CITY AND COUNTY OF SAN FRANCISCO
BOARD OF SUPERVISORS
BUDGET AND LEGISLATIVE ANALYST
1390 Market Street, Suite 1150, San Francisco, CA 94102
(415) 552-9292 FAX (415) 252-0461

Policy Analysis Report
To:
From:
Re:
Date:

Supervisor Gordon Mar
Budget and Legislative Analyst’s Office
Initial Public Offerings and Income Inequality
April 23, 2019

Summary of Requested Action
Your office requested that we review initial public offerings (IPOs) in San Francisco and the Bay
Area over the last ten years along with trends in gentrification and income inequality during
that time. You asked that we provide details and assess the potential impact of anticipated
IPOs in 2019 on housing prices, gentrification, and income inequality in San Francisco, relying
on third party studies recently completed on this topic.
For further information about this report, contact Fred Brousseau, Director of Policy Analysis, at the Budget and
Legislative Analyst’s Office.

Project Staff: Fred Brousseau, Linden Bairey, Elizabeth Brousseau

Executive Summary


At least six high visibility initial public offerings, or IPOs, of San Francisco companies
have occurred or are expected to take place in 2019. IPOs can generate significant
infusions of cash for growing companies and new wealth in a compressed time frame
for IPO company founders, initial investors, and employees. However, this wealth does
not necessarily benefit the entire community. Population, job, and income data show
that income inequality persists in San Francisco even as the City continues to experience
a sustained period of population, job, and wage growth, particularly for highly-paid
technical occupations.



While not the sole cause of income inequality in San Francisco, the occurrence of up to
six IPOs in 2019 will not improve San Francisco’s situation by adding a one-time injection
of wealth to many well-paid workers, investors, and founders and likely experiencing
some level of increase in housing prices.

Initial Public Offerings


An initial public offering (IPO) is the process of offering shares in a private company to
the public for the first time. Going public is a costly and time consuming process but can
provide companies with a significant source of capital to expand their business. IPOs
Budget and Legislative Analyst
1

 Memo to Supervisor Mar
April 23, 2019

may also generate significant wealth for a company’s founders and early private
investors.


The following companies have or are expected to execute IPOs in 2019. Although
estimated numbers of employees in San Francisco are not available for every company,
the information that is available shows that an estimated 7,990 employees work in San
Francisco. Depending on companies’ equity compensation practices, many of these
employees could be eligible to exercise stock options over the next one to two years.

Exhibit A: San Francisco Companies with Expected or Actual IPOs in 2019
Company

Valuation

Airbnb
Lyft*
Pinterest*
Postmates
Slack
Uber
Total

$35 bn.
$24 bn.
$12.7 bn.
$1.85 bn.
$7.1 bn.
$100 bn.

Estimated #
employees in SF
2,300
1,600
unknown.
340
750
3,000
7,990

*IPO has occurred; valuation at IPO. Others estimated by various sources and reported in
media.
Sources: Number of employees reported directly in SF Business Times or indirectly through
other information about leases or facilities.



Private companies do not disclose the number or percent of shares that employees are
awarded through stock options or restricted stock units. However, industry practice and
company valuations suggest that some of the employees at the subject companies have
or will receive stock options or other equity-based compensation that will, in some
cases, provide them with sufficient cash to make purchases such as housing in San
Francisco.



At least three recent studies by academics and real estate economists have found that
real estate prices increase in areas near IPO company headquarters, after accounting for
increases that would have occurred absent the IPO.



We applied the most conservative average increase per IPO found in the three studies
(1.8 percent) to San Francisco’s $1.3 million median sale price for a home to estimate
the possible impact of up to six high visibility IPOs that have occurred or could occur in
2019.



Applying the average 1.8 percent increase, we conclude that the impact on sale prices
for a home in San Francisco could range from $23,476 if only one IPO occurs to
$147,346 if all six occur, as shown in Exhibit B. The higher estimate would represent a
11.3 percent increase in the median sale price for a home in San Francisco. This increase
would be spread over one to two years, depending on when employees would be

Budget and Legislative Analyst
2

 Memo to Supervisor Mar
April 23, 2019

allowed to exercise their options (the lock-up period) and choose to cash out their
shares.
Exhibit B: Impact of Six Expected IPOs on Median Housing Sale Price in San Francisco
Median Value $ change
Actual: Feb. 2019
after 1st IPO
after 2nd IPO
after 3rd IPO
after 4th IPO
after 5th IPO
after 6th IPO

$ 1,304,200
1,327,676
1,351,574
1,375,902
1,400,668
1,425,880
1,451,546

Total

% change

$23,476
23,898
24,328
24,766
25,212
25,666

1.8%
3.6%
5.5%
7.4%
9.3%
11.3%

$ 147,346

11.3%



The rate of increase applied in Exhibit B is an average and, therefore, the actual impact
could vary depending on which companies go public. The full impact shown depends on
all six companies going public, with the impact of the smaller companies balancing out
the impact of the larger ones to arrive at the average 1.8 percent impact each.



Companies with larger valuations, more employees, and higher share prices would likely
have a larger impact than the 1.8 percent rate applied. Conversely, the impact of smaller
companies with fewer employees and lower share prices could have an impact of less
than 1.8 percent.
Income Inequality in San Francisco



1

Data on jobs, income, housing, and related measures shows that San Francisco and the
Bay Area region is experiencing an increase in income inequality. The Brookings
Institution reports that San Francisco had the sixth highest income inequality rating of
the largest 100 cities in the country. Their ranking is based on the fact that the $567,824
household income at the 95th percentile of all San Francisco households is 15.9 times
greater than the $41,820 San Francisco household income at the 20th percentile. As
shown in Exhibit C, this multiple worsened between 2014 and 2016, increasing to 15.9
from the 14.5 in 2014.1

th

The 20 percentile represents the income amount that 20 percent of all San Francisco households earn less than,
th
and 80 percent of all San Francisco households earn more than. The 95 percentile represents the income amount
that 95 percent of all San Francisco households earn less than, and five percent of all San Francisco households
earn more.
Budget and Legislative Analyst
3

 Memo to Supervisor Mar
April 23, 2019

Exhibit C: Differences between 20th and 95th Percentile San Francisco Household
Income, 2014 and 2016
Household Income
2016
2014
% increase

20th percentile 95th percentile Multiple
$31,840
$567,824
15.9
$26,730
$388,491
14.5
19.1%
46.2%

Source: Brookings Institution



A more recent 2019 analysis of inequality prepared by Bloomberg of the 100 largest
metropolitan areas in the United States found that San Francisco has the widest gap
between extremely high earners and “middle-class” earners based on 2017 Census
data.2 Bloomberg’s analysis found that the top 5 percent of San Francisco households
earned $632,310 in 2017, compared with $102,785 for middle-class households.



As shown in Exhibit D, the number of housing units did not keep up with population and
job increases in San Francisco between 2010 and 2017. Median house sale prices
increased by 60.7 percent during the same time period.
Exhibit D: Changes in San Francisco Population, Jobs, Housing Units and Median Home
Price, 2010-2017
2010
2017 Change % Change
Population
805,235
884,263
79,028
9.8%
Civilian employed population 444,628
504,914
60,286
13.6%
Housing units
376,942
390,376
13,434
3.6%
Median house price
$697,700 $1,120,900 $423,200
60.7%


Between 2012 and 2017, jobs and wages increased in San Francisco at both the high and
low ends of the wage scale. However, there were more jobs added at the high end and
the rate of wage increase was higher for those earning over $40 an hour than those
earning $25 or less per hour, as shown in Exhibit E.

Exhibit E: Differences in job and wage growth in San Francisco based on hourly wages,
2012-2017

Occupations $25/hr & under in 2017
Occupations over $40/hr. in 2017
Multiple: highest:lowest
Total Jobs

2

Mean
Employment
Employment
Hourly
Estimate May '12 Estimate May '17 % Chg. Wage '12
329,410
376,100
14% $ 18.27
267,280
336,910
26% $ 49.05
4.2
1,000,430
1,116,390
12%

Mean
Hourly
Wage '17 % Chg.
$ 21.08
15%
$ 59.02
20%
5.2

Middle-class earners are defined as the middle 20 percent of earners, or third quintile.
Budget and Legislative Analyst
4

 Memo to Supervisor Mar
April 23, 2019



The two fastest growing occupation categories in the San Francisco-Redwood CitySouth San Francisco area between 2012 and 2017 are the relatively high paid
Computer and Mathematical occupations (e.g., software developers, systems
analysts, etc.) and the lowest-paid Personal Care and Service occupations (e.g.,
personal care aides, childcare workers, etc.), in which the number of jobs increased
by 107 percent. Employment increased by 48 percent for Computer and
Mathematical occupations and by 107 percent for Personal Care and Service
occupations. However, as shown in Exhibit F, wages increased by 24 percent for
Computer and Mathematical occupations but decreased for Personal Care and
Service occupations by six percent.

Exhibit F: Differential in Mean Wages: Computer and Mathematical vs. Personal Care and
Service Occupations, 2012 and 2017: San Francisco-Redwood City-South San Francisco
2012
Mean Annual Wages
Computer and Mathematical Occupations
Personal Care and Service Occupations
Multiple: difference in wages


$102,096
$33,942
3.0

2017
$126,812
$31,743
4.0

% Change
24%
-6%

One of the results of the changes discussed above in San Francisco appears to be
displacement of lower income households. As shown in Exhibit F, households earning
less than $100,000 per year made up 63 percent of all San Francisco households in 2010.
By 2017, these households had decreased to 51 percent of all households. The largest
increase in households were those with incomes of $200,000 or more, which nearly
doubled from 12 percent of all households in 2010 to 21 percent in 2017.

Budget and Legislative Analyst
5

 Memo to Supervisor Mar
April 23, 2019

Exhibit G: Household Income Groups in San Francisco, 2010 and 2017

Source: 2006-2010 American Community Survey 5-Year Estimates and 2013-2017
American Community Survey 5-Year Estimates



Research by Karen Chapple, et al. found that in the Bay Area, housing price
appreciation has a strong relationship to the location of start-up jobs, particularly
those that provide information-intensive services.3 Chapple et al. argue that
because the occupational patterns of information-intensive services are skewed
toward the high end of wages, prices that employees can pay for housing are
inflated, and expansion of these firms therefore creates a disproportionate share of
highly skilled, highly paid workers into the local housing markets.



This pattern identified by Chapple et al. has occurred in San Francisco, as presented
above, with relatively faster growth in recent years of more highly paid workers in
San Francisco’s information-intensive businesses. Though San Francisco real estate
can still be costly for these highly paid workers, they are better positioned to buy
real estate in the San Francisco market based on their higher salaries, and become
even more able if they have stock options to exercise if and when their employers
go through an IPO.

Policy Options
While not the sole cause of income inequality in San Francisco, the occurrence of up to
six IPOs in 2019 will add a one-time injection of wealth to many well-paid workers,
investors, and founders further increasing the wealth gap and likely causing some
3

Karen Chapple, John V. Thomas, Dena Belter, and Gerald Autler, “Fueling the fire: Information technology and
housing price appreciation in the San Francisco Bay area and the Twin Cities,” 2004, Housing Policy Debate 15(2):
347-383.
Budget and Legislative Analyst
6

 Memo to Supervisor Mar
April 23, 2019

increase in housing prices. Some policy approaches the Board of Supervisors could
consider to address and mitigate the effects of income inequality are to:
1. Enhance funding for existing City workforce development, economic development,
affordable housing, and subsidized childcare programs, particularly those aimed at
the low and moderate income workforce.
2. Enhance City programs aimed at financial empowerment, such as those
administered by the Treasurer-Tax Collector including Kindergarten to College, Bank
on SF, and Smart Money Coaching.
3. Enhance taxes such as the City’s payroll or gross receipts tax for companies above a
certain size and/or gross receipts level, with the proceeds used for programs and
services designed to minimize the impacts of income inequality.
4. Establish a gross receipts surtax on companies with differentials between CEO and
average worker pay above a certain level, with the proceeds used for programs and
services designed to minimize the impacts of income inequality.
i. The City of Portland, Oregon has established such a surtax.
5. Establish other taxes on large or high-value companies in San Francisco, with the
proceeds used for programs and services designed to minimize the impacts of
income inequality. Some examples could include:
i. In November 2018, East Palo Alto voters approved an annual parcel tax of
$2.50 per square foot on commercial office property over 25,000 square
feet, with tax revenue reserved for spending on affordable and supportive
housing programs and programs that facilitate residents’ access to jobs in
the science, technology, engineering, and mathematics sectors and the
building trades.
ii. Also in November 2018, Mountain View voters approved an annual peremployee business license tax, with larger companies paying more per
employee. (Revenue from this tax is considered unrestricted general
revenue.)
iii. In 2018, the Seattle City Council passed a “head tax” ordinance that would
have levied a $275 per full-time employee4 tax on Seattle businesses that
have at least $20 million in annual taxable gross revenue. Tax revenue
would have been spent on affordable housing and services for Seattle’s
homeless population. The Seattle City Council later repealed the head tax
legislation.

4

The proposed tax, formally an Employee Hours Tax, would have applied to all Seattle employees at a rate of 26
cents per hour for 2019 and 2020, with the annual tax per employee capped at $275.
Budget and Legislative Analyst
7

 Memo to Supervisor Mar
April 23, 2019

6. Provide incentives or assistance to households under a certain income level to enroll
in retirement plans.
7. Lobby for changes to federal and state tax policies.

Budget and Legislative Analyst
8

 Memo to Supervisor Mar
April 23, 2019

Background on Initial Public Offerings and Equity Compensation
Initial Public Offerings
An initial public offering (IPO) is the process of offering shares in a private
company to the public for the first time. IPOs are used to raise money for the
company, to monetize the investments of the company’s early investors and
founders, and to facilitate trading of existing holdings. IPOs are underwritten by
investment banks, such as Morgan Stanley and Goldman Sachs; these banks also
help determine the value of the company’s shares and assist the company in
establishing a public market for shares.
Before an IPO, the company drafts an initial prospectus, also known as an S-1, and
files it confidentially with the U.S. Securities and Exchange Commission (SEC).
Once the SEC and the company agree on any necessary edits to the S-1 and
disclosures that need to be added, the preliminary registration statement is made
public; however, at that point, the prospectus only contains a potential range for
the amount and pricing of the offering.
After meeting with potential investors, the investment banks and the company
formally set the price at which shares will be offered and the total number of
shares that will be made available for purchase. The company then files a final
registration statement with the SEC, usually the evening before shares can
officially be traded. The following morning, the SEC approves the final terms of the
offering, and shares are formally offered for sale and the company becomes
public.
Once a company is public, employee-owned shares are typically restricted from
being sold for the first three to six months from a company’s IPO date, in order to
prevent individuals who already have company stock or stock options from
flooding the market with shares and depressing the company’s post-IPO share
price. This period of time is also called the employee lock-up period.

Stock Compensation in the Technology Industry
After a technology startup company is founded, a number of the company’s
shares are commonly reserved for an employee equity incentive plan. A
technology company employee may be allocated equity-based compensation
from the plan as part of his or her compensation package. Generally speaking, the
amount of equity allocated to an employee will depend on his or her role in the
company as well as when he or she was hired, with early and more senior
employees allocated proportionally larger percentages.
The form of equity compensation depends on the company, and can vary
significantly. Stock options or restricted stock units, defined below, are two of the
Budget and Legislative Analyst
9

 Memo to Supervisor Mar
April 23, 2019

most common ways technology companies grant equity to employees. Awards of
stock options, restricted stock units, and other equity compensation are almost
always subject to a vesting schedule5 and other restrictions imposed by the
company.


Stock options are contracts that allow employees to buy a specified
number of shares in the company at a fixed price (also known as the strike
price or exercise price).6 For example, as part of a compensation offer, a
company might offer a potential employee the option to purchase 50,000
shares at $4 per share.



Restricted stock units, or RSUs, are an agreement by a company to issue
an employee shares or stock or the cash value of shares of stock at a
future date (the settlement date). Each unit represents one share of stock
or the cash value of one share of stock that the employee will receive on
the settlement date, which may be the time-based vesting date or a later
date based on, for example, the date of the company’s IPO.

History of IPOs in the Bay Area and San Francisco
Initial public offerings have been an important aspect of the Bay Area business
climate for decades and have been associated with some dramatic waves of wealth
creation and economic growth. However, income and job data for San Francisco and
the region, detailed further below, show that not all residents have benefitted from
the wealth creation during this period. While IPOs are not the sole cause of such
problems, increases in San Francisco’s housing costs, congestion, and income
disparity have occurred simultaneously with significant gains for some of San
Francisco’s residents who have benefitted greatly from economic growth.
IPOs can provide a major supply of capital to growing companies, but they are
looked upon unfavorably by some entrepreneurs because becoming a public
company is very costly, involves complying with more public reporting and
regulatory requirements, and, some believe, requires founders to give up control of
aspects of their company. Other sources of capital for startup and private
companies include private venture capital investing and selling company ownership
to another firm. These approaches can also generate wealth for company founders
and employees.
IPOs have gone in and out of favor for Silicon Valley companies and nationally over
the last few decades. Between 2016 and 2018, the number of IPOs in the United
5

Under a vesting schedule, employees gain rights to their award of equity incrementally over time.
Stock options are only exercisable during the exercise window, or the period of time during which an individual
can buy the shares at the strike price.
6

Budget and Legislative Analyst
10

 Memo to Supervisor Mar
April 23, 2019

States increased and, for 2019, the number of high visibility IPOs expected in San
Francisco has increased.
The exact number of IPOs that have taken place in the Bay Area over time is not
readily available from government sources, but national and statewide level
statistics are available from private sources for approximations. According to
Renaissance Capital, a private IPO tracking firm, there were 1,372 IPOs between
2010 and 2019 (year-to-date) nationally, for an average of 167 per year (excluding
the partial year data for 2019). As shown in Exhibit 1, the most IPOs occurred in
2014 between 2010 and 2019, though the trend has also been upward for three
successive years starting in 2016.
Exhibit 1: Number of IPOs in U.S.: 2010-2019

Source: Renaissance Capital
*2019 data as of April 2019

For California, a study by three business and finance professors who analyzed the
impact of IPOs on housing prices, discussed further below, obtained IPO data from a
private financial services research firm for their analysis. They report that that were
1,987 IPOs in California between 1970 and 2017, or an average of 45 per year.7 It is
reasonable to assume that a high proportion of those IPOs would have been
companies in the Bay Area and the Los Angeles area, the state’s two major
population centers. In 1999, geographer Richard Walker reports that approximately

7

Barney Hartman-Glaser, Mark Thibodeau, Jiro Yoshida, “Cash to Spend: IPO Wealth and House Prices”, November
30, 2018. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3329651
Budget and Legislative Analyst
11

 Memo to Supervisor Mar
April 23, 2019

one fourth of all IPOs nationwide were for Bay Area companies.8 Applying Walker’s
estimate to Renaissance Capital’s reported average of 167 IPOs per year nationally
results in an estimated average of 42 IPOs per year for Bay Area companies between
2010 and 2018.
A unique feature of growth and wealth creation in Silicon Valley in the 1960s and
1970s was the establishment of local funding sources particularly aimed at serving
electronics and high technology company startups. The establishment of venture
capital firms and investment bankers located primarily in the Bay Area brought
funding to a number of startup companies. The local presence of the venture capital
firms and investment banks, along with their familiarity with the industry,
represented a departure from the traditional New York-focused investment and
financial services industry and fueled significant growth of Bay Area companies. By
the end of the 1990s, 800 venture capital firms were established in the Bay Area,
mostly oriented to the high technology industry including its then current iteration:
the dot.com boom.9
Three investment banking firms were established in San Francisco in the 1980s that
specialized in assisting high technology and other companies in Silicon Valley grow
and, in some cases, go public: Robertson & Stephens, Hambrecht & Quist, and
Montgomery Securities. All three firms were eventually purchased by large
international banking firms, making even more resources available for Bay Area
company IPOs. Total venture capital investment in the U.S. in 2000 was valued at
approximately $100 billion, of which approximately one-third, or $33 billion, was
invested in Bay Area companies.10

Dot.com boom
The late 1990s was a significant period for IPOs in Silicon Valley. Netscape, creator
of an early web browser, went public in 1995, and its IPO is considered a key
development in the history of Silicon Valley IPOs and an inspiration to many
technology company startups and entrepreneurs. Only one year old at the time and
at no point having achieved profitability, the company offered shares at $28 before
its IPO, but on its opening day, the stock price closed at $58.25 after rising to a high
of $74.75 earlier in the day. Its per-share price was $175 by the end of 1995. While
it was unusual at the time to go public before achieving profitability, this practice
would become more common in future years, particularly for the technology
industry.11
8

Richard Walker, “Boom and Bombshell: New Economy Bubble and the Bay Area”. Found SF.org. Originally written
for The Changing Economic Geography of Globalization, Ed., Giovanna Vertova, Routledge 2005.
9
Ibid.
10
Ibid.
11
Netscape: Web Browser, Wikipedia.com
Budget and Legislative Analyst
12

 Memo to Supervisor Mar
April 23, 2019

Netscape started a wave of IPOs and wealth creation through the late 1990s. Like
many technology companies, Netscape listed on the NASDAQ exchange rather than
the New York or American stock exchanges. NASDAQ grew significantly during the
late 1990s: its index grew from approximately 500 in 1990, to 1,000 in 1995, to
more than 5,000 by the year 2000. Many of the companies listed on NASDAQ were
Bay Area companies.12
During the second half of the 1990s, commercial and residential real estate prices
and vacancy rates in the Bay Area reflected the wealth creation of Silicon Valley,
fueled in part by venture capital and IPOs. Commercial and residential real estate
prices climbed, becoming among the most costly in the nation.

Dot.com decline
The expansion of the dot.com era came to a halt in the early 2000s, with a sharp
drop in the stock market, an exodus of people from the Bay Area who had come to
start or work in dot.com companies, and a decline in real estate prices and rents.
After the markets recovered, the Great Recession occurred starting in
approximately 2008. However, within a few years of the recession, trends of the
1990s reemerged, with extensive job and wealth growth in the technology industry,
fueled in part by venture capital and IPOs. This wave of growth affected San
Francisco as well as the traditional Silicon Valley communities in Santa Clara and San
Mateo counties.

Income Inequality in San Francisco: Changes in Jobs, Population, and Housing
Costs after 2010
The post-recession years in San Francisco have seen strong growth in population,
jobs, and housing prices, and strong but unequal growth in income. As shown in
Exhibit 2, population and jobs grew in San Francisco between 2010 and 2017 (most
recent year available for all data). During the same period, however, housing
production did not keep up with population growth, adding upward pressure to
already high housing costs. Reflecting this, the median home values grew by 60.7
percent between 2010 and 2017, and have continued to increase in 2018 and 2019.

12

Ibid., Walker, R.
Budget and Legislative Analyst
13

 Memo to Supervisor Mar
April 23, 2019

Exhibit 2: Changes in San Francisco Population, Jobs, Housing units and Median
Home Price, 2010-2017
2010
Population
Civilian employed population
Housing units
Median house price*

805,235
444,628
376,942
$697,700

2017

Change

884,263 79,028
504,914 60,286
390,376 13,434
$1,120,900 $423,200

%
Change
9.8%
13.6%
3.6%
60.7%

Sources: United States Census and American Community Survey 5-Year Estimates.
*Median house price: Zillow.com, San Francisco Median Sale Price (all homes): January 2010
and January 2017

As discussed further below, jobs and wages have increased in recent years for highincome earners in San Francisco, but middle- and low-income earners have not
experienced wage increases and opportunities at the same rate. When combined
with increased housing costs and changes in income distribution in San Francisco
during this period, with a greater share of households in the top quintile of income
distribution and more able to afford high priced housing, the City has become much
less affordable for those in lower income brackets.
High visibility IPOs occurred for companies such as Facebook, Twitter, and Square
between 2010 and 2017 when median housing prices increased by 60.7 percent.
While IPOs are not the sole cause of housing price increases, which depend on a
number of complex factors in the real estate market, at least three recent studies
have found that IPOs have impact on housing prices near IPO company
headquarters. These studies are discussed in more detail later in this report.

Household Income
Between 2010 and 2017, the number of households in San Francisco with incomes
less than $100,000 declined, despite an increase of 7 percent in total households in
San Francisco. During the same time period, the number of households with
incomes between $100,000 and $199,999 increased by 18 percent and the number
of households with incomes over $200,000 nearly doubled, as shown in Exhibit 3
below. The growth in higher income households has meant that there are more
residents better able to buy San Francisco’s high-priced real estate, making it more
difficult to do so for remaining lower income residents. The decrease in households
making less than $99,000 per year suggests that some of these households may
have been displaced between 2010 and 2017.

Budget and Legislative Analyst
14

 Memo to Supervisor Mar
April 23, 2019

Exhibit 3: Changes in Household Income in San Francisco, 2010 and 2017
Households earning:
Less than $35,000
$35,000 - $99,999
$100,000 - $199,999
$200,000+
Total households
Median income ($)

2010
93,487
119,691
83,504
39,274
335,956
$ 71,304

2017
83,489
100,396
98,812
76,075
358,772
$ 96,265

Pct. Chg
-11%
-16%
18%
94%
7%
35%

Source: 2006-2010 American Community Survey 5-Year Estimates and
2013-2017 American Community Survey 5-Year Estimates

As shown in Exhibit 4 below, households earning less than $100,000 made up
approximately 63 percent of San Francisco households in 2010. By 2017, these
households decreased to approximately 51 percent of all San Francisco households.
Again, the largest increase occurred in households earning more than $200,000:
these households made up 21 percent of San Francisco households in 2017, up
from 12 percent in 2010.
Exhibit 4: Household Income Groups in San Francisco, 2010 and 2017

Source: 2006-2010 American Community Survey 5-Year Estimates and 2013-2017
American Community Survey 5-Year Estimates

The Brookings Institution tracks income inequality trends for the 100 largest
metropolitan areas in the U.S., including the City of San Francisco and the Bay Area.
In its 2018 reporting on this topic, San Francisco was reported to have the sixth
highest income inequality level of the largest 100 cities in the country in 2016. San
Budget and Legislative Analyst
15

 Memo to Supervisor Mar
April 23, 2019

Francisco’s 95th percentile household income of $507,824 was 15.9 times greater
than the $31,840 household income at the 20th percentile.13 San Francisco’s 2016
differential of 15.9 reflects an increase from the 2014 differential of 14.5, as shown
in Exhibit 5, indicating a worsening of income inequality in San Francisco over the
two-year period.
Exhibit 5 also shows that the rate of increase in household income at the 95th
percentile between 2014 and 2016 (46.2 percent) was substantially greater than
the growth in household income at the 20th percentile (19.1 percent).
Exhibit 5: Differences between 20th and 95th Percentile San Francisco Household
Income, 2014 and 2016
Household Income
2016
2014
% increase

20th percentile 95th percentile Multiple
$31,840
$567,824
15.9
$26,730
$388,491
14.5
19.1%
46.2%

Source: Alan Berube, City and Metropolitan Income Inequality Data Reveal Ups and Downs
through 2016. Brookings Institution: brookings.edu. February 5, 2018.

According to a Bloomberg analysis of 2017 Census data,14 of the 100 largest
metropolitan areas in the United States, San Francisco has the widest gap between
extremely high earners and “middle-class” earners.15 Bloomberg’s analysis finds
that the top 5 percent of San Francisco households earned $632,310 in 2017,
compared with $102,785 for middle-class households. Bloomberg also found that
San Francisco has the widest gap between the highest 20 percent income group
and the lowest 20 percent income group.

Wages and Occupations
While the income distribution of San Francisco households has changed, skewing
upwards over the past several years, there are still a substantial number of lowwage jobs in the City and adjacent communities. In fact, the fastest growing
occupational classification between 2012 and 2017 reported for the San FranciscoRedwood City-South San Francisco area by the California Employment
Development Department was Personal Care and Service occupations (e.g.,
personal care aides, childcare workers, etc.), with 54,850 jobs in 2017, an increase
13

th

The 20 percentile represents the income amount that 20 percent of all San Francisco households earn less
th
than, and 80 percent of all San Francisco households earn more than. The 95 percentile represents the income
amount that 95 percent of all San Francisco households earn less than, and five percent of all San Francisco
households earn more,
14
Shelly Hagan and Wei Lu, “The Income Gap is Getting Worse in American Cities,” Bloomberg, March 22, 2019.
https://www.bloomberg.com/news/articles/2019-03-22/from-boise-to-nyc-uber-rich-are-leaving-the-middle-classbehind
15
Middle-class earners are defined as the middle 20 percent of earners, or third quintile.
Budget and Legislative Analyst
16

 Memo to Supervisor Mar
April 23, 2019

of 107 percent since 2012. This occupation category also had the lowest annual
mean wages reported for 2018, at $31,743 annually, or $15.26 per hour. The
second fastest growing occupational group between 2012 and 2017 was Computer
and Mathematical occupations (e.g., software developers, systems analysts, etc.),
with 93,670 jobs in 2017 and one of the highest annual mean wages reported at
$126,812, or $60.97 per hour.
Besides the gap in mean wages between the two fastest growing occupation
groups, the differential in wages between the two groups also increased during the
period reported. As shown in Exhibit 6, the difference between mean wages for
Computer and Mathematical occupations and Personal Care and Service
occupations increased from 3 times greater in 2012 to 4 times greater in 2017.
Further, as can be seen in Exhibit 6, unlike all other occupation categories, mean
annual wages for Personal Care and Service occupations decreased between 2012
and 2017 by six percent, whereas Computer and Mathematical occupation wages
increased by 24 percent.
Exhibit 6: Differential in Mean Wages: Computer and Mathematical vs. Personal Care and Service
Occupations, 2012 and 2017: San Francisco-Redwood City-South San Francisco
2012
Mean Annual Wages
Computer and Mathematical Occupations
Personal Care and Service Occupations
Multiple: difference in wages

$102,096
$33,942
3.0

2017
$126,812
$31,743
4.0

% Change
24%
-6%

Source: California Employment Development Department

Budget and Legislative Analyst
17

 Memo to Supervisor Mar
April 23, 2019

A report prepared for the Bay Area Regional Prosperity Plan Housing Working
Group by Alex Karner and Chris Benner16 found that there was a close association
between high-wage and low-wage job growth in San Francisco between 2008 and
2011. In fact, the study shows that San Francisco was responsible for the largest
growth in low-wage jobs in the Bay Area, but there was no net increase in the
number of affordable housing units, or the types of units that would be accessible
to low-wage workers. The result for some workers has been displacement, as
individuals who can no longer afford housing prices and rents are forced to move
out their neighborhoods or the City as a whole. Others remain but in an
increasingly unaffordable city and region.
SPUR’s 2014 Economic Prosperity Strategy report17 states that income inequality
has risen sharply in the past decade largely due to the decline in middle-wage jobs
and the stagnation of wages for many workers, plus a rapid increase in wealth for
workers and households at the top of the income spectrum. SPUR concludes that
the Bay Area’s driving economic sectors are increasingly split between high-skill,
high-wage jobs (in industries like technical and professional services) and low-skill,
low-wage jobs (in industries like hospitality, childcare, and retail).
The SPUR report’s conclusions are supported by Economic Development
Department data for the top 10 occupation categories by estimated employees in
the San Francisco-Redwood City-South San Francisco area, presented in Exhibit 7.
As can be seen, the top ten occupation categories contain a high number of jobs
earning both the highest and lowest wages. Overall job growth and wage increases
have occurred at both the high and low ends of the wage scale but both the
increase in the number of jobs and rate of increase in wages was greater for high
wage occupations.
Mean hourly wages in the top ten categories in 2017 ranged from $15.29 for
Personal Care and Service occupations at the low end to $79.77 at the high end for
Management occupations. Decreases in the number of jobs have occurred in the
middle of the wage range including Office and Administrative Support occupations,
Sales and Related occupations, and Education, Training and Library occupations.

16

Alex Karner and Chris Benner, “Job growth, housing affordability, and commuting in the Bay Area,” prepared for
the Bay Area Regional Prosperity Plan Housing Working Group, May 29, 2015.
17
SPUR, “Economic Prosperity Strategy,” 2014.
https://www.spur.org/publications/spur-report/2014-10-01/economic-prosperity-strategy
Budget and Legislative Analyst
18

 Memo to Supervisor Mar
April 23, 2019

Exhibit 7: Changes and Mean Hourly Wages for Top Five and Bottom Five Occupation
Categories by Earnings, 2012-2017, San Francisco-Redwood City-South San Francisco
Employment Estimate

Occupation Categories
Management
Computer and Mathematical
Healthcare Practitioners and Technical
Business and Financial Operations
Education, Training, and Library
Sales and Related
Transportation and Material Moving
Office and Administrative Support
Food Preparation and Serving-Related
Personal Care and Service
Subtotal
Other occupations
Grand Total
Occupations $25/hr & under 2017
Occupations over $40/hr, 2017
Multiple: highest:lowest

Employment Employment
Mean
Mean
Estimate
Estimate
Hourly
Hourly
May '12
May '17
% Chg. Wage '12 Wage '17 % Chg.
74,790
90,630
21% $ 69.25 $ 79.77
15%
63,470
93,670
48% $ 49.08 $ 60.97
24%
42,730
43,870
3% $ 49.02 $ 57.07
16%
86,290
108,740
26% $ 44.55 $ 49.85
12%
48,280
44,140
-9% $ 29.93 $ 35.51
19%
101,300
98,110
-3% $ 25.98 $ 28.38
9%
47,920
61,770
29% $ 20.22 $ 25.26
25%
156,590
151,820
-3% $ 22.19 $ 25.09
13%
98,410
107,660
9% $ 12.89 $ 17.07
32%
26,490
54,850
107% $ 16.32 $ 15.26
-6%
746,270
855,260
15%
254,160
261,130
3%
1,000,430
1,116,390
12%
329,410
376,100
14%
18.27
21.08
15%
267,280
336,910
26%
49.05
59.02
20%
4.2
5.2

Source: California Employment Development Department, OES Employment and Wages Data
st
st
Tables (2018 1 Quarter and 2013 1 Quarter)
Note: Minor changes in the geographic area boundaries occurred between the two years
compared.

Indicators of growing income inequality in San Francisco between 2012 and 2017
can be seen in a variety of the measures presented in Exhibit 7. Growth in the
number of jobs earning $40 per hour or more was higher at 26 percent compared
to 14 percent for those earning $25 or less between 2012 and 2017. Wages
increased for both of those groups, but at the rate of 20 percent for occupations
earning $40 per hour or more compared to a lower 15 percent increase in wages
for occupations earning $25 or less. Similar to the findings by the Brookings
Institution reported above, the differential in wages between the highest and
lowest paid occupations in the top ten categories increased from 4.2 times greater
in 2012 to 5.2 times greater in 2017, also shown in Exhibit 7.

Housing Prices
Reflecting the growth in population, jobs, wealth for some residents, and
occupations in San Francisco and the region and a relatively limited supply of
housing, Zillow reports that the median sale price for homes in San Francisco
increased from $677,400 in February 2009 to $1.3 million in January 2019, or by 93
percent. While this near doubling in the median home sales price outpaced wage
growth for all occupations, it certainly made homeownership significantly less
affordable for the many workers in San Francisco at the low end of the wage scale.
Budget and Legislative Analyst
19

 Memo to Supervisor Mar
April 23, 2019

Exhibit 8 below displays monthly median house sale prices in San Francisco
between 2009 and 2019.
Exhibit 8: Monthly Median Sale Price in San Francisco – All Homes

Source: Zillow.com, San Francisco Median Sale Price (all homes): February 2009 to January
2019

Zillow also reports that rent listings for a two-bedroom apartment have increased
between 2010 and 2019 from $3,300 to $4,500, or by 36 percent.
In her article “Income Inequality and Urban Displacement: The New Gentrification”
(2016),18 Karen Chapple argues that what is widely viewed as a housing crisis is
actually an income crisis, because at the core of the urban displacement crisis is
income inequality, driven by declining real wages. Chapple notes that housing
prices have risen much faster that incomes, particularly in strong market regions,
and in regions with high income inequality, low-income households find housing
less affordable. Chapple argues for a multi-pronged policy response to the
displacement crisis, with interventions to improve incomes and preserve housing
affordability, in addition to building new supply.
This phenomenon affects not only San Francisco, but other urban centers of the
Bay Area as well. A Silicon Valley Rising report19 found that between 2009 and
2015, the inflation-adjusted average rent in San Jose increased by 32 percent, while
adjusted median incomes for renters decreased by almost 3 percent. Rents have
18

Karen Chapple, “Income Inequality and Urban Displacement: The New Gentrification,” December 9, 2019, New
Labor Forum 2017, Vol. 26(I) 84-93, DOI: 10.1177/1095796016682018.
19
Silicon Valley Rising and Working Partnerships USA, “Soaring Rents, Falling Wages,” undated.
https://siliconvalleyrising.org/files/SoaringRentsFallingWages.pdf
Budget and Legislative Analyst
20

 Memo to Supervisor Mar
April 23, 2019

risen nearly four times faster than wages, and nearly five times faster than Social
Security payments. According to the report, nearly 57 percent of families in San
Jose earning less than $50,000 spend more than half their income each month on
rent, and the gap between wages and rent affordability is nearly three times what
it was in 2009.
IPOs and the Technology Sector
Research by Karen Chapple, John V. Thomas, Dena Belzer, Gerald Autler found that
in the Bay Area, housing price appreciation has a strong relationship to the location
of start-up jobs, particularly those that provide information-intensive services20
(urban livability factors, such as access to food stores and entertainment, also play
a role). Chapple et al. argue that because the occupational patterns of informationintensive services are skewed toward the high end of wages, prices that employees
can pay for housing are inflated, and expansion of these firms therefore creates a
disproportionate share of highly skilled, highly paid workers into the local housing
markets.
The data presented in Exhibits 6, 7, and 8 above support this pattern identified by
Chapple et al. in San Francisco: in recent years, the City has seen relatively faster
growth of more highly-paid workers in information-intensive businesses. Due to
their higher salaries, these workers are better able to afford rent or to purchase
real estate in the San Francisco market. These employees may also benefit from
sudden cash windfalls when the technology companies they work for go public, are
acquired, or have significant rounds of private equity and venture capital
investment. As these events occur and high-earning employees are able to exercise
stock options or take advantage of other equity compensation, they may become
even wealthier and even more able to purchase expensive housing.

IPOs in San Francisco in 2019
Reflecting the resurgence of IPOs in recent years, the following San Francisco
companies either have gone public or may be planning to do so in 2019, according
to various company and media reports. Private companies do not disclose the
number of employees who have received or will receive stock options or other
equity-based compensation (who may experience cash windfalls or other benefits
when their companies go public), nor do companies publicize the number of shares
that employees have or will be granted. However, industry practices indicate that
20

Karen Chapple, John V. Thomas, Dena Belter, and Gerald Autler, “Fueling the fire: Information technology and
housing price appreciation in the San Francisco Bay area and the Twin Cities,” 2004, Housing Policy Debate 15(2):
347-383.
Budget and Legislative Analyst
21

 Memo to Supervisor Mar
April 23, 2019

many of the estimated 7,990 employees in San Francisco at these companies should
be or have been awarded stock options, which would increase their wealth over a
short timeframe and, according to research on IPOs cited below, make at least some
of these employees eligible and motivated to purchase real estate in San Francisco.
Exhibit 9: San Francisco Companies with Expected or Actual IPOs in 2019
Company

Valuation

Airbnb
Lyft*
Pinterest*
Postmates
Slack
Uber
Total

$35 bn.
$24 bn.
$12.7 bn.
$1.85 bn.
$7.1 bn.
$100 bn.

Estimated #
employees in SF
2,300
1,600
unknown.
340
750
3,000
7,990

Sources: San Francisco Business Times, July 2018, as of January 2018 (estimated number of
employees in San Francisco). For private companies, valuations taken from recent media
reports (including New York Times, Reuters, Recode, San Francisco Business Journal, Forbes,
and the Wall Street Journal) and may be subject to change. For companies that have already
gone public, valuation listed is the valuation at the date of IPO.
*Valuation at IPO

Estimated Impact of IPOs on Housing Prices
The median sale price of a home in San Francisco as of January 2019 was $1,304,200
according to Zillow. Three recent studies have quantified the impact of IPOs on
housing prices in areas adjacent to IPO company headquarters. The results of the
three studies are:
1. Economists Barney Hartman-Glaser, Mark Thibodeau, and Jiro Yoshida, cited
above, found that real estate prices increased within 10 miles of IPO company
headquarters by 1.0 percent more than in the surrounding area after the
company filed to go public, by 0.8 percent more after issuing, and by zero
percent after the lock-up expiration, or the date when employees can exercise
their options and sell their shares. This conclusion was based on the economists’
analysis of 725 California company IPOs between 1970 and 2015 and 6.5 million
real estate transactions.
2. An economist at Zillow analyzed the impact of IPOs on home values in census
tracts that were home to many Facebook employees at the time of Facebook’s
IPO. He found that between March 2012 and March 2013,21 home values in

21

The Facebook IPO was in May 2012.
Budget and Legislative Analyst
22

 Memo to Supervisor Mar
April 23, 2019

census tracts likely to be home to Facebook employees increased 21 percent,
compared to a 17 percent increase in all other Bay Area census tracts.22
3. A third study found that IPOs have spillover effects on real estate outcomes,
including home prices and mortgage originations,23 and found that home price
growth increases by more than one third in zip codes located within two miles
from the IPO headquarters, but only for expensive homes. The study also
showed that home price growth accelerates after the expiration of the
employee lockup period, which supports the hypothesis that some of the
increase in housing prices is due to employees who see cash windfalls as a
result of IPOs, especially because home price growth is higher for listings when
the firm’s stock market price increases after the IPO.
While there are differences in the details of the outcomes of these studies, the
general conclusion is that IPOs affect real estate prices in areas close to IPO
company headquarters. We applied the most conservative results of the three
studies to San Francisco housing prices to provide an estimate of the potential
impact of the IPOs believed likely to occur in 2019. As shown in Exhibit 10 below,
applying the 1.8 percent average impact from the Hartman-Glaser et al. study to San
Francisco’s $1.3 million median home price as of January 2019 could add between
$23,476 for a single IPO and up to $147,346 to San Francisco’s median housing price
if all of the six anticipated IPOs occur. This impact would likely occur over a one- to
two-year period, allowing for completion of all employee lock-up periods (often six
months after IPO date). This impact is based on an average and would depend on all
six anticipated IPOs occurring so that those with larger impacts balance out those
with smaller impacts.
Limitations of this estimate include: 1) the study on which it is based focused on all
IPOs in California, many of them in Silicon Valley, where housing characteristics and
the real estate market is different than in San Francisco, and 2) there has been
significant publicity about the likely and actual 2019 IPOs in San Francisco already,
and some media reports indicate that sellers may be adjusting their timing and
upward pricing of their properties in advance of the IPOs in anticipation of buyers
with newly-liquid assets. If this is the case, some of the impact may already be
incorporated in pre-IPO real estate prices.
While the San Francisco housing market is different than the market in Silicon Valley
and other parts of California, we conclude that those differences would not affect
buying behavior associated with an IPO. We would expect buyer behavior to be
22

Tucker, Jeff. Post-IPO, Home Values Grew Faster in Areas Home to Lots of Facebook Employees. February 4, 2019.
www.zillow.com/research
23
Alexander W. Butler, Larry Fauver, and Ioannis Spyridopoulous, “Local Economics Spillover Effects of Stock
Market Listings,” SSRN, December 5, 2018. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2695464
Budget and Legislative Analyst
23

 Memo to Supervisor Mar
April 23, 2019

similar to what was observed in other studies: a certain number of employees at San
Francisco companies would choose to buy a home close to work in San Francisco
when given the resources to do so after an IPO. Regarding the second caveat, to the
extent sellers are pricing their properties in anticipation of the IPOs, there is still an
upward impact on prices due to the IPOs; it just may be occurring before the actual
IPO event.
Exhibit 10: Impact of Six Expected IPOs on Median Housing Prices in San Francisco
Median Value $ change
Actual: Feb. 2019
after 1st IPO
after 2nd IPO
after 3rd IPO
after 4th IPO
after 5th IPO
after 6th IPO
Total

$ 1,304,200
1,327,676
1,351,574
1,375,902
1,400,668
1,425,880
1,451,546

% change

$23,476
23,898
24,328
24,766
25,212
25,666

1.8%
3.6%
5.5%
7.4%
9.3%
11.3%

$ 147,346

11.3%

Source: Median value: Zillow.com, San Francisco Median Sale Price (all homes): January
2019. IPO impact factor: “Cash to Spend,” Hartman-Glaser, Thibodeau, and Yoshida.
November 30, 2018.

The results shown in Exhibit 10 assume that the increase would be at the average
level for all IPOs reviewed in the Hartman-Glaser, et al. “Cash to Spend” study, or
1.8 percent per IPO. However, Hartman-Glaser et al. also report that:
“The impact on housing prices is larger when the wealth increase by IPO is
larger: e.g., higher offer price, larger IPO proceeds, and larger share
underpricing at filing. The impact is also larger for younger and smaller
firms.”
In other words, since some of the San Francisco companies’ IPOs may be large and
will likely have larger IPO proceeds than average (such as Uber), the impact on
housing prices would likely be even larger, according to the analysis. Conversely, for
companies that may have smaller IPOs, the impact of a particular IPO may be less
than Hartman-Glaser et al.’s estimation of 1.8 percent. As mentioned above, the
estimated impact shown in Exhibit 10 depends on all six companies going public,
with the impact of the larger companies balancing out the lesser impacts of the
smaller companies.
To the extent housing prices increase beyond their already high levels due to the
impact of the IPOs, the pattern associated with gentrification in San Francisco will
be exacerbated, making housing less affordable, particularly to lower income
residents. While the IPOs may help workers in the companies that go public be
Budget and Legislative Analyst
24

 Memo to Supervisor Mar
April 23, 2019

better able to purchase housing in San Francisco, these are generally well paid
employees to begin with. As stated in the Zillow research cited above:24
“Today home prices have never been higher though appreciation is slowing
dramatically. Buyers often need exceptionally high incomes, or a sudden
windfall that an IPO could provide, to break into the market.”

Policy Considerations
While not the sole cause of income inequality in San Francisco, the occurrence of up
to six IPOs in 2019 will add a one-time injection of wealth to many well-paid
workers, investors, and founders further increasing the wealth gap and likely
causing some increase in housing prices. Some policy approaches the Board of
Supervisors could consider to address and mitigate the effects of income inequality
are to:
1. Enhance funding for existing City workforce development, economic
development, affordable housing, and subsidized childcare programs,
particularly those aimed at the low and moderate income workforce.
2. Enhance City programs aimed at financial empowerment, such as those
administered by the Treasurer-Tax Collector including Kindergarten to College,
Bank on SF, and Smart Money Coaching.
3. Enhance taxes such as the City’s payroll or gross receipts tax for companies
above a certain size and/or gross receipts level, with the proceeds used for
programs and services designed to minimize the impacts of income inequality.
4. Establish a gross receipts surtax on companies with differentials between CEO
and average worker pay above a certain level, with the proceeds used for
programs and services designed to minimize the impacts of income inequality.
a. The City of Portland, Oregon has established such a surtax.
5. Establish other taxes on large or high-value companies in San Francisco, with the
proceeds used for programs and services designed to minimize the impacts of
income inequality. Some examples could include:
a. In November 2018, East Palo Alto voters approved an annual parcel tax
of $2.50 per square foot on commercial office property over 25,000
square feet, with tax revenue reserved for spending on affordable and
supportive housing programs and programs that facilitate residents’
access to jobs in the science, technology, engineering, and mathematics
sectors and the building trades.
24

Ibid., Tucker.
Budget and Legislative Analyst
25

 Memo to Supervisor Mar
April 23, 2019

b. Also in November 2018, Mountain View voters approved an annual peremployee business license tax, with larger companies paying more per
employee. (Revenue from this tax is considered unrestricted general
revenue.)
c. In 2018, the Seattle City Council passed a “head tax” ordinance that
would have levied a $275 per full-time employee25 tax on Seattle
businesses that have at least $20 million in annual taxable gross
revenue. Tax revenue would have been spent on affordable housing and
services for Seattle’s homeless population. The Seattle City Council later
repealed the head tax legislation.
6. Provide incentives or assistance to households under a certain income level to
enroll in retirement plans.
7. Lobby for changes to federal and state tax policies.

25

The proposed tax, formally an Employee Hours Tax, would have applied to all Seattle employees at a rate of 26
cents per hour for 2019 and 2020, with the annual tax per employee capped at $275.
Budget and Legislative Analyst
26