On the surface, I agree with the author's reasoning that eventually many millennials will "grow-up" and want suburban homes for their kids with better schools.

On the surface, I agree with the author's reasoning that eventually many millennials will "grow-up" and want suburban homes for their kids with better schools. And this makes sense because it would follow trends in the past. However the author is missing a couple key components that lead me to believe the future will not be like the past. For example... 1.) In the past there were suburbs to move to. For example, in the 80s and 90s, Contra Costa (Walnut Creek, Danville, etc) were being developed as a viable suburb to SF. However, as far as I know there are no new suburbs being developed that provide a reasonable commute time or distance from SF. 2.) Of the suburbs that exists, those that are a reasonable commute time from major hubs are still pretty expensive. Again Walnut Creek maintains its value quite well. So it's not really a "cheaper option" from the city. 3.) Baby Boomers are living longer and staying in their suburb homes longer, which doesn't give an opportunity for millennials looking to start a family to move in. I've long since thought that instead of building new suburbs, we should be looking for options to move empty-nesters from their 4-bedroom homes that they don't need anymore. All of my friends parents still live in their suburban homes because they have no other options for the most part. 4.) Author assumes all millennials are waiting to have kids. (They're not.) 5.) We have plain old population growth to rely on. Cities collapsed in the past because there were places for people to go and a relatively smaller population. I strongly feel that we've exhausted a lot of those options while our population continues to grow strongly (at least in the US). Therefore, cities will remain full simply because of lack of other options and population growth (both by birth and immigration).