Modular construction has never failed on engineering. It fails on order books. Canada is about to run the largest test yet of whether guaranteed demand is the missing ingredient, and the answer matters well beyond Canada.
Six federal sites across Canada are due to break ground in the second half of 2026. Together they represent up to 4,000 homes, and the procurement for them explicitly prioritises modular and factory built construction.
That number is not large. Canada needs vastly more than 4,000 homes, and nobody involved pretends otherwise.
What makes it worth watching is not the volume. It is that a government has decided the way to build a factory built housing industry is to guarantee it customers.
The problem this is actually trying to solve
Modular construction has a long history of technical success and commercial failure. The factories work. The tolerances are tighter than site built. The build times are faster. The safety outcomes are better.
And the companies keep going broke.
The reason is boringly consistent. A modular factory is a fixed cost. It needs continuous throughput to make sense, and continuous throughput requires a forward order book measured in years. Residential development does not produce forward order books measured in years. It produces lumpy, project by project demand tied to finance approvals and sales rates.
You cannot run a factory on a pipeline that arrives in unpredictable batches. Every modular business that has failed has failed at exactly this seam.
So the factory gets built on optimism, runs below capacity, cannot amortise the fixed cost, and closes. Then the next operator arrives, correctly identifies that the engineering is sound, and repeats the exercise.
This is the same demand certainty problem that has stalled modular everywhere it has been tried, from Britain to Australia to the United States. It is not a technology gap. It is a demand structure gap.
What Canada is doing about it
Build Canada Homes was launched in September 2025 as a federal agency to build and finance affordable housing at scale, with an initial $13 billion capitalisation. Its Investment Policy Framework explicitly prioritises modern methods of construction. The Build Canada Homes Act received Royal Assent in June 2026.
The specific mechanism is the interesting part. Rather than subsidising modular manufacturers directly, the agency is using its own procurement to create the forward order book the market has never been able to produce on its own.
In February 2026 it launched a Request for Information on modern methods of construction, aimed at volumetric modular builders, panelised system manufacturers and design build firms, with responses due 5 March. Those responses were intended to inform a prequalification process and identify high potential firms for future projects.
The stated purpose is worth reading precisely: to generate long term, predictable demand and provide certainty to manufacturers to scale up production, strengthen supply chains and create jobs.
That is a direct attempt to fix the seam. Not better factories. Guaranteed customers.
What is actually being built
The six Direct Build sites sit on Canada Lands Company land, which removes land acquisition and much of the approvals friction in one move. Through Canada Lands, the agency has access to 88 federal properties spanning around 463 hectares, with portfolio capacity cited at up to 45,000 units. The initial six are in Dartmouth, Longueuil, Ottawa, Toronto, Winnipeg and Edmonton.
Requests for Qualifications have been posted for five of the six: Dartmouth, Longueuil, Ottawa, Winnipeg and Edmonton. Toronto is selected but has not yet been listed with a posted RFQ, which is a small detail worth noting given Toronto carries the largest associated infrastructure commitment.
On the announced numbers, Arbo in Toronto is set to deliver 540 homes on part of the Downsview site, with at least 40 per cent affordable, alongside a $283 million infrastructure investment intended to catalyse up to 63,000 homes in the broader area. Pointe de Longueuil is around 1,055 homes on the St Lawrence waterfront, 40 per cent non market. Heron in Ottawa is around 1,100 homes on an 18 acre site. Shannon Park in Dartmouth is up to 630 mixed market homes in its first phase. Naawi Oodena in Winnipeg is a partnership with Treaty One First Nations and Canada Lands Company.
Beyond the direct builds, the agency has partnered with British Columbia, securing $170 million in federal capital that unlocks up to $200 million in provincial capital, delivering more than 700 supportive and transitional homes, with a further 400 through BC’s Digitally Accelerated Standardized Housing program. Over 700 units are planned in Nunavut with the Nunavut Housing Corporation, roughly 30 per cent built off site. A $1.5 billion Canada Rental Protection Fund will preserve at risk affordable rentals, $1 billion goes to transitional and supportive housing, and around $1.7 billion is directed to Indigenous housing in urban, rural and northern communities.
Federal land, standardised design, off site manufacture, and a single agency coordinating the pipeline. Every friction point modular has historically died on, removed by design. The agency’s own claim is that these methods can halve construction timelines and deliver around 20 per cent savings.
The case against
This is where it gets genuinely contested, and the criticism deserves a fair hearing.
The Parliamentary Budget Officer projected that Build Canada Homes would add approximately 26,000 units to Canadian housing supply over five years, about 13,000 of them affordable for low income households. Interim PBO Jason Jacques put that at a 2.1 per cent increase in housing completions against the office’s baseline projection, addressing 3.7 per cent of an estimated 690,000 unit housing gap by 2035.
Against a shortfall that size, 26,000 units is a rounding error, and the PBO said so in the plainest terms available to a budget office: the contribution will likely be modest.
There is a second finding in that report that got less attention and matters more. Total federal planned spending on housing programs is forecast to fall 56 per cent, from $9.8 billion in 2025-26 to $4.3 billion in 2028-29, as existing programs end and Budget 2025 cuts take effect. The PBO noted the 13,000 affordable units would only partially offset the decline in overall affordability support.
So the agency is not simply additional. Some of what it delivers replaces support being withdrawn elsewhere.
Researchers Austin Thompson and Jake Fuss at the Fraser Institute have argued the mandate is internally contradictory: that Ottawa cannot credibly promise cost effective affordable housing while simultaneously steering builders toward its own preferences on materials, sustainability targets and construction methods. Their sharper point is about crowding out. The agency has a large enough budget to bid workers, materials and equipment away from private homebuilders, even where the projects it backs deliver worse value per dollar.
They also argue that headline figures have been announced without timelines or benchmarks against which anyone could assess whether the agency is generating additional affordable housing at reasonable cost.
That last criticism is the one to take most seriously, and not because of Canadian politics. If the whole premise is that guaranteed demand builds an industry, then the industry needs to know the demand is real, sized, and scheduled. An unbenchmarked promise is not an order book. It is a press release, and modular manufacturers have built factories on those before.
There is also the workforce question sitting underneath everything. Canada’s broader housing push aims to double homebuilding, and industry leaders have openly questioned whether the country has enough workers to deliver it. Factory built housing changes where the labour is needed. It does not conjure labour.
What Australian builders should take from this
Australia has been circling modular for years. We have had the technology conversation repeatedly. We have not had the demand conversation.
Canada is running the experiment we have been arguing about, at scale, with public money, and with a real chance of visible failure. That makes it valuable regardless of the outcome.
If it works, the lesson is that modular does not need better engineering or more enthusiasm. It needs a customer willing to commit to volume across years, and government is the only actor structurally capable of being that customer.
If it fails, the failure mode will be instructive too, and it will probably be one of three things: the order book turns out to be smaller or later than promised, the crowding out effect bids resources away from private builders faster than the agency adds supply, or the workforce simply is not there.
Either way, the question of whether factory built housing can hold quality at volume gets tested in public rather than in a pitch deck.
The immediate practical relevance for an Australian builder is narrower. Watch the procurement structure, not the politics. If Canada demonstrates that prequalified panels of suppliers with guaranteed multi year volume produce lower costs and faster delivery, that structure is portable and the argument here changes overnight.
And it is worth being honest about the domestic parallel. We have started building industrial capacity again on the strength of long horizon commitments rather than spot demand. The logic is identical. Guaranteed volume is what makes fixed cost manufacturing rational.
The thing worth watching
Construction on the Direct Build sites was expected to begin in the second half of 2026. The first honest data point will be whether it did, and on what schedule.
Not the announcements. The slabs.
The Good Builder Take
Modular has never had an engineering problem. It has had an order book problem, and no amount of enthusiasm has ever fixed it.
What Canada is testing is the only intervention that plausibly could: a customer big enough and patient enough to guarantee a factory continuous work. That is a genuinely interesting idea. The Fraser Institute’s crowding out objection is a genuinely interesting rebuttal. Both can be right.
The number that should give everyone pause is not the 26,000 units. It is the 56 per cent fall in total federal housing spending sitting behind it. An agency that adds 26,000 units while $5.5 billion a year exits the system elsewhere is not a housing program. It is a restructure with a ribbon on it.
For Australian builders, the useful move is to stop watching the modular technology and start watching the procurement. If guaranteed volume works in Canada, the case for it here stops being theoretical. If it does not, we will have learned something expensive without paying for it.
Frequently asked questions
Build Canada Homes is a Canadian federal agency launched in September 2025 to build and finance affordable housing at scale, with an initial $13 billion capitalisation. It combines financial tools, access to federal land through Canada Lands Company, and development expertise, and its Investment Policy Framework prioritises modern methods of construction such as modular and factory built housing. The Build Canada Homes Act received Royal Assent in June 2026.
Modern methods of construction, or MMC, refers to building approaches that shift work from the site into a controlled factory environment. It includes volumetric modular construction, where complete three dimensional units are manufactured off site, and panelised systems, where wall, floor and roof elements are made in a factory and assembled on site.
Six Direct Build sites on federal land represent up to 4,000 units, with construction expected to begin in the second half of 2026, and portfolio capacity across 88 federal properties cited at up to 45,000 units. The Parliamentary Budget Officer projects the agency will add approximately 26,000 units over five years, about 13,000 of them affordable for low income households, representing a 2.1 per cent increase in housing completions and addressing 3.7 per cent of Canada’s projected 690,000 unit housing gap by 2035.
Modular factories carry high fixed costs and require continuous throughput to be viable, which means they need a forward order book measured in years. Residential development typically generates lumpy, project by project demand tied to finance approvals and sales rates. The mismatch between fixed factory capacity and irregular demand, rather than any engineering limitation, is the most common cause of modular business failure.
Australia has no single federal agency operating as a direct housing developer on the Build Canada Homes model. Australian modern methods of construction activity is driven by state level initiatives and private sector investment rather than a centrally coordinated federal procurement pipeline with a guaranteed multi year order book.
RELATED ARTICLES
Canada’s $25 Billion Bet: Can a Government Housing Agency Actually Build Homes?
Robots on the Tools, Homes in the Factory: The Construction Revolution the World Is Already Running
Steel Is Coming Back to Newcastle, and This Time There Is No Gas in the Furnace
Hear more conversations with builders working through exactly these conditions on The Good Builder Podcast, available on Spotify and Apple Podcasts.
Last updated: 17 July 2026. Sourced from Housing, Infrastructure and Communities Canada, the Parliamentary Budget Officer (RP-2526-020-S), the Government of British Columbia and the Fraser Institute.
This article is intended for general information purposes only and does not constitute legal, financial, or professional advice. Laws, regulations, and industry requirements vary by state and territory and change over time. Builders and trades professionals should seek independent advice relevant to their specific circumstances before making business, legal, or financial decisions.
0 Comments