Housing Starts Fell 3%: What That Actually Tells Us About Supply
The Canada Mortgage and Housing Corporation tracks starts using a seasonally adjusted annual rate, which means August's figure of roughly 240,000 units represents what the year's total would be if every month matched August's pace. That number fell from July, continuing a pattern that has held through most of 2026: the trend line is flat or slightly negative, even as the Bank of Canada has cut rates three times since early 2025.
The disconnect matters because housing starts are supposed to respond to cheaper borrowing. When a developer can finance a project at 4.5% instead of 6%, the arithmetic of the pro forma shifts. More projects should clear the hurdle, yet August's data shows the starts did not move.
The lag runs through the development timeline
A housing start, in CMHC's definition, occurs when the foundation is poured or the structure's footprint is erected. What precedes that moment is 12 to 18 months of work: land assembly, rezoning applications, architectural drawings, pre-construction sales if the project is a condo, and then securing construction financing once the lender believes the project will sell. Rate cuts announced in early 2025 are only now filtering through to projects that were conceived in late 2024 under different assumptions.
Urban multi-unit projects, which account for the majority of Canadian starts, are especially slow to respond. A 200-unit apartment tower in Toronto requires roughly $80 million in financing, and lenders price that debt based on expected rental income two years from now, not current conditions. If the builder started planning in 2024 when rates were 6% and vacancy was tight, they are breaking ground in 2026 with financing priced at 4.5% but construction costs that have risen 9% since the plan was drawn. The net effect is that easing monetary policy has been absorbed by input inflation, leaving the developer's margin roughly where it was.
The multi-unit segment drove August's decline. Single-family starts, which respond faster because they require less capital and shorter permitting timelines, held steadier. The gap between the two suggests that the supply response is happening, but only at the smaller scale where a builder can move quickly.
The six-month trend removes the noise
CMHC publishes a six-month moving average precisely because monthly figures bounce. A single large-scale project in Vancouver that delays its start by two weeks can shift the national number. The trend measure smooths that out, and what it shows through August is stagnation: starts have hovered between 230,000 and 250,000 units for nine consecutive months.
Against the backdrop of CMHC's estimate that Canada needs 3.5 million additional units by 2030 to restore affordability, the current pace is running at roughly one-third of what would be required. Ontario and British Columbia, the two provinces with the highest housing pressure, also carry the highest development charges, the longest permitting backlogs, and the most restrictive zoning. A builder in Calgary can move from land acquisition to shovel-ready in 14 months. The same process in Toronto averages 26 months.
Labour is the other ceiling. Ontario does not have enough framers, electricians, and plumbers to build 400,000 units annually, even if financing were free and permits issued overnight. The shortage of skilled trades workers sits at the center of what can be built.
What the August figure actually measures
A start is not a completion. The gap between the two has widened in 2026 as developer insolvencies have risen. Projects that "started" in 2024 are sitting unfinished because the builder could not carry the debt through the higher-rate period. August's 3% decline does not account for those stalled sites, which means the effective supply coming to market is even thinner than the headline suggests.
The number tells us that the policy response, rate cuts, GST rebates on purpose-built rentals, federal funding through the Housing Accelerator, has kept starts from collapsing. It has not, however, created the conditions for a surge. That would require lower construction costs, faster permitting, or a much larger skilled trades workforce. None of those move on a central bank timeline.
Read Next
Ohio Lost $4.8 Billion to Trump's Tariffs in 90 Days
How Much Identity Theft Coverage Your Family Needs, Based on Who's Actually at Risk
CMHC's Latest Housing Supply Gap Report Shows Construction Slowing Faster Than Demand
Birch Hill builds a regional aviation network in British Columbia through Harbour Air's Pacific Coastal acquisition