CMHC Puts Canada's Affordability Target at 417,000 Starts as Toronto Needs 50% More Homes Built

Toronto is the clearest case of a narrowing headline gap concealing a widening tenure problem. In the first half of 2026, CMHC's detailed market tables show the City of Toronto recorded just 156 condominium-unit starts, against an annual average of roughly 7,000 over the preceding decade. Ground-oriented freehold starts sat at record lows. Meanwhile, purpose-built rental starts across the Toronto census metropolitan area rose 82% year over year and exceeded condominium starts for the first time since 1994.

That is not a modest shift in the mix of otherwise healthy construction. It is the ownership channel going quiet while the rental channel runs hot, which is precisely why CMHC describes Toronto's remaining supply gap as increasingly concentrated in the ownership market even as the total gap narrowed.

Vancouver runs the same pattern with a sharper edge. Condominium apartments are the largest source of new homeownership supply in that market, and their construction has fallen to its lowest level in more than a decade. Purpose-built rentals now account for about 60% of housing starts there, compared with under 20% a decade ago. Condominium-apartment starts fell another 40% in the first half of 2026 after an already weak 2025, making it the weakest first half since 2011. CMHC's own conditional language on what follows is worth quoting in substance: if low starts produce fewer future completions, buyers will likely face more competition for limited ownership housing and renewed upward price pressure.

The mechanics behind the slowdown are not a mystery, and they are partly financial rather than physical. Presale thresholds, construction costs and completed unsold inventory are all constraining condominium development right now — pressures that have already prompted calls for the banks to lower the 70% condo presale threshold that governs when a project can secure financing at all.

The three remaining markets in the report reach ownership pressure by different mechanisms, which is a useful corrective to reading every city through the Toronto lens.

Montréal's affordability has deteriorated significantly since 2019, according to the September release. Rental construction there has been substantial in recent years, but homeownership starts are near multi-year lows — and Montréal now carries the largest absolute supply gap of any market in the report.

Ottawa's problem is composition, in an unusually stark form. Rental units represented 54% of starts, 64% of completions and 69% of units under construction in 2026, while condominium apartments made up only 12% of starts. Year over year, rental starts rose 3% while condominium starts fell 49%. Ottawa's pending-start inventory is large, but CMHC notes some approved projects may be delayed while developers wait for better financing, costs or market certainty — so the size of a pipeline says little about when, or whether, it converts.

Halifax arrives at ownership pressure from the demand side instead. Record construction in recent years has eased rental conditions, but stronger population growth has caused demand to outpace new supply, particularly in the ownership market. The pattern of affordability strain reaching well past the two markets that usually define the conversation is now well documented — CMHC's own housing affordability composite index showed the crisis spreading beyond Toronto and Vancouver earlier this year.

Any honest version of this analysis has to say clearly where it stops working, and it stops working in Alberta.

Calgary's total starts fell by more than 20% in the first half of 2026 after three consecutive record years, which sounds like the same story as everywhere else. It is not. Ownership demand there stayed comparatively resilient: condominium-apartment starts rose 4% and completed unsold condominium inventory fell 59%. Rental starts fell more than 30% but still represented almost 60% of Calgary starts, so the ownership counterpoint is relative rather than a claim of unlimited supply. CMHC estimates Calgary needs only 4,000 to 5,000 additional units annually to restore affordability by 2036 — a small fraction of the Toronto or Montréal requirement. Buyers there face a real near-term challenge from sharply lower for-sale inventories, but new construction is responding to demand rather than retreating from it.

Edmonton is further still from the national narrative. Condominium-apartment starts increased 14% year over year in the first half of 2026 while completed unsold condominium inventory fell by almost half. Condominiums represented nearly one-quarter of apartment starts, up from about 3% in 2023, and units under construction reached a record 17,900. Developers continued launching ownership-oriented projects on the strength of steady demand and the region's relative affordability. Edmonton is the only large market in the report with no measured supply gap — though CMHC still identifies affordability difficulties for the city's lowest-income households, which a market-wide supply measure does not capture.

The tenure split is not incidental to the affordability target — it is most of it. CMHC's 2025 modelling framework estimated that the extra starts needed beyond business as usual would be predominantly ownership housing: 83.7% in Toronto, 77.8% in Vancouver, 71.7% in Ottawa–Gatineau and 61.8% in Montréal. Nationally, that framework allocated 74.9% of the additional starts to ownership housing, 21.3% to primary-market rentals and 3.7% to secondary-market rentals.

Those shares come from a June 2025 exercise with a 2035 horizon rather than from the updated fall 2026 estimates, and scenario models are simplifications, not forecasts. But the direction is unambiguous and it lines up with everything else in the report: maintaining strong purpose-built rental output, on its own, does not close the modelled gap in the markets where the gap is largest. Roughly three-quarters of the incremental need is ownership housing, and ownership housing is the category currently being started least.

This is also why supply-side and demand-side policy pull against each other when the pipeline is thin. CMHC has previously warned that government homebuyer incentives could push prices higher without roughly 28,000 additional annual starts to absorb the demand they create — a caution that lands harder when ownership construction is at the levels this report describes.

Start with what this report is not. It is not a price forecast, it does not model resale values, and nothing in it constitutes a reason to buy, sell or refinance. CMHC's framing throughout is conditional — a likely crunch, a probable increase in competition, a risk of underbuilding — and that conditional register is the honest one to carry into any personal decision.

What the report does supply is a piece of context that is usually missing from housing coverage: the composition of what is being built where you live, and the multi-year lag before today's construction decisions show up as finished homes. CMHC's summer 2026 outlook projected national starts falling from 259,028 in 2025 to 241,400 in 2026, 223,400 in 2027 and 211,900 in 2028 under its baseline, attributing the weakness to soft demand, high costs and elevated inventories while expecting demand to recover gradually as economic and income growth improve. Set that baseline forecast against a required 417,000 and the gap is not closing on the current trajectory — it is widening before it narrows.

For an owner weighing a move-up, a renovation refinance, or a hold-versus-sell question, the practical translation is narrow but real. The national number is the wrong input. The composition of starts in your own census metropolitan area is the right one, and CMHC breaks it out for seven of them. In Toronto, Vancouver, Montréal and Ottawa, the ownership pool is being added to slowly while the rental pool grows quickly. In Calgary and Edmonton, it is not. Those are genuinely different situations and they deserve genuinely different conclusions.

It also helps to know that many households have already reached their own conclusion, which is to do nothing — a CPA Canada survey found 55% of homeowners planning to stay put indefinitely as the market locks up. Low turnover in the existing stock and weak additions to the ownership pipeline are two halves of the same tight-inventory picture, and the next Housing Supply Report is where to look for evidence that either half has started to move.