San Francisco Chronicle - (Sunday)
Where declining home values hit hardest in S.F.
Home values are up seemingly everywhere in the Bay Area except San Francisco. Yet while the city is an exception in the region, the “San Francisco exodus” — people leaving the city, causing rents and home values to drop precipitously — has actually been the norm within major U.S. metropolitan areas, according to research.
That is, most U.S. cities have seen widespread migration out of their denser centers into less crowded suburban surroundings.
To understand the extent to which the pandemic affected home values in San Francisco and the rest of the Bay Area, The Chronicle analyzed
Zillow data on monthly median home values for 456 neighborhoods across the ninecounty Bay Area.
While most of the neighborhoods we looked at saw home values increase from January 2020 through April 2021, estimated values declined in 74 of them — 55 of which were in San Francisco. Home values stayed flat or increased in just 14 of the city’s 69 Zillowdefined neighborhoods.
Additionally, San Francisco made up the top 37 neighborhoods in the Bay Area with the steepest declines in home values.
The Tenderloin neighborhood saw the steepest decrease in home values over the pandemic; the neighborhood’s estimated median home value went from about $780,000 at the end of January 2020 to $690,000 at the end of April, a decrease of almost 12%.
Home values sank in the more affluent neighborhoods of Cow Hollow and the Marina as well; both neighborhoods saw value declines of more than 11%, from roughly $3.7 million to $3.3 million, and $2.5 million to $2.2 million, respectively.
San Francisco’s pandemicera exodus and subsequent decline in rents and home values has been covered exhaustively, and the phenomenon has a lot of distinct regional causes, like high rents and high home values to start with, plus a concentration of tech companies with work-from-home policies. But it certainly isn’t the only dense major city that experienced a hollowingout during the pandemic.
In May, economists Arjun Ramani and Nicholas Bloom from Stanford University published a study examining pandemic migration patterns and real estate markets within the 12 biggest metropolitan areas.
“Within large US cities, households, businesses, and real estate demand have moved from dense central business districts (CBDs) towards lower density suburban zipcodes,” the authors wrote. They labeled this phenomenon the “doughnut effect,” as a visual representation of people moving from an urban center to its suburban surroundings.
Nationwide, the study found that the “doughnut effect” had created home value losses of around 15 percentage points in major cities’ densest urban
ZIP codes relative to changes in lessdense surrounding ones.
In particular, the paper found a “striking pattern of outflows” from San Francisco, which it categorizes as the Bay Area’s central business district, to more suburban areas within the ninecounty region.
This strong outflow from San Francisco makes sense, given that the city is a dense urban hub; in fact, it’s the seconddensest city in the U.S. after New York City.
S.F. also has many nearby cities and suburban areas that are less densely populated, which could be attractive to workers who still have to come to the office occasionally but now have more flexible work-from-home arrangements.
“Working patterns post pandemic will frequently be hybrid, with workers commuting to their business premises typically three days per week,” the researchers wrote. “This level of commuting is less than prepandemic, making suburbs relatively more popular, but too frequent to allow employees to leave the cities containing their employer.”
Bolstering this research, The Chronicle previously found a strong relationship between population density and increased outward migration within San Francisco’s ZIP codes. In fact, ZIP codes in the Marina and Cow Hollow, both of which are denser than San Francisco’s median density, and both of which saw home values decrease substantially, had about 150% more moveouts during MarchNovember 2020 compared to the same period in 2019.