I don't think the demand is literally boundless, but I think the idea that we could easily outpace demand with construction to dramatically increase affordability is wildly optimistic.
I'd rather build new Dallases, or denser Dallases, than try to double the density of Manhattan, because it will both be easier to find/acquire un-utilized land to use, and because the amount of work required to double the density of Manhattan is much more intense, construction-wise.
Of course density should be pursued wherever it's cheapest to do so. The point of thinking about the Manhattan case is that this is the natural endpoint of any sort of "more housing and higher density always leads to higher prices" argument. It clarifies to what extent that argument can possibly work, as a matter of basic logic.
As far as I can tell, the Manhattan case tells us that we aren't even close to the "building more will cause prices to fall" point anywhere in the US. And because of that, building more where it's cheaper is a better strategy for controlling costs than building more where it's already expensive.
Manhattan was brought up to mock the idea that construction could make prices worse for existing residents, I don't think it shows that at all.
Is it any wonder that "you're dumb, building more will lower prices" hasn't been a persuasive argument to people who've seen construction push up prices around them for decades without the promised lowering of prices yet? That's just asking for a leap of faith that this time will be different, we'll hit the magic tipping point, without any actual evidence for that.
> Is it any wonder that "you're dumb, building more will lower prices" hasn't been a persuasive argument to people who've seen construction push up prices around them for decades without the promised lowering of prices yet?
You're making an assertion about causality that often gets thrown out, but typically without being backed up.
Certainly construction often correlates with prices rising. However, the arrow of causation could run in either direction. It's possible that construction itself increases demand in the area enough to raise prices. However, it's also possible that increased demand (due to agglomeration effects, etc) raises prices and results in construction to meet that demand, which lowers prices in the area relative to what they would have been without the construction.
This doesn't necessarily mean that construction will lower housing prices, but it does mean that without it prices will be even higher, because the construction is attempting to meet new demand that's already there regardless.
"Expensive" city centers in the U.S. have comparable density to bog-standard suburban towns in the rest of the developed world. By all indications, we're nowhere near the "tipping point" where trying to increase density there would physically be so costly as to be unsustainable. That's something NYC can perhaps legitimately worry about, but I'm not sure how that could apply to other high-demand places in the U.S.
The tipping point of concern for the anti-gentrification crowd isn't "is it profitable to upzone", it's "will upzoning make things cheaper" and those are VERY different questions.
But just going on the numbers for the former question, the "expensive" city centers in the US generally start with Manhattan, Brooklyn, SF, and whatever you'd call "center" of LA. And they don't seem that far behind a lot of the rest of the developed world...
Manhattan is >70k/sqmile.
Brooklyn is 37k/sqmile.
SF is ~19k/sqmile.
LA is 8k/sqmile.
So what are the "bog-standard suburban towns" you think they're comparable to?
I can't think of many, and my searching isn't turning many up. Milan appears to be 20k/sqmile while Rome is 6k/sqmile. And those aren't "bog standard suburbs" like a Marietta, Georgia; those are central cities! Let's go bigger and even more central - London is 15k/sqmile, Paris is 53k/sqmile. Tokyo is ~17k (and larger, more like a double-size LA, there), but not overall denser than the US's expensive city centers. Manchester, in the UK, is 5450/sqmile. Stuttgart is 7,800/sqmile.
Manchester is apparently comparable to San Jose! So again, seems like the way to fix is it to focus on the places further down the list, like an Austin, TX, at 3k/sqmile.
I'd rather build new Dallases, or denser Dallases, than try to double the density of Manhattan, because it will both be easier to find/acquire un-utilized land to use, and because the amount of work required to double the density of Manhattan is much more intense, construction-wise.