Amid a drumbeat of reports documenting the exodus of lower- and middle-income earners from California’s urban regions, the state’s political leaders finally passed a battery of tenant protections in the recent legislative session. However, marquee legislation aimed at expanding the supply of housing has stalled for the second year in a row. Newly published data from the United States Postal Service (USPS) is a flashing red sign about the state’s housing shortfall: vacancy in the state’s largest housing markets is approaching zero.
Low vacancy rates exacerbate housing affordability issues. Renters compete against one another for available units, home buyers bid above asking in an effort to beat out other offers, and both renters and buyers of modest means lose out. This process is driving a disproportionate number of low-income residents out of California’s cities. While tenant protections can soften the blow, the affordability crisis requires policymakers to do much more work to counteract the fundamental forces pushing up rents.
Our analysis of the USPS data[1] suggests that California cities have extremely low housing vacancy rates — lower even than those reported by other government agencies.
Why be concerned about high vacancy rates?
While many rural and industrial cities face the challenge of very high vacancy, much of California has unusually low vacancy rates as the economy has steadily grown. Notably, California’s major cities see very low vacancy even in comparison to other regions that with strong economic indicators. Low vacancy rates exacerbate housing affordability issues. Renters compete against one another for available units, home buyers bid above asking in an effort to beat out other offers, and both renters and buyers of modest means lose out.
Vacancy data is notoriously noisy, and what is considered a “normal” vacancy rate is dependent on the mobility of a region’s labor market and the composition of its housing stock. For highly dynamic economies in the state’s urban areas, we can imagine that a higher rate of baseline vacancy might help absorb the large number of workers who are moving there for work. As for housing type, we expect owner-occupied housing to be associated with less turnover, and rental housing to see more turnover. Although this dataset does not allow us to differentiate between homeowner and rental vacancy rates, we might infer from this pattern that the owner-occupied homes might have even lower vacancy than what we have reported for the overall residential rate.
Across three booming economic centers, changes over time show sharply decreasing housing availability between the end of the Great Recession and June 2019. The City and County of San Francisco’s gross average vacancy rate fell from 2.4% in 2011 to 1.3% in today. Average vacancy in Alameda County is currently just 0.4%, compared 2% in 2011. During this period, Los Angeles County’s vacancy rate fell from 1.7% to 0.9%.
San Francisco
Alameda County
Los Angeles County
Remarkably, we see this decline in vacant housing across all kinds of census tracts. In fact, this pattern holds in whiter areas as well as in neighborhoods that are mostly comprised of racial minorities. Differences in neighborhood wealth, as measured by median household income, also had little effect on vacancy rates. The whiter and upscale City of Malibu near the edge of Los Angeles County might be the sole outlier; although vacancy rates there have also declined, they remain elevated compared to every other neighborhood in the regions we studied.
Comparing home sales across regional markets
Low vacancy in California is consistent with two other measurements of tightness in the owner-occupied housing market:
● The number of months of supply of homes available for sale
● The median time on market
In the California Association of Realtor’s latest report, there was only a 1.9 month supply of homes for sale and homes sold after only 14 days on the market across Alameda County. In San Francisco, there was only a 2.1 month supply of homes for sale and homes sold after only 15 days on the market. These metrics show far more supply and lower prices in markets with higher vacancy rates:
While the market continues to show signs of short supply, demand for housing in the Bay Area remains strong. Employment in the San Francisco-Oakland region is now well above the pre-recession peak in 2007. The Department of Housing and Urban Development (HUD) estimates a deficit of 48,950 housing units in the region in its three-year forecast period.
Implications for California policymakers
The USPS vacancy dataset corroborates a growing body of evidence which demonstrate that supply is historically tight across California’s largest housing markets. Even as lawmakers have begun taking measures to address the cost of housing through a rent cap, and banning Section 8 discrimination, these reforms do little to fix the simple fact that the need for housing continues to outpace the production of housing.
These data also call into question the efficacy of vacancy taxes, a step that some California cities have already taken. In theory, a tax on vacant parcels could force landowners to rent their empty homes or pursue infill development of underutilized lots, all while raising revenues for local government. The USPS data set does not provide any new insight into the prevalence of undeveloped plots of land, as the data are only collected on existing addresses. An earlier analysis by a Terner Center researcher did find about 4000 vacant parcels in Oakland, most of them only large enough to develop into a single-family home or duplex.
On the question of the existing housing stock, however, the data do suggest that we do not currently have a problem of widespread underutilization. Moreover, with the exception of a few coastal communities on the outskirts of metropolitan Los Angeles, our findings run counter to the claim that large numbers of wealthier property owners are keeping their homes vacant for speculative purposes. Very simply, counts of vacant homes are in decline across all neighborhood income types. Certainly, vacancy taxes could be deployed to incentivize development of empty plots of land, but this analysis shows little opportunity for improving the use of existing homes.
As local and state elected officials take stock of housing market dysfunction, they have no choice but to refocus at ways to unlock additional supply: expanding by-right designations, streamlining a shifting kaleidoscope of rules and fees, and fostering innovative ways to bring down construction costs.
[1] Unlike the vacancy data reported by the US Census and by Department of Housing and Urban Development (HUD), which have a lag of about a year, the USPS data are published every fiscal quarter. This data can be thought of as a more real-time estimate than those provided by other administrative data sources. Furthermore, the USPS data are generated by the Postal Service workforce, providing us with a dataset that results from constant canvassing of the entire nation. Note that although vacancy estimates vary widely across data sources, the USPS data set tracks the other data sources across all the dimensions we studied, including change in geography and change over time.