Looking at a piece that went viral on housing Twitter.
TL;DR: The number of net new households is limited by the number of net new homes. So it should not be used as an indicator of the adequacy of the housing supply.
Consider this a bonus piece.
The piece Want to solve the housing crisis? Address super-charged demand went viral on housing Twitter this week, and boy, is it a bit of a head-scratcher. You know you’re in for a treat when you see this strawman:
This “lack of supply” view draws on basic Economics 101 textbooks, where using the example of widgets and a simple supply and demand curve, an increase in supply causes a reduction in price.
I have never referred to a supply curve once in my housing work, and please help me if I ever do. Just check out Baby Needs a New Home — 178 pages on Ontario housing, which examines demographic change, immigration and international student policies, and housing completions. No economics, not even Economics 101, and the only reference to a curve is Curve Lake First Nation.
But enough about that… it then follows-up with this display of data.
Nationally between 2006 and 2016, Canada added 1.636 million households and built 1.919 million new homes, according to the Canada Mortgage and Housing Corporation and Census data. So, on average, almost 30,000 extra homes were constructed each year compared to the increase in the number of households.
There are three big problems with using housing completions relative to the number of net new households as a test of whether we are (or are not) building enough housing. In order from least to most problematic, we have:
- This analysis treats all units the same, despite a one-bedroom apartment housing far fewer people than a family-sized home.
- Household growth is a net number (number of new households minus the number of households that left), but new dwellings completed is a gross number. Until you subtract out the number of homes that were knocked down, this is an apples-to-oranges comparison.
- And… most importantly… the number of net new households formed is a function of the number of net new homes. They’re not independent of each other!
Some Twitter users caught this issue instantly:
The definition of a household is important here. Statistics Canada defines it as follows:
Household refers to a person or group of persons who occupy the same dwelling and do not have a usual place of residence elsewhere in Canada or abroad. The dwelling may be either a collective dwelling or a private dwelling.
Imagine a tornado rips through a town and destroys half of the houses in the community, but miraculously no one is hurt. Half of the town, now homeless, is invited to move in with the residents who did not suffer losses.
Here is what the data would show as far as the number of households and the number of homes.
Does anyone want to seriously argue that this graph shows the community does not having a housing shortage, as the ratio between households and housing units is unchanged?
Tornados aside, there are many ways people can respond to a shortage of homes. The first thing they could do is move somewhere else (that is, go be a household someplace else). And, guess what, that’s exactly what we’ve seen in housing constrained markets such as the City of Toronto and Peel Region:
Between July 1, 2019 and July 1, 2020, 60,000 people moved out of the City of Toronto and Peel Region to other parts of Ontario. As the chart above shows, it was primarily people in their late 20s-early 30s, and kids under the age of two. Young families, in search of homes.
Or, maybe, instead of starting new households, those young people continue to live with their parents (so we have one household, instead of two). That is exactly what is playing out in Ontario, as shown in the piece More 25–34-year-olds are living with their parents in cities in Ontario… but not in Alberta and Quebec:
This phenomenon of young people living with their parents longer due to housing shortages also helps explains why fertility rates are falling faster in Ontario than they are in the rest of Canada:
In short, the number of households in a jurisdiction is a function of the number of housing units, so as such it should not be used as an indicator of the adequacy of the housing supply!