Halifax Housing Starts Down 35%: What Landlords Should Know

The national headline from CMHC's latest survey of rental developers was blunt: 72% of them have paused or cancelled projects in the past two to three years because costs rose faster than the rents they could charge. The question for a Halifax owner is whether that is happening here. It is, and the evidence is CMHC's own count. Halifax housing starts are down 35% this year.

That is only half the picture, though, and the other half matters more to anyone renting out a unit in the next two years. Halifax has never had more homes under construction than it does right now.

Halifax starts fell 35%, from a record

Here is the Halifax CMA from CMHC's monthly construction count, January to September of each year:

Two things are true at once in that table.

New launches have dropped sharply. 2026's 3,448 starts are 35% below 2025 and 20% below 2024. They are still above any January-to-September before 2024, so this is a step down from an exceptional peak, not a return to the old normal. 2025 was the biggest year for Halifax starts in that series, at 7,000 for the full year; the previous high, in 1977, was 4,277.

The pipeline is the largest it has ever been. Units under construction reached 14,751 in July 2026, the highest in CMHC's monthly Halifax series, which runs back to 1972, and stood at 14,573 in September. CMHC's Fall 2026 Housing Supply Report puts the first-half figure at a record 14,400, against fewer than 5,900 in 2021. Completions are running at under 3,000 in nine months, so most of that pipeline is still to arrive.

And it is overwhelmingly rental. In the same report, CMHC says rentals account for 78% of all Halifax housing starts and 89% of all units under construction.

What developers told CMHC

The 2026 Rental Housing Development Survey was run for CMHC by EY from 28 May to 13 July 2026 and published on 29 September. It drew 110 responses from private developers, non-profits, fund managers and other rental builders across Canada. Asked what they had done in response to rising development costs over the past two to three years:

The sentiment split explains the pause. Only 37% of respondents were optimistic about the feasibility of rental development over the next five years, and 49% were negative. Over the longer term the picture reverses: 53% optimistic, 19% negative. EY's own reading is that the problem is project economics rather than demand. Developers still expect renters to be there; they doubt today's costs can be covered by today's rents.

The family-size gap is sharper still. Asked what limits them from building units with three or more bedrooms, 72% cited development costs that achievable rents do not cover, and 60% cited insufficient demand at the rents those units would need.

The Atlantic answers, and how far to trust them

The survey breaks results out by region, and Atlantic Canada is one of them. Hold these loosely. Only 8% of the 110 respondents said they operate in the Atlantic provinces, roughly nine firms, and EY states plainly that the results are "directional indicators" rather than a statistically representative sample. With that caveat, the Atlantic answers stand out in three ways:

• Atlantic developers are among the most optimistic in the country. Over the next five years, 44% were optimistic and 34% negative, the highest optimism of any region; in the big markets, around half were negative. Over the longer term, 67% were optimistic.
• Their top barrier is the same as everyone's, but the second is different. Government regulations, fees and approval delays ranked first in every region. Nationally, adverse market conditions came second; in Atlantic Canada, construction material costs ranked second, while interest rates ranked fifth and access to capital ninth.
• Their favourite strategy is renovating old buildings. Repositioning ageing assets ranked first in the Atlantic provinces, ahead of buying raw land. EY suggests that in eastern markets with older rental stock, upgrading an existing building can offer a better risk-adjusted return than building new.

The policy wish list is consistent: faster and more predictable approvals first, lower or deferred development charges second, then zoning certainty and upfront servicing. CMHC's Halifax analysis names the servicing problem directly: limits on water and wastewater capacity are "limiting project sizes" here, alongside financing and building capacity.

One caution on "approvals". When we timed HRM's own process, the median Halifax building permit took 27 days from submission to issue. The survey's approval complaint is broader than that final step: it covers planning, rezoning and servicing, which come before a building permit is ever applied for.

Permits held steady. Starts did not.

Our engine also reads HRM's building-permit records, and they tell a different story from the starts count. Net-new residential units on permits issued between 1 January and 23 October, leaving out any later withdrawn, cancelled or expired:

Issued permits are down about 4% on last year. Large-building permits are slightly up. Big applications are still arriving too, including a four-building, 1,174-unit proposal on Quinpool Road, submitted in May and still in review.

A permit is permission to build; a start is when construction actually begins. Permits holding steady while starts fall 35% is the gap you would expect if developers are getting approvals and then waiting, which is exactly what 72% of the survey's respondents said they had done. Our data cannot tell us which permitted Halifax projects are paused, so treat that as a consistent reading rather than a proven one. But it does mean the slowdown is not, so far, a collapse in what developers are asking HRM for.

What the survey does not say about rent control

Because this survey landed in the middle of a loud argument about Nova Scotia's rent cap, it is worth being exact about what it covers. The survey asks no question about rent control, rent caps or Nova Scotia's cap, and none of its findings mention them. Its closest item is a single combined barrier, "government regulations, government fee, and approval delays", which cannot be split into its parts. Anyone citing this survey as evidence about rent control is reading something into it that CMHC and EY did not put there.

What Halifax's own numbers can say is narrower. The 5% cap on increases for sitting tenants has applied since 1 January 2024, so the same cap was in force in 2024 and 2025, the two biggest years for Halifax starts on record. The cap did not change in 2026, and starts fell anyway. That does not prove the cap is harmless to investment, a question we tested in more depth in Does Nova Scotia's rent cap work?. It does mean the 2026 drop needs another explanation, and the survey offers one: costs, approvals and the rents new buildings can achieve. The cap also limits increases for sitting tenants only; it does not set the first rent on a newly built unit. On 23 October, Rental Housing Providers Nova Scotia, the landlords' association, wrote of the province "transitioning away from its temporary rent cap", which is scheduled to lapse after December 31, 2027.

What it means for a Halifax rental owner

Through 2027, expect more competition, mostly at the top. About 14,500 homes are under way and nearly nine in ten are rentals. As they finish, they add choice for tenants who can afford new construction. CMHC notes, nationally, that the easing in rental markets "is most evident in the newer, more expensive segments", and Halifax's last CMHC count put vacancy at 2.7% (October 2025), just below the 3.0% to 4.5% CMHC treats as balanced. If your unit is older, the competition is less about price than condition, and the Atlantic responses suggest developers here plan to compete in exactly that space, by renovating older buildings.

Price from current asking rents, not last year's. As of October 2026, the median Halifax asking rent is $1,825 for a one-bedroom, $2,237 for a two-bedroom and $2,500 for a three-bedroom. Rents already paid are still climbing: Statistics Canada's rent index for Halifax was 7.0% higher in September 2026 than a year earlier. Those are two different measures, and the gap between them is where pricing mistakes happen.

Three-bedroom units are the ones developers are least likely to build. In the survey, 72% said the cost of a unit with three or more bedrooms is not covered by the rent it can achieve. We have not measured the unit mix of Halifax's pipeline, so this is a national signal rather than a local count, but it points one way: if you own a three-bedroom house, townhouse or large apartment, new construction is less likely to be competing directly for your tenant than it is for a one- or two-bedroom.

After 2027, the pipeline thins if the slowdown holds. Fewer starts in 2026 means fewer buildings left to finish in 2028 and 2029. That would arrive at about the time the rent cap is due to lapse. Neither is certain: the cap could be extended, and starts could recover if costs or rates ease. They are two things to plan around rather than bet on.

The number that matters is the one for your own unit, and it moves with the competition on your street rather than the national survey. Get a free rental analysis and we will price it against current Halifax asking rents.

Sources: CMHC, monthly housing starts, completions and units under construction for the Halifax CMA, Statistics Canada table 34-10-0154-01, to September 2026. CMHC, Fall 2026 Housing Supply Report, Halifax section, published 10 September 2026. EY / CMHC, 2026 Rental Housing Development Study, survey conducted 28 May to 13 July 2026, published 29 September 2026; regional figures read from its published charts and tables. HRM building-permit open data as tracked by our market engine, net-new residential units on permits issued 1 January to 23 October of each year, excluding permits later withdrawn, cancelled or expired. Statistics Canada Consumer Price Index, rented accommodation, Halifax, table 18-10-0004-01, September 2026. Rental Housing Providers Nova Scotia, news post of 23 October 2026. Asking rents as of the date shown above. Informational estimates, not investment advice.