Paying $100M/acre for land would only be the case in the most expensive parts of the city, but there is nothing (outside of zoning) requiring you use the most expensive land, and most of that land already has a tall building on it which doesn't make for an ideal site when you want to build a new one. There exist properties for sale in San Francisco with more than an acre of land that cost less than $10M, and indeed the average cost of land is less than $10M/acre.

Your method of calculating costs is also flawed. Things like loan interest, taxes and realtor fees are a percentage of the other costs, not a fixed amount. If construction and regulatory costs were lower then you could take out a smaller loan and pay less interest etc.

The dominant costs here are the premium on sites zoned for tall buildings, which is the cost of zoning rather then cost of land and could be alleviated with the stroke of a pen, and construction costs which are themselves significantly higher than necessary as a result of regulatory requirements.

The premise here is flawed. Due to zoning, most housing in SF / the Bay is not dense. The post I originally replied to blamed union labor for the price of housing. For non-dense housing, land is a huge portion of the price.

Yes, as a thought experiment, you could build an infinitely tall building which pushes the per unit cost for land to zero. That’s not what’s happening in SF, and it has nothing to do with labor. Another obvious point: building tall buildings requires more engineering, more expensive materials, more difficult labor, and more safety than small ones.

Your argument is all pointless hypotheticals. Put up or shut up: show me the numbers that say labor is driving the price of housing in SF. You’ve done nothing to support the idea that labor is driving the actual costs of real housing built in SF.

> For non-dense housing, land is a huge portion of the price.

That isn't the cost of the land, it's the cost of the zoning rules that prohibit density. And, more to the point, construction costs, because now you have a problem where the only place you're allowed to put multi-story buildings already have multi-story buildings and to be profitable you need construction costs low enough to justify replacing a 10 story building with a 20 story building instead of replacing a single-family home with a 10 story building for the same benefit. Which, as you point out, costs more because the building has to be taller, and also costs more because you have to knock down the existing building and do twice as much construction to add the same number of units. Multiplying the effect of any increase in construction costs.

> Yes, as a thought experiment, you could build an infinitely tall building which pushes the per unit cost for land to zero.

It doesn't have to be infinitely tall, buildings tall enough to make even high land prices irrelevant are routinely built in practice.

> You’ve done nothing to support the idea that labor is driving the actual costs of real housing built in SF.

Your claim was that the primary cost was land, but the primary costs are zoning, construction costs and their interaction.

My claim was that labor was not the primary cost, and I guessed that land was the biggest driver. You have confirmed it basically is, although as a function of things like zoning. That’s trivially obvious. That was what I meant when I wrote the post. Obviously zoning is not the same thing as labor.

Lumping zoning in with “construction costs” makes zero sense. Everything in some sense fits that definition. To be clear, zoning costs definitely are not labor costs. The way you are using these terms is incorrectly conflating things.

Again because you don’t seem to get it: the post I was responding to blamed (union) labor for the price of construction. Clearly incorrectly.