You have it a little off. This is based on census data where they literally just ask "does the owner of this residence live here?" and "does anyone live here?".

You have it a little off. This is based on census data where they literally just ask "does the owner of this residence live here?" and "does anyone live here?". The "homeownership rate" is just the yes's to the first question divided by the yes's to the second question. So, by definition, this excludes homeless people, people living in illegal dwellings and counts 7 roommates sharing a single apartment as one "non-owner-occupied household". Also, this is based on randomized legally required self-reporting. I'd imagine there is a skew towards people who are stable enough to own a home among the people who respond. People assume it's "percentage of people who own a house", but it's only about the houses themselves. As to a better measure, a simple better measure is to look at the ages involved. In 1982, 70% of household based around 35-45 were owner-occupied household. By 2021 that number was 60%. It was pretty much a straight line down. Meanwhile, older people saw that rate continue to level off or even grow. Hence, even the flawed data shows that the trend is worsening. So the post you are responding to, once you look at the affected generations, doesn't even say what the original poster was claiming.