From Hype to Hard Numbers: The True Scale of Ottawa’s New Housing Push

Hawkeye Wealth Ltd. • May 31, 2025
SUBSCRIBE TO OUR NEWSLETTER

Introduction


The Liberal Government is in and we are starting to get more clarity on what that means for housing in Canada.


In our last article, we compared the
Liberal vs. Conservative Housing Platforms, and discussed how the majority of the Liberal housing platform would be positive for housing investors, but that the Build Canada Homes program had the potential to negatively overshadow everything else.


One month later, our opinion has softened. The limited documentation available about Build Canada Homes indicates that the government will be (directly) building far fewer homes than we initially anticipated, which has materially lowered our level of concern.  Build Canada Homes looks to be far too small to displace private builders or upset private markets.


In this edition of the Bird’s Eye View, we review publicly available information on the Build Canada Homes program to determine its scale and potential impact.  We then turn to the secondary question of how successful that program is likely to be as we review two of the models that the Liberals have used as inspiration for Build Canada Homes; the Wartime Homes Limited program that saw the Federal Government get directly involved in homebuilding post-WWII, as well as the Singaporean Public Housing model.


Build Canada Homes


“The Liberal housing plan will double Canada’s current rate of residential construction over the next decade to reach 500,000 homes per year”.

Liberal Housing Plan, March 31, 2025


We begin as we so often do with a caveat. It’s important to recognize that there is uncertainty about what this program will look like, as the entire housing plan (at least what is publicly available) is a mere two-page document. The truth is that we really don’t know what this program will look like, even if we know its goals and now have cost estimates.


Canada built ~245,000 homes in 2024, which is near the all-time high for annual construction (257,453 units were built in 1974). Getting to 500,000 units by 2036 feels like it borders on impossible, and is potentially much higher than what’s necessary.


When we saw the 500,000 homes per year target, alongside the words “deeply affordable,” and the announcement that “the Federal government will get back in the business of building homes”, we saw a very real potential for the heavy disincentivization of private development. If the government is going to compete with private industry while subsidizing costs, why would private industry build anything? Why would private investment fund it?


On further review, those concerns are now much smaller than we initially feared. Since we won’t see the 2025 federal budget
until the fall, we are limited to the Liberal Housing Plan as well as the Liberal Fiscal and Costing Plan to get a sense for the program itself and how much money the Feds will be allocating to it, but those documents indicate that funding allocations will be small.

Here are some of the housing highlights from the Liberal Fiscal and Costing Plan:


Source: Liberal Fiscal and Costing Plan, Chart recreated by Hawkeye Wealth for legibility.


The full cost of housing initiatives, including GST cuts, DCC reductions, the new MURB tax incentive, and the Build Canada Homes program, is expected to fall between $4.9B  and $6.8B in spending per year, while the Build Canada Homes portion is expected to cost ~$2.85B to $3.0B annually.  With the 2024 budget at ~$449B, the Build Canada Homes program is smaller than we anticipated at 0.65% of the annual budget.


At the proposed funding levels, the Build Canada Homes program simply doesn’t have the funding to build an appreciable number of homes and displace private investment, which we see as a positive. At $250,000 per unit, the Feds could build ~12,000 (small) units per year with this funding.  It’s not nothing, but it’s also only 2.4% of the 500,000 target, or ~5% of 2024 building levels.


Unless funding levels increase, we would be surprised to see more than 60,000 homes directly constructed by the government under the Build Canada Homes program over the next four years.


With the question of the size of the program answered, let’s look at whether Build Canada Homes is likely to be successful by reviewing the Wartime Homes Limited program, as well as the Singapore public housing model.


Wartime Housing Limited


“Mark Carney’s Liberals unveil Canada’s most ambitious housing plan since the Second World War.”


“Canada has solved a housing crisis before, and we can do it again”.


Much of the rhetoric around the Build Canada Homes plan is based on the Wartime Housing Limited (WHL) program, which has been credited for pulling Canada out of the 1940s housing crisis.


What might surprise you is that for all the hype, WHL only oversaw the development of 31,192 homes between 1941-1947.  This represents approximately 8.2% of the total number of homes constructed during that time.


By most accounts, WHL was highly successful in achieving its mandate.
Jill Wade’s 1984 thesis paper details and evaluates the WHL program from when it began in 1941 to when it wrapped up in 1947, and we’ve provided highlights below:



  • In the early 1940s, Canada experienced a housing shortage that stemmed from a lack of new supply built during the Great Depression, followed by labour shortages during wartime.


  • The intent of the program was to build ‘temporary housing’ at the lowest possible cost and rent the homes to veterans.  Homes were only intended to last 10 years, even though many of the units lasted far longer, and some still stand today.
  • Panels were pre-fabricated, and workers needed ~16 hours (not a typo) to put up a small house. Neighbourhoods of 20 homes could go up in 30 days.
  • The intent behind building ‘temporary’ homes was to keep permanent housing the purview of private development, so as not to disincentivize it.

Source: City of Kelowna, WHL Era Home (Built in 1946) - Kelowna, BC

  • These units weren’t ever intended to be ‘subsidized housing’.  Leonard Marsh, a program official stated that, “The corporation does not supply low-rental housing”, and that WHL had no intention of subsidizing its tenants, only building the lowest cost housing possible.


  • Even in the WWII era, “many municipal governments reserved hostility, or at best passive tolerance for WHL projects”.  The primary concern from municipalities was the perceived quality of WHL housing, as municipalities wanted more permanent (and more expensive) solutions.


  • The program was largely successful at delivering low-cost, temporary housing, but once the war was over and the immediate need passed, the government privatized and sold most of the homes, and what didn’t sell was transferred to CMHC.


How will WHL differ from Build Canada Homes?


1. WHL homes were temporary and not subsidized.

  • In contrast, we assume that all homes under Build Canada Homes will be permanent, and given the language that has been used around ‘affordability’, we anticipate that they will be subsidized to some degree.  This is especially likely after hearing Housing Minister Gregor Robertson’s comments last week that ‘he doesn’t want to see home prices decline and that the Federal Government needs to focus on affordable housing’.


2. WHL era permitting was comparatively simple and build times were extremely short.  The first National Building Code was adopted in 1941 to accompany the WHL building program (and along with it, the broad adoption of building permitting).

  • Modern building codes, permitting and inspection processes aren’t conducive to speed, even when using partially pre-fabricated solutions.
  • If the government is subject to the same planning and permitting queues as private industry, it will add to costs and timelines. If they aren’t, private developers will be furious.


3. Availability of land, particularly in urban areas.

  • Modern projects will need to put more units on smaller footprints, as developable tracts of land, even federal lands, are increasingly rare and expensive in urban centres.


4. Much of the oversight on WHL projects was provided by voluntary committees that supported land selection, negotiation, tenders, etc.  This kept costs down (until they started using more staff in 1945).

  • In 2025, all of these functions would likely be paid, not volunteer.


5. During and after WW2, there was general agreement and support from municipalities regarding the necessity for federal intervention as a wartime effort, despite not being overly enthusiastic about WHL projects.

  • Today, many municipalities are likely to be less enthused about federal intervention.


Government can implement the same strategy that it did in WW2, building the cheapest pre-fabricated buildings possible, but there are still challenges that exist today that either didn’t exist or were smaller in the 1940s.


We highly suspect that both government and hopeful citizens will be surprised at how expensive these units will be to build and buy. While new technology may drive efficiencies that we don’t account for, we think it will take considerable subsidization to hit an ‘affordable’ price tag.


Singapore Public Housing Model


Anytime public housing is discussed, Singapore is held up as the poster child of success. You’ve probably read Canadian news articles or social media posts about the Singapore public housing model, echoing common refrains like ‘if Singapore can do it, why can’t we’?


It’s true that Singapore’s public housing program has been extremely successful.  With home ownership rates at 90%, it’s easy to argue that they’ve done many things right.  The Housing and Development Board (“HDB”) has built 77% of the homes that Singaporeans live in today.



While there are some elements of the Singapore program that could likely be ported to Canada, there are fundamental differences between Singapore and Canada that would make implementation of a similar program much more difficult in Canada. 


Here are some important stats that help paint a picture as to why Singapore’s program has been so successful, and why it might be more challenging in Canada:

Source: Statistics Canada, Singapore Department of Statistics, Metro Vancouver Regional District. Chart compiled by Hawkeye Wealth


Singapore is less than 1/10,000th the size of Canada, and is less than a quarter the size of Metro Vancouver. Setting up an effective public building program in an area smaller than Metro Vancouver, but with 8x more population density leads to incredible economies of scale efficiencies.


Couple this with the fact that the national government owns approximately 90% of all land in Singapore and that there is one building code and no municipal or provincial governments (just one national government), and you have a situation where the government controls its own development and permitting process entirely.


The 12-person HDB board is capable of having a high level of knowledge and oversight for all projects. Their manufacturing and design is for a single climate, and pre-fab solutions have minimal freight considerations. Highly skilled development and construction teams don’t have to be mobile, allowing deeper specialization.


In short, Singapore is a near perfect place to run a vertically integrated government housing system.  In that context, Singapore’s policy decision to have the government become the main developer and run multi-billion dollar losses to build ~15,000 - 25,000 units annually is both wise and admirable. They only rely on private developers to build the much smaller proportion of housing types not provided by the Housing and Development Board (luxury homes, single family homes, etc).


By contrast, Canada is massive, has a huge variety of different climates that require different building solutions, along with inter-provincial variation in building codes.  Public projects require coordination with 3 layers of government (as well as first nations in some instances), all with differing ideas, values and roles. The personnel structure that would be required to provide a similar level of local specialization and expertise across Canada would be enormous compared to Singapore.


Canada is not Singapore. If Canada tried to run an aggressive public housing program that in any way threatens private development, it would fail miserably. Thankfully, it doesn’t seem like Canada is moving that direction.


Conclusion


For investors, the majority of the Liberal Housing Plan is positive for investors, and the Build Canada Homes is the only item that has the potential to discourage private building and investment.


However, given that the Fiscal and Costing Plan preliminarily allocates only ~$2.9B to that program annually for the next 4 years, we think its impact on private markets will be minimal.


Despite its limited scope, we genuinely hope that the Build Canada Homes program is efficient and successful. That said, a modern WHL program would be much more costly and difficult to replicate in the current era, and there are enough differences between Canada and Singapore to believe that their model would also be much more costly here.  Ultimately, Canada has an uphill battle to deliver cost-effective public housing.


The industry is a little nervous right now, and rightfully so given the current market challenges. But in our view, with the MURB tax exemption and DCC cuts around the corner, we believe the future is at least as bright as it was before the Build Canada Homes announcement.

SUBSCRIBE TO THE BIRD'S EYE VIEW

By Hawkeye Wealth Ltd. April 19, 2025
Introduction Election season is here, and while housing affordability and availability have taken a backseat to how Canada should approach its relationship with the United States, changes to housing policy still feature as central pillars of both the Conservative and Liberal party platforms. What makes their proposed changes particularly notable is that since the 1980s, the Federal Government has played a smaller role in housing compared to Municipal and Provincial governments, influencing markets indirectly through immigration and monetary policy. Those days look to be over, as both parties have introduced proposals that would see the Federal Government take a much more active role. In this edition of the Bird’s Eye View, we review the housing platforms for both the Conservative and Liberal parties, and offer our opinion on how these policies will impact development generally, and real estate investors specifically. Note: We recognize that other parties also have housing platforms, but for brevity, we are only covering the Conservative and Liberal platforms. Policies in Common Between Conservatives and Liberals Before we dive into the novel proposals from each party, we begin with three policies in common: 1. Elimination of GST on new homes Both parties have proposed to eliminate GST on new homes, but there is a massive difference in the size and scope of the two programs:
By Hawkeye Wealth Ltd. February 22, 2025
Most investors would be thrilled with the outcomes forecasted in the CMHC 2025 Housing Market Outlook given the level of uncertainty ahead. The question is, how likely is CMHC to be right?
By Hawkeye Wealth Ltd. January 25, 2025
Demand is high and has nearly chewed through the supply overhang in many markets, which should result in rising rents and falling vacancies over the next few years.
By Hawkeye Wealth Ltd. December 21, 2024
While GDP isn’t a perfect predictor of housing prices, the two tend to run in the same direction. If we do in fact see a decline in GDP from 2024, it would take a unique set of circumstances to see anything more than flat housing prices in 2025.
By Hawkeye Wealth Ltd. November 23, 2024
It doesn’t take a genius to hypothesize that population decreases could cause rental rates and housing prices to soften over the next two years. However, a look at historical data shows that changes in population growth often don’t result in immediate housing price changes
By Hawkeye Wealth Ltd. August 24, 2024
While we will continue to watch and seek to understand how investors may be affected by who ends up in government and any resulting policy shifts, there is something reassuring about knowing that in the past, real estate has performed well for investors regardless of who is in power.
By Hawkeye Wealth Ltd. July 23, 2024
In this edition of the Bird's Eye View, we look at deal structure changes that developers are making to attract LP investors, and how these features are making deals more attractive on a risk-adjusted basis.
By Hawkeye Wealth Ltd. June 21, 2024
In this edition of the Bird’s Eye View, we cover how real estate prices have historically moved in falling rate environments and explore some of the perils both in forecasting rates and what prices will do as a result of changes in rates.
By Hawkeye Wealth Ltd. May 25, 2024
In this edition of the Bird’s Eye View, we review and analyze the 2024 Housing Market Outlook report, prepared by CMHC, where they forecast major price and rent increases in Canada over the next three years.
By Hawkeye Wealth Ltd. April 27, 2024
In this edition of the Bird’s Eye View, we provide summaries and commentary on federal budget initiatives through the lens of private real estate investment. We focus on items that have seen less commentary from media, such as the potential freeze of development cost charge increases in communities over 300,000 population, and the Public Lands for Homes Plan, parts of the recently released Solving the Housing Crisis: Canada’s Housing Plan .
Show More