Here is an interesting housing chart from Ryerson University's Centre for Urban Research (CUR) using data from CMHC:

What it shows is (1) the number of new housing using created through the addition of secondary suites, such as basement apartments and laneway suites; (2) the number of housing units lost to demolition or "deconversions", such as when a duplex or triplex gets converted (back) to a single-family home; and then (3) the net new units added over the last three years.

In looking at the chart, you'll see that the City of Toronto actually lost about 2,000 units from its existing housing stock between 2019 and 2021. Again, these numbers only consider what's happening in the city's existing low-rise residential housing stock. They don't factor any of the housing supply being delivered through new condominiums and multi-family apartments.

Still, it's evidence for something that is perhaps already well known: many of Toronto's low-rise neighborhoods are losing people. They are losing people because the existing structures are housing fewer residents and they are losing people because we make it difficult to build new housing. We want them to be "stable." But stable built form doesn't necessarily mean that things aren't changing on the inside.

Now compare this to what's happening in Brampton (a suburb of Toronto). CUR is calling Brampton the land of secondary suites. Over the last three years, it added nearly 11,000 housing units and was on pace (at the time the data was published) to create nearly 6,000 last year alone (most of which are basement apartments). This is all within its existing housing stock.

With all of this, I think there's an interesting question about about how much of this is being driven by market demand and how much of this is being driven by land use policies. There's obviously demand for expensive single-family homes in Toronto, which is why "deconversions" are happening. But to what extent does this change if/when we become more permissive around multi-unit dwellings?

I think it depends on how we craft the policies.