San Francisco is Doomed: A Cautionary Tale for Growing U.S. Cities

Graffiti on a development rendering in San Francisco's Mission District. Photo by Flickr user Torbakhopper.

The fact that San Francisco is home to the most obscene rents in the country is old hat by now; it's not really news to anyone. That said, the latest battle in the affordability war—currently being waged in the Mission between those that believe more housing is needed and those that want a moratorium on nearly all new construction—got me to thinking about where this is all leading. And those thoughts are bleak. San Francisco, by any reasonable measure, is doomed.

More than a year ago I wrote about the general phenomenon that rents tend to be more stable than housing prices, and that rents very rarely actually decrease, whereas housing prices can swing wildly. Short of a mass exodus of residents, which is unlikely for a variety of reasons, current demand for housing in San Francisco is unlikely to decline. Even if it did, the immediate response would be for developers to lose interest in building more housing. They would bide their time, waiting for demand to pick back up and profits to return before constructing any more new housing. It's very hard—probably impossible—to go back to the good old days.

The Bay Area is still booming despite the incredible housing crunch, and for all I or anyone else knows, that may very well continue. But it definitely weighs on the minds of potential future residents and workers, myself included, deterring future talent that's interested in someday owning a home, or even just paying less than half their income on rent. I'm sure that many current residents say to that "good riddance," but this extends to new industries and employers as well. The region may have a lock on the tech sector, but startups for unaffiliated industries would generally be foolish to locate in the Bay Area. San Francisco is on its way to becoming a monolithic populace, not just socioeconomically but commercially as well.


Since rents are unlikely to ever dramatically decline in San Francisco (or just about anywhere else—even Detroit's median rent has consistently increased over the past 25 years, while home prices fell by about half from 2006 to 2011), there are really two possible outcomes. In one, the supply-siders get their way and lots of new market-rate housing, and a decent share of affordable housing, is built in the Mission and across the city. In the other, the anti-development folks get their way and very little new housing is built in the coming years, with most new construction coming in the form of affordable housing constructed with city, state, and federal funds. The first outcome is extremely bad, the second is utterly disastrous.

Path 1: Increased Supply, Increased Short-term Displacement, Bifurcated Population

If the supply-siders' dreams come true, San Francisco will get off its ass and at least make a show of trying to build enough to meet the phenomenal demand for housing in the Bay Area. The city's housing production hit a 20-year peak in 2014, with 3,514 new units in a city of about 850,000. Seattle, a smaller city of about 670,000 that has also struggled with rent increases (though not nearly to the degree of San Francisco and other Bay Area cities), built 8,311 new units last year.

20-Year new housing construction trends from 1995 to 2014, from the 2014 San Francisco Housing Inventory.

What many anti-growth, pro-moratorium advocates fear—justifiably—is that more new construction will need to more displacement of existing residents, many of whom are low income and could not afford to rent a market-rate unit in nearly any part of the Bay Area, to say nothing of San Francisco itself. Owners of some apartments will evict current residents in order to build larger multifamily developments—those residents are generally compensated by up to $50,000, though even that maximum payout is not enough to afford unsubsidized rents in the city for very long. A share of those new units must be set aside as affordable to low income households, but the number of new affordable units may or may not exceed the number of households that are displaced.

Following this path, more total housing is provided in the city, which somewhat moderates rent increases in the long run. A pretty sizable amount of affordable housing is built, funded entirely by private investment, because of the inclusionary zoning program mentioned above. The majority of homes are market-rate though, and with median prices of $3,000 or more per month in many areas, they are far out of reach even to middle class households. In neighborhoods with a lot of development, we end up with a lot of rich people housing and a decent amount of poor people housing, and basically nothing in between.

Path 2: Limited Supply, Decreased Short-term Displacement, Homogeneous Population

At the other extreme there is the anti-growth path, which I should note is not necessarily the long-term vision of all those in support of the Mission moratorium—instead, it's viewed by many as a "cooling off" period where the community and city planners can create a framework for more equitable development. That said, in a regional market that's not providing enough housing overall, there's not much one neighborhood in one city can do to limit rent increases in the long run.

If moratorium supporters succeed at limiting the pace of development for years to come, it will be a Pyrrhic victory. Fewer residents will be evicted to have their units torn down or renovated, which is no doubt of great value to those that are protected, but that won't mean the neighborhood will maintain its character. Rent control only stays in effect as long as people stay in their units, so when they move out, those units will revert to market rates, forever out of reach for low income residents. Many of these units will be fixed up as rent controlled residents move out, pulling even higher rents. For those who don't move out, well, no one lives forever.

Private development is a relatively cheap way to provide income-restricted affordable housing, and there's just no way that fully publicly-subsidized construction is going to fill the gap at upwards of $300,000 per unit. Nor should San Francisco or any other city want to rely on public funds for all of its affordable housing needs: Every dollar spent on affordable housing is one that isn't spent on education, parks, public health, or infrastructure. Over time, without a meaningful amount of private housing construction, the city will consist of essentially nothing but luxury housing—or rather, pretty standard housing at luxury prices. The transition will take longer than in Path 1, but the transformation will be absolute in a way that even private sector developers cannot hope to match.

The Third Path: Something in Between

The examples above are, to some degree, two extremes on a spectrum of infinite possibilities. Many of the in-between approaches are probably superior to either of the extremes, and I imagine that many of the pro-moratorium advocates and community members are just looking for some breathing room to develop an in-between framework that suits their needs and recognizes the validity of their concerns. A revised development framework has the potential to affect the lives of real people in positive ways, so I don't mean to discount it entirely. But whatever the outcome, it will amount to tinkering at the edges compared to the scale of the problem.

Anyone who's read my blog knows that my opinions are decidedly on the "build more housing" side of the argument, but this post isn't about villainizing moratorium supporters in the Mission. It's about acknowledging the fact that rents are very unlikely to decline, and that, as a result, things are unlikely to get any better for San Francisco or its low income residents. No city can afford to subsidize the rent for half its population.

San Francisco is in a particularly unenviable position because, despite its failures, it has actually done more to combat increasing rents than other Bay Area cities—places like Palo Alto and Mountain View that have greedily attracted tech employers while leaving the un-sexy, and much more costly work of providing housing to their larger, more compassionate neighbors. Not to absolve SF entirely, but no single city should be forced to completely transform itself to appease the greed of its neighbors, especially when those neighbors make no sacrifices of their own. Because of the limited size of San Francisco relative to the Bay Area as a whole, even a complete "Manhattanization" of the peninsula might only have delayed its fate in lieu of similar efforts from Oakland, San Jose, and all the smaller cities in between. San Francisco was never in a position to fix the problem of housing affordability all by itself, but this is not the case for many other places in the U.S.

So, growing coastal cities throughout country, take note.


Nothing new here: Growing cities need to build more housing where and while they can, period. The more they build, the less competition there is for each available housing unit, and the less rents will increase. Period. More development also means more opportunities for inclusionary zoning and other value capture mechanisms. The takeaway here isn't anything you haven't heard a million times before, but hopefully this framing adds some sense of urgency for cities that are still in the early and middle stages of across-the-board gentrification.

Rental affordability is clearly related to the supply of new housing, captured here by the housing "permits per 1,000 new residents" metric. Graph from Zillow.

San Francisco is a harbinger, but the fate it foreshadows is not an inevitable one, as Tokyo and other cities have demonstrated. Places like Los Angeles and Seattle are on the cusp, and Chicago and Philadelphia are moving in the same direction, but they're all still salvageable. Boston and Washington, D.C. are further along than even LA or Seattle, and they'll have to adopt a regional housing strategy to address housing supply in their regions. This will be considerably more challenging than a go-it-alone approach, but their small size necessitates partnerships with neighboring cities that share responsibility for the affordability and equity of their respective regions.

These and many other cities have significant control over their fates because they represent a much larger geographical footprint in their respective regions, though they've yet to really take advantage of it. They have more of an opportunity to allow for redevelopment while still not growing at an unmanageable rate. Even Seattle, which has done a decent (though still not quite sufficient) job of providing lots of housing to respond to surging demand, has only grown by a few percentage points each year recently—and that's with only a tiny slice of the city's total land available for redevelopment.

As demand cools off in the coming years, homes built during this boom will serve as a buffer against future rent increases and filter down from high-end to middle-income markets. Better still, if we "overbuild" during the boom years, when the current real estate cycle ends—they always do, sooner or later—there is a good chance of rents actually declining somewhat. Not enough to dramatically change the landscape of any city's affordability, but every little bit helps.

Cities change, and for some people that's not good news. Those committed to sticking with their cars might deal with more congestion even as other transportation options improve. Those who love the feel of a largely low-slung neighborhood—or even those who prefer a five-story streetscape to a ten-story one—might have to accept the change and just make the most of the new shops and park space that accompany increased density. Call me selfish, but if it means I don't have to pay the majority of my income on rent, or fight to secure a rent-controlled unit and squat in it for the next few decades just to stay in the city, or watch my neighbors leave the city and never have enough money to come back, I'd say that's a pretty small price to pay.