CP101: Assignment #2

Analyzing Evictions and Displacement in San Francisco

by Daniel Campbell and Ferdie Taruc

Building in the Mission District of San Francisco. Source: https://www.parcbay.com/real-estate/san-francisco/mission-district-san-francisco-homes-for-sale/

Do certain types of evictions in San Francisco lead to lasting gentrification?

As the tech bubble grows in the Bay Area and the housing shortage continues, we hope to explore any relationship that evictions have on the increasing trends of gentrification or displacement in San Francisco.

Types of Evictions:

We used administrative data filed by San Francisco Rent Board to analyze trends in evictions from 1997 to 2020. As noted by the creator of the data set: “A notice of eviction does not necessarily indicate that the tenant was eventually evicted, so the notices below may differ from actual evictions.”

There are quite a few reasons why people get evicted. For clarification, we have described a few of them below.

  • Breach: Any breach of the lease.
  • Nuisance: Creation of a nuisance with the landlord or other tenants in the building.
  • Capital Improvement: Usually temporary to rehabilitate/improve the unit. Tenants have the right to reoccupy the unit when rehabilitation is completed.
  • Good Samaritan Ends: Evicted after period of good Samaritan laws, “a temporary rental for a maximum period of 24 months, after which the tenant may have to move out or pay higher rent.”

In San Francisco, we want to focus on two categories in particular:

  • Owner Move-In (OMI): Owner or relative of owner moves into the unit as their primary place of resident. The owner or relative “must move into the unit within three months and occupy the unit as that person’s principal residence for at least 36 continuous months.” After the three-year period, the owner is permitted to re-rent the unit.
  • Ellis Act Withdrawal: The landlord removes all their units in their building from the rental market, evicting all tenants. The property must claim to be going out of business and cannot control rental units for five years after filing the eviction notice. Under loopholes, it is usually used to convert rental units to condos or convert multi-unit buildings into large single family households like mansions.

These categories are the two most common types of no-fault evictions in San Francisco, meaning that the tenant is evicted through no fault of their own. No-fault evictions also include demolition of the rental unit and capital improvement of the property, though these are significantly less common than OMI and Ellis Act evictions. Since August 10, 2006, San Francisco tenants are entitled to relocation payments from their landlords if they are evicted through a no-fault eviction. The current relocation payment amount is $7,225 per tenant, up to a maximum of $21,674. An additional $4,817 is due for each tenant that is over 60 years old or disabled, as well as for a household with minors. For Ellis Act evictions, the relocation payment amounts are slightly higher.

Through the data, we believe that we might see a spike of these two types of evictions before the installment of the many tech hubs and offices that surround that area. Given that both OMI and Ellis Act evictions are the most efficient in removing tenants from their units, we will explore if areas were gentrified due to these tech companies.

Number of Evictions:

Below, we mapped this data set in Carto, finding that the most common reason for eviction since 1997 is owner move-in. Ellis Act withdrawal is the fourth most common eviction type, behind OMI, breach, and nuisance (these other reasons can be seen by unlocking the reasons widget). Looking at just OMI and Ellis Act evictions, there seems to be a high concentration in the Mission, North Beach, Richmond, and Sunset neighborhoods. We can also see that these evictions saw their peak between around 1997 and 2001, coinciding with the dot-com bubble. The evictions declined sharply around the time of the 2008 financial crisis, before rising in number again since 2013.

Evictions since 1997 in San Francisco. Reasons included can be changed in widget on right side.

If we analyze based on neighborhood, we find that the top three number of total evictions occur in Mission, Tenderloin, and Sunset/Parkside. This follows suit with the active perception that Mission has been generally gentrified over recent years. In the Tenderloin, there is a high concentration of breach and nuisance evictions, while owner move-in and Ellis Act evictions are relatively less common. Both the Mission District and the Tenderloin contain a larger percentage of minorities than the majority of San Francisco, as over 50% of the population is Hispanic or Latino in certain census tracts in the Mission District, while over 30% of the population is African American in parts of the Tenderloin, according to 2018 American Community Survey 5-year estimates.

With more time with analysis, we would have hoped to explore the types of composition of evictions per neighborhood, checking if it was predatory, or if it was just cause due to rising rents and cost of living.

Number of Evictions based on Neighborhood since 1997.

By tracking the number of evictions over time, we see that it remains relatively consistent for all neighborhoods, except for different spikes for South of Market in 2011, 2012, and 2013 and a huge spike at Lakeshore in 2016.

Number of Evictions based on Neighborhood over Time.

Interestingly, Twitter moved its headquarters to South of Market around June of 2012, consistent with the large spike of evictions seen above. Many other companies like Spotify and Square followed suit, where they moved near Twitter’s HQ in 2013. The main reason for these moves was the payroll tax break that Twitter and many other companies were to receive for relocating there.

As areas with a once-high vacancy rate, both SoMa and Mid-Market were revitalized, but this led to a drastic increase of rent prices that negatively impacted surrounding areas like the Tenderloin, an impoverished neighborhood filled with rich diversity. It was once coined as ‘Twitterloin’ due to the lasting effects on it had on the area.

Although we cannot isolate any causal effects of these tech companies moving into both SoMa and Mid-Market on the number of evictions (as other confounders still exist like lingering effects from the recession of 2008), we see that these spikes remain consistent with the story.


Analyzing Buyouts:

Similarly, we sourced administrative data by San Francisco Rent Board that filed both buyout declarations and agreements. “A ‘buyout agreement’ is an agreement wherein a tenant is paid money or given other consideration (for example, a waiver of rent) to vacate a rental unit.” Starting in 2015, the Rent Ordinance mandated that landlords file a Pre-Buyout Disclosure Form before negotiating a buyout with their tenants.

However, take note: Signing a pre-buyout disclosure form does not obligate any tenant to pay anything, as they are not bound to any given negotiation through this form.

Below, we mapped completed buyout agreements between 2015 to 2020. The size of the bubble indicates how large of a payment (in thousands) the tenant received from the buyout. We have color coded the map based on the neighborhood.

Unfortunately, given that the data started from 2015 onward, we cannot track how much buyouts played a role in the initial arrival of the tech boom in San Francisco, especially accounting for Twitter and other major companies’ arrival in the SoMa area.

Completed buyout agreements in San Francisco since 2015. Size of point indicates buyout amount. Color indicates neighborhood.

We see a similar trend from earlier, where the Mission and Sunset/Parkside have a large number of both buyouts and no-fault evictions. Overall, the neighborhoods with the most OMI and Ellis Act evictions seem to coincide with the areas with the most buyout agreements. It seems that these areas might be experiencing a large amount of displacement, as the eviction and buyout data shows that many tenants are being driven out through no fault of their own.

We also notice that the Tenderloin has only 106 total buyouts. Though it has the second-most total evictions of any SF neighborhood, most of these are not due to no-fault reasons. To counter the lack of buyouts, the average buyout in the Tenderloin is the second highest at $130,000, so it is possible that tenants are not willing to leave the area unless they receive high compensation. It could be the case that there are more rent controlled units in this neighborhood.


Buyout Attempts and No-Fault Evictions:

We also wanted to consider the relationship between failed buyouts and no-fault evictions. Over 70% of buyout declarations in 2019 did not result in a buyout agreement. The average buyout amount for completed agreements was $45,000, so it seems that most tenants are unwilling to relocate unless they are paid a relatively large sum.

In general, landlords who are hoping to vacate their units may prefer to reach a buyout agreement with their tenants rather than go through a no-fault eviction. Both OMI and Ellis Act evictions require relocation payments (though these are less than most buyout payments), and they also come with restrictions on the landlord, as noted above. To avoid the three-year residency requirement that comes with OMI and the five-year rental restriction that comes with the Ellis Act, we thought that some landlords might attempt a buyout with their tenants before resorting to one of these evictions.

In the Carto map below, we mapped the instances of buyout declarations that did not result in an agreement (black points) in 2019, as well as 2019 eviction notices for owner move-in and Ellis Act withdrawal (red points). We also included the median gross rent of census tracts, taken from the 2014-2018 American Community Survey 5-year estimates, as the background color for context on the rental makeup of each neighborhood.

Owner move-in and Ellis Act evictions and uncompleted buyouts in 2019. Black dot indicates a buyout declaration that did not result in an agreement. Red dot indicates an eviction notice with reasons specified as OMI or Ellis Act withdrawal. Background color indicates median gross rent in census tract.

We can see that the eviction and buyout points are generally concentrated in the same neighborhoods, with the highest concentration in the Mission District. By zooming into individual points, we can find some instances where it appears that failed buyouts directly led to no-fault evictions. Since the public dataset on eviction notices only includes the block group rather than the address, we cannot confirm that points line up to the same unit, but we can make some inferences based on the proximity of point locations and dates.

For example, in the image below, we can see two instances of failed buyouts lining up with no-fault evictions in the Parkside neighborhood of San Francisco. On Wawona Street, a buyout declaration was filed on March 12th but no agreement was reached, and an Ellis Act withdrawal eviction notice was filed on the same block on May 10th. On the Lower Great Highway, a buyout declaration was filed on January 16th and an owner move-in eviction notice was filed on March 5th.

Two instances of failed buyouts and evictions on the same block in San Francisco.

Though these make up a small number of total evictions, there seems to enough instances where failed buyouts likely resulted in evictions a couple months later that we can conclude that this is a real dynamic occurring in San Francisco. It is very concerning that tenants can be forced out of their homes and apartments through no fault of their own, even after turning down large sums in exchange for moving out. It seems that many landlords may be taking advantage of the eviction rules in order to free up their properties, especially when they have just purchased the property.

When investigating this further, we found that many of these buyouts and no-fault evictions occurred after the building changed ownership. For example, the home at 1132 Alabama St. in the Mission District sold for $1.435 million on January 11, 2019, and a buyout declaration was filed there on January 22nd. There was an eviction notice filed on that side of the block on January 23rd, with the reason being owner move-in. The new owner has just recently sold the property for $1.795 million on February 27, 2020, and there are pictures showing that much of the interior was renovated over the past year. In this instance, it appears that this landlord wanted to capitalize on the hot housing market in San Francisco, so they bought this property with the intention of renovating and flipping it, evicting the tenants in the process. Since a buyout was not agreed upon, it is likely that the tenant was unwilling to move out, but they were forced out through the owner move-in eviction and given a relocation payment that maxes out at $21,674, depending on the number of tenants. Given how expensive this property is, it is likely that this payments would only cover a few months of rent. This is just one example of an eviction after the property changed ownership, but we found many other instances where buyout declarations and eviction notices were filed just months after the property sold, such as 736-738 Natoma St. in SOMA, 82 Woodward St. in the Mission District, and 870 South Van Ness Ave. in the Mission District, to name a few.

Home at 1132 Alabama St. Source: https://www.redfin.com/CA/San-Francisco/1130-Alabama-St-94110/home/2031915

Connection to Displacement and Gentrification:

After determining which areas have experienced the most displacement due to no-fault evictions and buyout agreements, we hoped to see if this lined up with overall trends in the displacement of low income households and gentrification. To make this assessment, we looked into the Urban Displacement Project and their findings for the San Francisco Bay Area. Some indicators for displacement that the UDP looks for are a change in low income households and the loss of affordable housing. For gentrification, they look for a change in income, a change in educational attainment, and rent increases, all relative to the regional median. Below is the Urban Displacement Project’s assessment of the displacement typologies of San Francisco’s census tracts.

The neighborhood with the highest concentration of no-fault evictions and buyouts, the Mission District, is classified as experiencing ongoing gentrification and / or displacement in its lower income tracts and advanced gentrification in its high income tracts. It appears that many lower income households are being displaced in this neighborhood, and are being replaced by higher income people such as tech workers. This is consistent with our findings that the new property owners are using buyouts and no-fault evictions to vacate their properties either to use for themselves or to capitalize on the housing market. The construction of new condos in this area is also associated with the process of gentrification, and can lead to no-fault evictions and buyouts.

Other areas with concerning exclusion statuses according to the UDP are the North Beach area and parts of the Sunset District, both of which have experienced a high number of buyouts and no-fault evictions as well. It seems that no-fault evictions and buyout agreements may be a useful indicator for neighborhood displacement and exclusion.

Conclusions:

Analyzing evictions and buyouts in San Francisco can help us better understand the nature of the displacement and gentrification that is occurring, as well as identify neighborhoods with displacement concerns.

No-fault evictions such as owner move-ins and Ellis Act withdrawals unfairly force out tenants with minimal compensation. With the shortage of housing options in the San Francisco Bay Area, displacing tenants who haven’t done anything to break their lease agreements is an extremely concerning phenomenon. Looking at specific instances where this occurred, we found that many no-fault evictions took place after a new owner purchased the building and after a buyout proposal was already rejected by the tenant. In these cases, it is clear that the tenant is unwilling to relocate, but are forced out against their will and not through any fault of their own.

Since no-fault evictions and buyouts are often used by landlords to convert rental units to upscale condos, they contribute to both the displacement of existing residents and the evolution of the housing stock. Limiting the occurrence of unfair eviction methods and buyouts is an important step to promoting equity and inclusivity in San Francisco. The City of San Francisco should take additional action to close any loopholes surrounding the eviction framework and to prevent landlords from forcing out their tenants for the purpose of flipping the units or converting the units into condos.

Create your website with WordPress.com
Get started