San Francisco’s Planning Department has completed its first annual “Jobs-Housing Fit” report, a newly required analysis of “the fit between the housing needed by workers of new jobs located in San Francisco and housing produced in San Francisco,” segmented by affordability.

And in fact, based on the pipeline of entitled projects in San Francisco proper, the 2020 report projects that “development from 2020-2040 will meet and exceed overall housing need” by 18,417 units, with the majority of the surplus occurring within the next decade, as summarized above.

That being said, the projected surplus is net of a 3,048 unit deficit in the number of needed affordable units, which includes units affordable to households earning up to 120 percent of the Area Median Income (AMI), which is currently $111,900 for an individual and $127,850 for a household of two.

And while the projected net surplus over the next decade would not offset the deficit in production from 2009 to 2019, relative to the growth in local employment, which drove property values up, it would make a dent.

42 thoughts on “A Projected Surplus in San Francisco Housing Production”
  1. Would love to know when in the last 50 years our “pipeline” of approved projects actually resulted in exactly that many units being constructed. Case in point: Zero of the approved Central SoMa housing projects are under construction three years later. Anyone who pays attention knows that some projects won’t be built & the pipeline is an undercount of units to be constructed, so why doesn’t SF Planning take that into account? So bizarre.

  2. Also if our “unmet need” was 125K units in 2019, why don’t those negative units carry over into the next decade? Don’t those folks still need housing? This makes no sense.

    1. The vast majority of those folks are already housed, which drove up occupancy rates, property values and rents, while pushing others out of the city (or onto the streets). In other words, the impact of the historic “unmet need” has already been realized and/or priced in to the current market.

      1. But to Hunter’s point, that’s essentially saying that the status quo is an acceptable baseline – inflated rents and property values and people pushed out of the City or onto the streets. It’s like saying there were 125K unemployed people in 2019, but the City will have more jobs than bodies in 2030 because we’re going to create 22,000 jobs but only have 18,000 people move into the city – it discounts / assumes away the existing demand.

        So the alternative view is that the acceptable baseline would be affordable housing for at least the middle class, which will not be achieved by the current pipeline (and/or by lopping off the 125K of unmet needs as of 2019).

  3. I mean, that’s certainly one take. The bottom line here is that on the current trajectory there will still be a severe housing shortage – a deficit of over 100k homes – in the year 2040, in other words for the foreseeable future. And that’s not even counting the deficit accumulated from 1980-2010, which was larger than the 2010-present net deficit.

    1. That’s certainly another take. But the impact from the accumulated deficit to date has already been realized and/or priced in to the current market and it’s the net change going forward that really matters and will have the most impact.

      1. that’s a totally obtuse take on this analysis, which is that our crazy market is the result of a cumulative deficit of over 100K units. The future net only matters in isolation if you think our current state of affairs in 2021 is acceptable and the benchmark we should aspire to maintaining.

        1. Speaking for myself, I find the current state of housing market abysmal for anyone but extremely well-heeled, and that reducing the accumulated deficit of 100K units, in addition to making sure future growth doesn’t exacerbate it, is what matters.

        2. That’s incorrect. While understanding how we got here is important, the future net is the only thing that matters going forward, in terms of moving the market, especially if one thinks our current state of affairs is unacceptable and needs to change.

      2. I’m not sure I understand your statement. The impact from the accumulated deficit may be realized in current prices, but that doesn’t mean it isn’t still felt continuously by everybody paying insane rents, and everybody that can’t move here due to prices. 1990 seems like a much better reference to me than 2021 in deciding what the optimal jobs/housing fit actually is. How can 2500+ one bedroom rental prices represents a reasonable balance between supply and demand? Only in SF…

        1. “The impact from the accumulated deficit may be realized in current prices, but that doesn’t mean it isn’t still felt continuously by everybody paying insane rents, and everybody that can’t move here due to prices.”

          That’s correct. But once again, it’s the net change going forward that matters, in terms of moving the market and forecasting. The accumulated deficit is history. And of course, “optimal” and “reasonable” are subjective policy measures, we’re objectively focused on the impact, implications and trends.

  4. @Hunter: As the population of SF was in decline from the early 1950’s until the early 1980’s, when it hit a minimum around 660K, the number of new residential units being built was not a problem until the late 1990’s. It was only in the 1990’s that the population of the City started to increase noticeably again.

    I remember one feature in the Sunday Chron back in the late 1990’s which mentioned in passing that due to the conversion of subdivided houses back to single occupancy the actual number of residential units had fallen in SF since 1990 even though the population had increase in the intervening time by about 40K. Due to changing demographics at the time I would guess a lot of the increase was in-laws and multi generational families in places like the Sunset as the original owners from the 1940’s died / retired and moved.

    The building of new residential units only became noticeable by the mid 2000’s. Mostly in places only incomers or short term City residents would want to live.

    Nothing really changes in SF. And nothing will until the next big earthquake solves the nimby / no new development in my neighborhood problem. When at least 1/3’rd of the City will burn down. Just like last time. All those EV’s doing the job the failed water system did in ’06. And SF’s still incompatible fire hydrants making mutual aid almost impossible. Do they still have the fireboat? Or have they sold it again? The one that prevented most of the Marina burning down in ’89.

    Plus ca change…

    1. 1/3 of the City will not burn down in the next earthquake. Building codes basically didn’t exist in 1906. San Francisco has built multiple redundant ways to deliver water to the fire department. The City will not burn.

      1. Keep in mind that the building codes that were in effect in the 1910s and 1920s when the city was being rebuilt were much weaker than the current codes. Also know that in 1906 almost 100% of firefighters lived in San Francisco. Now the majority of SFFD staff live a bridge or tunnel away.

        Analyses of how a 1906 sized quake will affect SF show that the city will be completely cut off due to failed bridges, tunnels, and even 101 southbound will be eaten by liquefaction. That leaves a single SFFD shift within the city to battle fires for the first week it takes to restore transportation and utilities.

        1. That is not true. The city will not be completely cut off. The Bay Bridge was rebuilt specifically to allow emergency vehicles across after an earthquake because the old bridge collapsed.

          San Francisco has an entire backup set of fire hydrants and fire system specially to protect us from a fire caused by the earthquake. It sounds like you’re not aware of this, but that’s ok. Not many people are. There are also huge underground cisterns all over the city with extra water specifically for fighting any fires. As yet another backup, fire boats can pump salt water into the system. Not to mention that fire trucks today don’t compare at all to what was available in 1906.

          There will be damage from the next earthquake, of course, but the City will not burn.

          1. Not that I agree with this gloom-n-doom premise, but why should we assume aid will (even) be available to come to SF ?? Certainly a major quake will have impact far beyond SF city limits, and other cities’ First Responders will have their own issues to worry about.

            Perhaps the remarkable sense of self-importance of the San Francentric POV was what ‘tf’ meant by plus ca change

          2. Keep in mind that in the Loma Prieta quake (The Medium One) only took the failure of a small 50′ part of the Bay Bridge to close the whole connection for a month. For The Big One the weak points aren’t the main bridge structures but rather the approach ramps which one or both ends lie on liquefaction prone earth. There’s nearly a century of accumulated seismic knowledge acquired that was not available when SF’s main bridges were constructed. Also the E in WETA stands for Emergency. Those ferries are earmarked for disaster duty to take over when the cross-bay connections fail.

            Yes, I know about the cisterns marked by brick circles in the pavement. But all the infrastructure in the world is no good if most of the firefighters are stuck in the east bay. The designers of SF’s firefighting infrastructure didn’t take into account that their staff would be priced out of the city they defend. If they did there would be a several hundred unit “Firefighter’s Colony” on one of the city’s blocks to provide affordable housing for their staff.

          3. Since the vast majority of homeowners do not have earthquake insurance but do have fire insurance, I suspect they will ensure that the city burns.

          4. We are all lucky that you are not the only person who is aware of the importance of the approach ramps to the bridge. This is why the century of knowledge was recently put to good use rebuilding those approach ramps.

            Actually infrastructure matters a whole lot. Improved building codes, both on keeping buildings standing and making them less prone to burning, mean that there will be far fewer fires to begin with. In the Loma Prieta earthquake, there was exactly one large fire in the city, and building codes have got progressively stronger over the last 30 years.

          5. That’s great that the western ramps were upgraded. That leaves the eastern end which was built on the old Key System mole (i.e. landfill), where the toll plaza is sited. I don’t recall any upgrades east of the new eastern span project limits.

          6. There’s only so much of your research that I’m willing to do. Look a little harder.

            I understand that you’re really hung up on the fact that many San Francisco first responders live in the East Bay. Yes, that’s not ideal. But to claim that “Analyses of how a 1906 sized quake will affect SF show that the city will be completely cut off” is obviously false. (Remember, that was your first argument, before you shifted to the first responders one.)

            SF will not be cut off, nor will it burn.

      2. Were you around in ’89? I’ve found that direct experience of a major earthquake makes people take them more seriously.

        You need to read the USGS and FEMA studies. That started in earnest after Loma Preita. Once had a long talk with the lead seismologist on the first big FEMA study. Would have been around 1994 when the USGS etc were still in the Presidio. Went over the maps (I have more than a passing interest in geology) and basically got the unexpurgated version of what did not go in the final report. Also got the unvarnished story of how close the Marina came to an uncontained fire. No fireboat and many blocks would have burned. And this was with a very moderate quake. If we had got the full 45 seconds of shaking of a normal mid M6 (rather than 15 secs) a lot more stuff would have come down and a lot of blocks would have burned. Even with modern “fire codes”.

        The USGS, FEMA, Cal Bureau of Mines (or whatever they are called now) etc reports in the last 25 years really have added little to likely damage scenarios. The 1/3 number I quotes is of total residential units and is the number you will find in the SF City disaster plans as the likely number for the worst case next earthquake scenario. A M7 on the San Andreas segment in San Mateo County. Although if the Hayward pops a M7 and the weather gods are not smiling the numbers in SF could be just as bad. Basically after a big earthquake you only need one small fire to break out to burn down a block. Because there are so many other small fires and the SFFD will take at least 6 to 12 hours to effectively regroup.

        So the City will burn next. Its just a matter how much and for how long. Now ’06 was a special case. As the many lawsuits brought by the insurances companies right afterwards proved. But in the end it burned not because of fire code but because a lot of small fires cannot be contained in a post earthquake environment. Look at what happened in Kobe, Japan in 1995. With very strict fire codes. The majority of fatalities died in fires. And large parts of the city burned.

        Earthquakes happen. So best to be prepared. And be aware of the actual risks.

        1. I do take earthquakes seriously, as does the City of San Francisco.

          Yes, there will be some fires when the big one hits. No, 1/3 of the city will not burn down.

          Yes more of the city would have burned in 1989 if not for the fire boat. But we had a fireboat then and now.

          And if you think our codes have not improved since 1995, you can start your research here:

          1. So you have not been reading the various local earthquake risk /response reports/ papers put out over the last three plus decades then it seems.

            Even with a causal online search the first search return item is a City report that estimates loss of at least 25% of the 330K plus residential units in SF. Thats red tags losses. And thats the best case worse case scenario. Most losses due to loss of structure / foundation integrity and partial collapse like in 1989. No large scale uncontained fires.

            In more detailed discussions of various scenarios a hot spell with high winds situation would mean large scale uncontrolled area fires. Just like 1906. Without the winds, maybe we get lucky. But once those fires start those damaged building soon become burning buildings. Just like in ’06.

            I watched the Marina burning while crossing the GG Bridge in ’89. For a few hours it was a close run thing. The people I knew who were on the ground that evening thought so too. And that was the absolute best case scenario. A few months later and the fire boat would have been in Australia.

            I mentioned the Kobe earthquake in 1995 just to see if you were around at the time. It got very big coverage in the local media back then precisely because despite all those very strict modern seismic and fire codes lots of buildings collapsing and many many blocks burning. A lot of very modern mid rise buildings failed. Not just large parts of the central freeway in the docks. Which made the most dramatic TV footage.

            There was a lot of discussion of these failures in Kobe in the seismic engineering and disaster planning literature. It caused a big shock at the time. As did the near failure of several high rises in LA the year before during the Northridge quake. Just like with the Embarcadero Freeway and 280. Another ten to twenty seconds of shaking and they would have started coming down.

            Lots of good work has been done in the last 30 years but at least in SF the lack of preparedness and the fragility of the infrastructure is in many ways much worse than it was 30 years. For a start there is no Potrero Hill power plant to restart the City power grid. Like it did in ’89. Lets hope the Trans Bay Cable does not fail. No power for weeks to months if it does.

            And the City preparation for dealing with mass displacement of residents is beyond a joke. At least 30 years ago there were public lists of designated local shelters. Last time I looked those public lists are gone. Due to ADA it seems. And the prepositioned disaster supplies like other Bay Area cities have are almost non existent in SF. For purely ideological reasons. This pattern is repeated all though the City disaster prep.

            I suggest you do your research on the subject. Some of us have been reading the academic as well as the general literature for many years. For me it’s mainly been the geology as well as the structural engineering angle that is of most interest. A fascinating subject with a lot of first rate work being done. But as someone living in one of the areas most likely to burn in the worst case scenario I do keep up on the local disaster scenarios literature. For the day when it will happen. As it surely will. Because this is earthquake country.

          2. Thank you for clarifying that San Francisco is earthquake country. No one would have guessed it otherwise.

            Yes, the city will be damaged. But no, 1/3 of the city will not burn. Just making stuff up is not the way to improve preparedness. Even you are now claiming that there will only be major fires with the right wind conditions. In any case, people will die, building will fall down, but 1/3 of the city will not burn.

    2. Nothing really changes in SF…except in the Southeastern portion of the city…

      – Entire Transbay / South Beach set of towers
      – Mission Bay / Chase Center / UCSF
      – Infill condos / apartments throughout Soma, Hayes Valley, Mission, Upper Market
      – Central Subway
      – New Giants ballpark / demolition of Candlestick

  5. It’s hard to believe the median income for a 1 person household is $93,250, particularly as that would seem to include all kinds of elderly/retired and – at the other extreme – young/newly graduated persons. Am I misinterpreting the data ?? (I hope I am, b/c if that’s the claim, it’s frankly suspect)

      1. Thanks (been there, done that…got the figure from there, not by dividing the post’s number by 1.25).

        then I’ll come right out and say it: I don’t believe it…either it’s a typo in the report or there’s an asterisk; and just intuitively, how could the median for a 2-person be only ~$16K greater than a 1 person?

        1. Many two-person households consist of a second person with no earnings, such as children, who are surprisingly widespread!

          1. Yes, I realized that – the stereotypical single mom fits the bill- but it’s hard to imagine it isn’t overwhelmed by two-earner “households”.

            But whatever… what I meant by “(there being) an asterisk”, and why I put household in quotes is b/c the numbers imply the traditional single person household – the kind we saw in ‘Vertigo’ – of younger or older or just unattached persons living independently has largely vanished from SF… at least for all but the wealthy.

            That point is often alluded to, of course, but it seems we seldom get such stark evidence of it.

  6. The table clearly shows a net deficit of almost 50K market rate units. How on earth can you (Socketsite) read this summary to suggest that the report shows a “surplus” of housing?

  7. So we added 200k jobs in the last decade but will only add 100k more jobs in the coming 2 decades. These projected housing estimates are surely going to be spot on! /s

  8. 250,000 out of county commuters come into SF every day for their job. A quarter of a million. It boogle’s the mind to suggest building a few thousand units will result in a surplus jobs housing balance.

    Like most City reports – this draws conclusions completely opposite of the facts presented. It’s no wonder why so many people have such little faith in our City leaderships ability to address our very real social problems.

    The MTC response? Link 21. Forcing the commute shed out to 21 northern California counties as far away as Sacramento and Monterey. This is not progressive planning.

    1. the City’s report did NOT say that building “a few thousand units will result in a surplus”. That was Socketsite’s completely misreading of it. The table above says the opposite: There is a deficit of at least 105K units! Look at the table.

      1. To quote the City’s report, as summarized and included above, “development from 2020-2040 will meet and exceed overall housing need by [18,417] units.” And YES, when the projected production exceeds the projected need, that would be a surplus.

        But as we also wrote, which you might have completely misread or missed, “while the projected net surplus over the next decade would not offset the deficit in production from 2009 to 2019, relative to the growth in local employment, which drove property values up, it would make a dent.”

        And now back to what really matters, the implications and impact…

  9. There will continue to be a jobs/housing imbalance just south of SF proper, and many of those employed there will continue to seek homes in SF.

    1. Excellent point. More than 5 million feet of additional biotech/office space have been proposed for SSF sotuth to Millbrae in the past several months. A massive office complex across from the See’s factory on El Camino as well as several other El Camino proposals. Including housing. SSF is supposedly aware of the housing disparity and is encouraging new housing development throughout the city including turning the single-family homes a few blocks from downtown into multi units complexes. No more single-family housing in SSF.

      Job growth will significantly outstrip SF for the coming decade. The same for Oakland. The key is getting major new housing developments in SSF/San Bruno/Millbrae. Oakland workers won’t choose to live in SF but some north Peninsula workers might. Given almost 11 million feet of biotech/office space has been approved for Brisbane/SSF alone housing must be facilitated. One option is to turn Lennar’s HP/CP project into a completely residential development. Replace proposed office use with condos and apartments.

Leave a Reply

Your email address will not be published. Required fields are marked *