The difference showing that Ireland, Canada, New Zealand, and the US had either the same, or more dwellings per thousand inhabitants in 2011 than 2022?

I don’t know if I’m just misunderstanding your point, or misreading the material, but that is also not an example of underutilization.

My point is that the relatively static proportion of people to dwellings in those countries over that period suggests that a supply shortage is not sufficient to explain the accommodation crisis.

What I am suggesting is that incentives to invest in property have led to an increase in the number of properties that are not being used as a primary residence (i.e. are underutilised) and that that is the most significant factor causing the shortage of available homes.

I think you’d want to look at household size - 4 five person households need fewer houses than 10 two person households, even though the population is the same.

That's a good point and I have done that in the past (for NZ). Household size has remained steady over roughly the same period (decade up to 2018 where census data is available). This doesn't account for a possible significant change in household size distribution of course (e.g. the possibility of many more elderly living alone while young people increasingly live in shared housing).

I see now through a US & REIT centric bias, I think.

Given a market like NZ’s size, growth/prices, and attractiveness I can see demand like speculation, STRs, and 2nd homes possibly affecting utilization more than supply.

Given US markets that are larger, cheaper, and less attractive where investment properties fueled on cheap credit were rented out or to a much smaller degree flipped, it’s truly, significantly more difficult to imagine vs simple lack of supply (especially given historically low US rental & homeowner vacancy rates).