The Mom-and-Pop Investment Trust

The financialization of housing is being driven by households — not corporations.

An abandoned home in St. Charles, PEI. From Flickr.

One of the old ideas that does not seem to die in Canadian housing discourse is that we do not need to build more because we are supposedly sitting on millions of empty homes. The idea emerged because of vagaries in how the census defines an occupied vs. unoccupied dwelling. Since Statistics Canada generally classifies dwellings as being (or not being) “occupied by usual residents,” everything else gets hastily clumped into the vacant category. The broadness of this category has fueled some misinformation, especially among supply-skeptic authors like Better Dwelling’s Stephen Punwasi, who wrote last year that Canada has 1.3 million vacant homes.

The idea has convinced some activists that Canadian cities are wellsprings of empty dwellings that are ripe for expropriation:

I wrote about this issue in the past, but TLDR — dwellings are far more likely to have occupants, especially in urban centres where housing crises are most acute. The “vacant” category includes a lot of innocuous reasons for the absence of usual residents in a dwelling. It could be because the dwelling is a seasonal/vacation property, a home that is listed as being for sale or rent, or a recently sold/rented unit whose new occupants have not yet moved in. Beyond these broad reasons for the emptiness of a unit, a property can also be under construction/renovation, unfit/unsafe to have occupants, or built as a granny suite on a lot with another regularly occupied home. Unfortunately, the lack of disaggregation means a lot has been left to the imagination of vacancy truthers.

Even in Vancouver, a city where conversations about empty condos tend to trump building more of them, only 96 condo units built between 2017 and 2022 had been kept empty and unsold at the time of completion. Despite accounting for less than a percent of all condo unit completions in the city during that period, uncontextualized stats on empty homes tend to drive the narrative that investors are building skyscraper condos only to leave them empty until they can turn a handsome profit.

The good news is that a new Statistics Canada table will hopefully demystify some of this. Published at the same time as Canada’s GDP data, we will now get housing stock unit totals “by institutional sector, housing type, dwelling occupation, dwelling type, and tenure type.” By using a combination of economic estimates and census data extrapolation, this table will improve the quality of housing debates.

A narrative on empty homes that is full of something else

For starters, we can say confidently that the number of truly unoccupied dwelling units in Canada is likely no more than 1.2 million. Since prior data did not distinguish between housing that was unoccupied vs. occupied by non-usual residents, Punwasi’s 1.3 million-home estimate of empty housing also includes almost 80,000 dwellings with students, temporary workers, and families of foreign consular staff living in them. While many of these cohorts would not have been counted in the census as usual residents, they are still families living in Canada that do not deserve to be othered for ideological convenience.

The second interesting observation is that unoccupied housing is generally found in rural and low-density areas rather than along stretches of high-rise condos. More than half of Canada’s unoccupied dwelling units are single-detached homes. Although the vacancy rate of apartments is slightly higher than it is for all dwelling types (8.4% vs. 7.5%), the dwelling type with the highest rate is mobile homes (16.9%). In Newfoundland and Labrador, the share of unoccupied mobile homes is almost 50%.

Much to Loreto’s dismay, Canada’s unoccupied housing is likely far outside the dense urban cores of large provinces like Ontario and BC. Expropriating a half-empty trailer park in Rural Newfoundland is not a way to reduce homelessness in big city downtowns.

The share of unoccupied dwellings is lower in more urbanized provinces such as BC, Ontario, and Quebec.

Taking ownership of ownership data

The new dataset points out some of the holes in the mythos of Canada’s vacant housing, but it goes further — it also pours water over some pervasive narratives on institutional ownership and financialization. Under the pretense of advancing housing as a human right, social media activist Leilani Farha recently told the Hill Times that: “the traditional rules of supply and demand no longer apply, as new supply turns into assets for large investors instead of housing for people who need it [because] institutional investors have achieved such a dominant position in Canada’s housing system.” But according to Farha, what constitutes a dominant position? According to the dataset, the combined share of dwelling units owned by financial and non-financial corporations is only around 4.1% of Canada’s total private rental stock.

Private rental stock is comprised of dwellings owned by households and corporations, while social housing stock is comprised of dwellings owned by governments and non-profits.

This stat may shock many, not least the small handful of activists who have built their brands on the premise that Canada’s housing market was just a few REITs in a trench coat. If we are to use the Canadian Housing Statistics Program definition of an investor, it is actually private households that are most responsible for the financialization of housing — not corporations.

NOTE: The line gets fuzzy on household-owned apartments because owner-occupied condos would not be considered investment units, while all other forms of owner-occupied apartment units would be. The investment property share of household-owned housing is also likely higher because there is no way to account for individuals who own multiple properties. Learn more here.

There are some gaps in this dataset. For starters, it only provides data at the provincial-territorial level. There may be more evidence about the corporate share of housing stock ownership in metro areas, but even moving the goalposts in that direction will likely not paint a narrative that is much more convenient. It is also not clear where co-operative housing was placed, as co-ops provide social housing but they are technically not non-profit institutions. Quarterly data on unoccupied dwellings is also difficult to track (especially on the reason for vacancy), meaning there is certainly some room for improvement.

But as a relatively new product, it will do a lot of good in debunking the dogma of old-school housing advocacy. The dataset comes at a crucial time when Canada needs to build more than ever — especially for students, seniors, renters, and low-income households. Statistics Canada’s products show a positive direction in providing more data to support these objectives. And shedding more light on what housing financialization actually looks like is the first step in addressing the problem.