San Francisco With 200,000 More People — Would we be Better Off?


You want something truly scary? Take a look at these mockups of what San Francisco might look like if we build all the housing that the developers say we need.

According to writer Greg Ferenstein,

The city probably needs somewhere north of 150,000 more units: most high-rises would be concentrated in the Eastern, Downtown, and mid-market areas, while every block in the entire city would need at least one 7-story building. Essentially, San Francisco would be Manhattan downtown and Paris everywhere else.

Set aside that I never want to live in Manhattan (at any price), and that the infrastructure to handle 200,000 more people would be horrendously expensive (and developers are already refusing to pay their fair share for far lower levels of need).

It’s not just “how to we build that much housing.” It’s how do we build maybe $20 billion or more worth of transportation capacity to handle that density. Manhattan has a citywide underground transit system with high capacity and no surface traffic issues. SF doesn’t, and won’t, as long as we can’t raise property taxes and refuse to charge developers for the cost of that new system.

Never mind, let’s take Ferenstein’s idea and play it out. Suppose we decided as a city that we are willing to accept a lot more density in exchange for affordability. (This is something the mayor is promoting). Let’s say that the city really needs to build highrises all over the eastern side of town (why only the east?) and put mid-rise buildings everywhere.

Let’s say we decide that 47.5 square miles of space are enough for1 million people, and that we are willing to give up everything about San Francisco that we would lose in the process.

Remember, the streets in the highrise districts in Manhattan are much broader than the streets in SF, able to handle more traffic, with big sidewalks that can handle more pedestrians – and still it’s often overwhelming.

Right now in SF, for example, I am able to walk down the streets at 5pm without being jammed in a pack of stressed-out pushing people, which is life in parts of NYC. It’s possible to able to take your young kids and your dog for a walk in a place where there’s actually room to walk.

Imagine Mission and Valencia, being packed with thousands more pedestrians. Don’t even think about the traffic.

In fact, unless we took entire streets and banned cars, forget about the bicycle lanes – they are narrow and limited and can’t easily handle say 200 percent more traffic.

But again, whatever. Let’s say that it’s elitist to try to keep the charm of a human-scale city in a world-class city like SF, which Ferenstein calls “quaint.” Let’s say that our only hope of avoiding being a city of just the rich is to build all the apartments and condos anyone could every want to build.

Let’s say we have that debate and decide that the need for affordable housing trumps all, and we will just have to live with the implications.

So what happens if we let the developers build 200,000 new units – and prices don’t come down?

That’s actually a pretty likely scenario. It’s happened in other places (NYC, for example, where lots of new housing is being built and prices are not in any way coming down.)

It’s happened in SF so far, where we have built more market-rate housing in the past four years than at any point since the 1960s, and prices continue to soar.

Ferenstein talked to an econometrics expert at a credit agency. Okay. No idea if this person has ever studied housing or housing price trends in San Francisco, but he has a model. It assumes that we have to build housing faster than the population grows. Nice.

Except that market-rate housing causes population growth as fast as it solves it – that is, if your model is the traditional capital-market model, you can’t keep up with population growth by building. You might as well try to decrease traffic by building freeways; never works, never has – not in San Francisco.

And how come we never talk about why the population is growing so fast, and why so much of that growth comes from one industrial sector that hires one type of workers?

I emailed Ferenstein with my questions, and here’s what he said:

Well, prices don’t fall here because we don’t build enough. It’s been an issue for decades. And, if you build enough units, prices will fall. You just have to build more supply than people. The question is whether it is possible to do so. But, I’m actually not advocating for that. I’m advocating for *some* solution. If the city decides it doesn’t want to grow, then it should be responsible for finding some solution where people can live and work in the same city–somewhere. Maybe it’s San Francisco. Maybe it’s Oakland. Maybe it’s a new city. But there has to be a giant metropolis somewhere. And, San Franciscans must realize, if jobs relocate elsewhere, they will suffer massive inequality and terrible commutes.

Interesting argument. Of course, we are not talking about a city where people live and work; San Francisco’s housing crisis in large part the result of people living here and commuting to Silicon Valley, on private buses. The Valley cities build no housing at all, and expect us to solve the problem.

And I would argue that if some tech jobs went elsewhere, we would have less inequality and less terrible commutes – it’s the displacement from too many people moving here for jobs when housing doesn’t exist that has created the problem. Most of San Francisco does better when there is slower growth in bubbly tech industries.

There’s a much more interesting question that we might want to address: Suppose we built may 20,000 new units, or 30,000, or 50,000, spread all over the city – and every one of them was social housing, that is, housing that was never in the private sector? Would that bring down prices? Would that provide the same level of affordability, or maybe much more, than the Manhattan West model?

Would that be a better deal?

At the very least, we would know that the new housing would be affordable, instead of taking a huge gamble that the (failed) free market, and the (failed) econometric projections of the past, would save us.

Oh, and what if we said that SF no longer wants to be the bedroom community for Silicon Valley, and will stop entitling things like private buses that make that trend possible?

That’s a bit of a different picture.

This piece originally appeared at

Photo: A mockup by Alfred Two for a Medium story on what an “affordable” SF might look like

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

How to increase housing stock

Around 1/3 of all SF housing is single-family homes. The City or the State could simply take one-half, say, of all single-family homes, divide the half taken into one or more apartment units, and pay the owners "just compensation" by assigning them the rents. If the City wanted to mandate that some of the new rental units be "affordable," then the foregone rent could be made up to the owners in the form of some kind of tax credit. The process for determining the "just compensation" would take decades probably, thereby making it a problem not for today's politicians who enact the scheme but for their successors. This is called a win in politics. You would double the housing stock overnight without any new construction, thereby preserving the optics.

Jerry Brown has recently touted the coercive power of the State. Well, here is a great opportunity for that.

EDIT: to clarify, I am proposing that 1/2 of *each* single-family home be taken.

Add housing and shake...

I'd like to see a mockup of what the proposed denser San Francisco will look like after the next big earthquake. There's a reason they don't have tall buildings all over the place in coastal California.

For that matter, look at the Manhattan skyline. Even without earthquakes, the height of the buildings goes up and down and up and down again, varying with the ability of the bedrock to support tall buildings. And that's with fierce economic incentives to build tall, incentives that have operated for generations.