The disappearance of San Francisco’s middle-class neighborhoods, 1990-2010

Over the past 30 years San Francisco’s middle-class neighborhoods have been decimated. In 1990, 60% of San Francisco residents lived in middle-class areas, but by 2010 that number had dropped to 41%. Given current trends, the next census report (in 2020) will almost certainly show an even smaller middle-class in San Francisco. These facts are not surprising, as the tech boom has made San Francisco a poster child for economic inequality. Countless think pieces have been written about the city’s problems with exploding rent, gentrification, and a widening income gap.

I wanted to add to this discussion by creating a series of maps showing how San Francisco’s inequality is distributed across the city. These were inspired by Daniel Kay Hertz fantastic series of maps visualizing inequality in Chicago. The maps below show a city that was solidly middle-class as recently as 1990, but has quickly become sharply divided between homogenous upper-income neighborhoods, and areas that are overwhelmingly low-income. Census block-group data was used, with the median income of each block-group compared to the median income of the San Francisco-Oakland metropolitan area. Low-income areas are colored red, middle-income areas are colored light-grey, and upper-income areas are colored green.

This slideshow requires JavaScript.

Map in gallery form.

In 1990 the map is dominated by a sea of light-grey middle-class areas, but by 2010 the map is much greener with many more upper-income areas. San Francisco has generally seen it’s middle-class neighborhoods transition towards the upper-income range, as opposed to becoming lower-income. In 1990 only 9% of San Franciscans lived in upper-income areas, but by 2010 that number was 32%. In contrast, the number living in lower-income areas declined from 31% to 26%.  Many of the residents of these lower-income and middle-income areas have been displaced to other parts of the Bay Area, or outside the metropolitan area altogether.

image (38).png
The table below shows the change in median income (as a % of the metropolitan area median) by neighborhood from 1990 to 2010, with middle-income values (between 75 and 125% of the metropolitan area median) bolded:

Change 2010 % of Metropolitan Median Income 1990 % of Metropolitan Median Income
The Presidio +79% 163% 84%
Potrero Hill +58% 148% 90%
South Beach / Financial District +55% 144% 89%
Noe Valley +50% 155% 105%
Mission Bay +50% 142% 92%
Lone Mountain/USF +37% 123% 86%
Marina +37% 145% 108%
Castro +36% 137% 101%
South of Market +35% 81% 46%
Haight-Ashbury +33% 128% 95%
Inner Sunset +33% 126% 93%
Hayes Valley +27% 89% 62%
Mission +27% 92% 65%
Pacific Heights +27% 152% 125%
Bernal Heights +27% 116% 89%
Russian Hill +21% 115% 94%
Japantown +20% 67% 47%
Twin Peaks +20% 128% 108%
West of Twin Peaks +20% 170% 150%
Glen Park +19% 132% 113%
Seacliff +18% 224% 206%
Outer Richmond +12% 103% 91%
Presidio Heights +12% 138% 126%
Outer Sunset/ Parkside +10% 110% 100%
Inner Richmond +10% 98% 88%
Outer Mission +9% 110% 101%
Western Addition +8% 68% 60%
North Beach +8% 83% 75%
Nob Hill +7% 75% 68%
Bayview/Hunters Point +5% 70% 65%
Excelsior +3% 100% 97%
Oceanview/Merced/Ingleside +2% 93% 91%
Portola +2% 94% 92%
Tenderloin -4% 28% 32%
Treasure Island -5% 64% 69%
McLaren Park -6% 25% 31%
Chinatown -6% 31% 37%
Visitacion Valley -12% 65% 77%
Lakeshore -14% 68% 82%

Neighborhoods that were middle-income in 1990 saw the greatest changes in their median income. 14 neighborhoods transitioned from being middle-income in 1990 to being upper-income in 2010, including Potrero Hill, the Financial District, the Marina, the Castro, and Inner Sunset, to name a few. A prime example of this trend is Noe Valley, which in 1990 had a median income that was 105% of the metropolitan area median, solidly middle-class. By 2010 Noe Valley had jumped all the way up to making 155% of the metropolitan area median, and was firmly in the upper-class range.

Some middle-income neighborhoods have remained fairly stable, but these neighborhoods are generally located on the fringes of the city, farther from workplaces and public transit. Neighborhoods that were middle-income in 1990 and remained that way through 2010 include Portola, Excelsior, and Outer Mission. The map below shows this pattern, as many of the neighborhoods with small changes in income are located on the southern edges of the city. Very few neighborhoods in the heart of the city have remained middle-class.


Two neighborhoods that were middle-income in 1990 became lower-income by 2010, Visitacion Valley and Lakeshore. These two neighborhoods sit on the border of San Francisco’s city limits, and their decline in relative incomes was likely caused by working-class residents from other parts of the city moving there seeking out cheaper rent.

As for those neighborhoods that were low-income in 1990, there have been a few that  became middle-income by 2010. But these neighborhoods mostly consist of rapidly gentrifying areas like the Mission District and South of Market. So while they have moved into the middle-income category, these neighborhoods are suffering high rates of displacement and internal divisions between wealthier newcomers and older working-class residents. They may not remain middle-income for long, but instead transition into the upper-income category as newcomers make up a greater and greater proportion of these neighborhood’s population.

As for the area’s poorest neighborhoods such as Chinatown and the Tenderloin, they saw their incomes fall slightly from 1990 to 2010 (relative to the metro area). These neighborhoods have been left behind in San Francisco’s boom times, which contributes to the huge gap between the rich and poor in the city. It will be interesting to see if their median incomes rise by the 2020 census due to an influx of young professionals seeking out cheap rent. It’s also possible that these areas may benefit from the city’s low unemployment rate, with a recent report suggesting that San Francisco’s poorest resident have seen some economic gains. Nonetheless, it is disturbing that San Francisco’s tech boom has not increased incomes much for the neighborhoods in the most dire need of a raise.

Lastly, there were three neighborhoods that were already upper-income in 1990, Seacliff, Presidio Heights, and West of Twin Peaks. These neighborhoods all saw moderate increases in their relative incomes. But generally, they did not see as big of rises as areas that were middle-income in 1990. This makes sense, since their rents were already expensive in 1990 and they had less room for their incomes to grow.

Overall these maps and statistics tell the story of a city that is moving farther and farther away from an egalitarian distribution of wealth. As rental prices continue to increase, more and more areas of San Francisco will become unaffordable to the middle-class. The speed of this transformation is terrifying and depressing, but it also means that it was not long ago that San Francisco was significantly more egalitarian. Looking at the 1990 map of San Francisco’s neighborhoods shows a city much more open and available to people of average means. It is important that we don’t let the memory of that older San Francisco disappear.

Data Source:

Minnesota Population Center. National Historical Geographic Information System: Version 2.0. Minneapolis, MN: University of Minnesota 2011.

23 thoughts on “The disappearance of San Francisco’s middle-class neighborhoods, 1990-2010”

  1. Forgive my ignorance, but, I was not aware that Chinatown was one of the poorer areas of town. There seems to be a lot of money there with the political and other influence over The City. Is it a matter of disparity between the upper ranges and lower ranges?


    1. Chinatown definitely has some wealthy and middle-class residents, but the median income is still very low in comparison to the rest of the city. A lot of people are recent immigrants with low incomes, or the elderly who may not be working at all, and many are living in SRO’s.

      A lot of the wealthier middle-class Chinese population lives in other parts of the city, such as Sunset.

      From wikipedia:

      “According to the San Francisco Planning Department, Chinatown is “the most densely populated urban area west of Manhattan”, with 15,000 residents living in 20 square blocks.[17] In the 1970s, the population density in Chinatown was seven times the San Francisco average.[18] The estimated total population in the 2000 Census was at 100,574 residents.[3]

      During the time from 2009 to 2013, the median household income was $20,000 – compared to $76,000 city-wide – with 29% of residents below the national poverty threshold. The median age was 50 years, the oldest of any neighborhood.[19] As of 2015, two thirds of the residents lived in one of Chinatown’s 105 single room occupancy hotels (SRO), 96 of which had private owners and nine were owned by nonprofits.[20]”


      1. Thank you, Nick, I knew there was a broad range between the high incomes and the low incomes, I had no idea it was THAT broad.


  2. The free market is more fair and honest across the board, to everyone equally, than some clumsy attempts to special entitlements to a select few touted falsely as “inclusiveness. There is nothing inclusive about this, it’s actually very exclusive and only for a chosen few.


    1. Hi Bob,

      Are you arguing that San Francisco has become more inclusive with the significant increase in median income?

      I’d love to hear how that works.




  3. This article has some lazy post hoc reasoning in it. Less than 10% of San Francisco workers are employed in the tech industry (Ted Egan, the City Economist said 6% in 2013) but the author implies that this small percentage has driven all the income inequality in The City. This is dubious at best and needs more rigor to be proven. There are still more people employed in Finance or Medicine than Tech in San Francisco, has the author considered what rising salaries in other traditionally well paid sectors of the economy might be contributing to the change?


  4. According to their post %BT% is generally
    well-accepted right away. How can what is this particular to start out extremely vlog and as
    well blog site?


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: