Yes, there is an endgame here -- and also, truth be told, construction costs per unit start to go up once you pass 4 or 5 stories, so there's other economics at play.

But that endgame is Tokyo or midtown Manhattan, and we're not close to those levels of density basically anywhere on Earth -- and certainly not in the USA, where you can count with one finger the number of cities that might not be able to increase density because the buildings are so tall it's hard to imagine building more-attractive taller ones.

> But does this not apply primarily to the land underneath the building? Once you reach a certain plateau a 700sqft apartment on the 10th floor of a 50 story building isn't worth any more than an apartment on the 10th floor of a 70 story building is it?

This is an interesting question -- in my example, you might think it's pretty clear that a single condo in a 4-plex would be less valuable than the same land area with a single family home (even if all 4 units together are worth more). But it's not clear why a single condo in a 25-story doorman building is inherently more valuable than a similar condo in a 50-story doorman building, which is basically the only density increase that's realistic -- and in fact, maybe the larger building supports more amenities, making the individual units more expensive as density increases? Also, this level of density supports rapid transit that allows for more effective geographical distribution than the "freeways choked with traffic" sprawl we're seeing today.

That said, this is not the situation that SF (or really, any market in the USA outside of Manhattan) finds itself in. In all these cities, developers could build a larger number of less-expensive units to replace more-expensive lower-density units on a given land area (and still profit! And increase housing supply!), if they were allowed to by zoning and other (typically local) policy.