This thread from demographer Lyman Stone on the definition of the US homeownership rate has stuck in my head for a couple of years now. Reading it produced a pretty profound “oh” for why this particular metric didn’t line up with my perception of the issue.
To put it simply, the definition of the homeownership rate is:
Take the number of households where the home is owned by the household head, divide by the total number of households.
The homeownership rate is based on households, not individuals. If an adult child lives with their parents (and their parents own their own home), they are counted as “homeowners” for the purpose of the homeownership rate. If more and more people in their 20s and their 30s move in with their parents (or never move out in the first place) rather than renting an apartment, this has the effect of increasing the homeownership rate, because you have reduced the denominator (number of households) without changing the numerator (number of owner-occupied households).
Canada uses the same definition:
The homeownership rate refers to the proportion of all households that are owner occupied.