Tying Office Development to
Affordable Housing Production:
Economic Impact Report

CITY & COUNTY OF SAN FRANCISCO

Office of the Controller

Office of Economic Analysis

01.27.2020

 2

Introduction
 Since the 1980s, large office development in San Francisco has been
subject to limitations imposed by Proposition M, which was approved
by voters in 1986. Prop M applies a limit to the amount of additional
new office space that the City can authorize in a year.
 The proposed ballot measure, known as the “San Francisco Balanced
Development Act”, would reduce this annual limit if the city did not
meet a certain target for affordable housing production that is set by
State and Regional agencies. If the affordable housing target was ever
exceeded, the limits established by Prop M would continue to apply.
 Given the level of affordable housing production in the city in past
years, the proposal has the potential to significantly reduce future
office development, and employment, in San Francisco. For this
reason, the Office of Economic Analysis has prepared this economic
impact report.

 3

Details of the Measure
 Prop M established an annual allocation of 875,000 square feet of new
office space in large development projects1.
 The allocation is reduced by the square footage of each approved
office building . Any allocation that is not used in a year is available in
the next year, along with an additional 875,000 of new allocation.
 Thus, Prop M ensures that the city does not add more than 875,000 of
square feet of large office space per year, on average over the long
term.
 The proposed measure would reduce the 875,000 annual allocation by
a percentage equal to the percentage by which the city missed its
annual affordable housing target in the prior year.
 For example, if the annual affordable housing target was 2,042 units,
and 75% of that target (1,532 affordable units) was produced in that
year, then the city would have missed the target by 25%. The new
allocation in the next year would be cut by 25%, from 875,000 square
feet to 656,250.

 4

The Affordable Housing Requirement
 The affordable housing target referred to in the measure comes from
the Regional Housing Needs Allocation (RHNA), a state program
implemented by regional agencies.
 The RHNA process establishes housing production targets for
California cities every eight years. The targets are broken down by the
income levels of their intended residents.
 Under the proposed measure, office allocation would be reduced
unless the city produced 2,042 affordable housing units per year for
low- and moderate-income households, or 1/8 of the city’s 8-year
RHNA target, whichever is greater, each year.
 The measure would not relax any of the City’s planning restrictions on
housing development, and would not generate any new funding for
affordable housing.
 Because new office development funds affordable housing through
the City’s Jobs-Housing Linkage Fee, restrictions of office development
will reduce funding sources for new affordable housing.

 5

Further Provisions and Exemptions
 The restrictions associated with reduced allocations would not apply to
new office projects which began before September 11, 2019. Much of
the office development which was planned as part of the Central
SoMa redevelopment would therefore not be affected by this
legislation.
 This provision will likely greatly limit the impact of the proposed
measure in the near term, though the measure will continue to apply
indefinitely into the future.
 Later projects within Central SoMa may only be approved after 15,000
housing units have been completed within the neighborhood.
 Additionally, future projects may be completed, irrespective of the new
limits on allocation, under several provisions. However, any office
development project that moves forward under these provisions will
reduce the allocation available in future years. Over the long term,
therefore, they will not change the measure’s impact on future office
development, or the city’s economy.

 Employment Growth and Housing Affordability

Jobs-Housing Ratio and Low- & Moderate-Income Housing Burden
in San Francisco, 2006-2018

2

San Francisco Jobs per Housing Unit

The proposed measure raises the issue
of the relationship between employment
growth and housing affordability in the
city.
59%

1.9

58%

1.8

57%

1.7

56%

1.6

55%

1.5

54%

1.4

53%

1.3

52%

1.2

51%

1.1

50%

1

2006

2007

2008
2009
2010
Jobs / Housing Unit

2011

2012
2013
2014
2015
2016
2017
% of Burdened Low/Moderate Income Households

Source: Bureau of Labor Statistics; Census Bureau, American Communities Survey, ipums.org;
Mayor’s Office of Housing and Community Development.

2018

49%

% of Households Spending More than 30% of Tehir Income on Housing

6

Employment growth clearly grew faster
than housing supply in San Francisco
during the 2010s. The chart to the left
shows that the city’s jobs/housing ratio
has risen by more than 25% since 2010.
However, while the housing burden
facing low- and moderate-income San
Franciscans remains very high, housing
appears to have become more, rather
than less, affordable as the jobs-housing
ratio has increased during the 2010s.
According to Census data, the
percentage of low/ moderate income
households in the city spending more
than 30% of their income on housing has
tended to decline since the end of the
recession2.
As shown on the next page, one reasons
for this seems to be that job growth has
fueled growth in household income,
which has been faster than growth in
housing costs, for most households in
the city.

 7

Growth in Incomes and Housing Costs Since 2012

Average Annual Change in Median Household Income, and Median Housing Costs,
By Income Group in San Francisco, 2012-2018
5.0%

4.5%

4.0%

3.5%

3.0%

Above-moderate income
households, making over 120% of
the area median income, have the
widest gap, indicating the most
growth in housing affordability.

2.5%

2.0%

1.5%

1.0%

0.5%

0.0%

The chart to the left shows the
growth in median housing costs,
and median incomes, that sampled
San Francisco households reported
to the Census in 2012 and 2018. In
the chart, households are broken
out by income group. A gap where
income has grown faster than
housing costs indicates that housing
has become more affordable to the
median household in that income
group sample.

Extremely/Very Low Income

Low Income
Growth in Median Housing Costs

Moderate Income
Growth in Median Household Income

Source: Census Bureau, American Communities Survey, ipums.org;
Mayor’s Office of Housing and Community Development.

Above Moderate Income

However, extremely/very low, and
low income groups also have
experienced income growth
exceeding their growth in housing
costs since 2012. Only moderate
income households, from 80-120%
of the area median income, have
seen essentially the same growth in
housing costs and income since
2012.

 Available Office Allocation Under Prop M

8

San Francisco Large Office Development, and Available Prop M Allocation,
1985/86 to 2019/20
5,000,000

4,500,000

Prop M establishes an annual
allocation for new office space, and
any amount that is unused is
available for projects in later years.
In fact, Prop M’s impact on office
development has been very limited
until the present year, 2019-20.

4,000,000

3,500,000

In the early 1990s, there was little
demand for new office space in
San Francisco, and the available
allocation grew. Since 1999, San
Francisco has added more than 1.2
million of square feet per year of
large office space, exceeding the
annual allocation, because of the
amount built up in earlier years.

3,000,000

2,500,000

2,000,000

1,500,000

1,000,000

500,000

0

Office Space Produced

Source: San Francisco Planning Department

Available Allocation Under Prop M

However, by the end of the 2018-19
period, only 21,752 square feet
remained. Even if the proposed
measure is not enacted, Prop M is
likely to constrain office
development and limit job growth
in the near future.

 Affordable Housing Production Versus the Target

9

Production of Affordable Housing in San Francisco, and Shortfall Relative to the Target in the
Proposed Measure, 1999-2018
2,500

Minimum Target: 2,042 affordable units per year
2,000

The city has never met this
target in a single year, and it
substantially exceeds the city’s
track record producing new
affordable housing. Over the
past 10 years, San Francisco has
produced between an average
of 712 affordable housing units
per year, which is 35% of the
target.

1,500

1,000

500

0

As stated earlier, the measure
specifies that the city needs to
produce a minimum target of
2,042 units of affordable
housing annually to prevent
further restrictions to office
development and job growth.

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

Low- and Moderate-Income Affordable Units Produced

2010

2011

2012

2013

2014

Shortfall Against the Target

Source: San Francisco Planning Department, San Francisco Housing Inventory, various years.

2015

2016

2017

2018

If either these recent or longerterm trends continue in the
future, the proposed measure
would lead to a significant
reduction in future office
development and job growth.

 10

Economic Impact Factors
 If past trends continue and San Francisco continues to demand
significant new office space while producing affordable housing at
levels below the target in the proposed measure, it will act to further
constrain office development in the city, beyond the effects of Prop M.
 If this happens, office supply constraints will put further upward
pressure on office rents in the city, and reduce employment below
what it would otherwise be.
 In addition, reduced office construction will have a negative impact on
the city’s construction industry.
 Finally, the proposed measure will have a fiscal impact to the City
government. A reduction in office space can be expected to reduce
the City’s property and business tax revenues, among others, while
also reducing the City’s General Fund costs of servicing new
development.
 Less office development would also lead to less funding for affordable
housing, through the Jobs-Housing Linkage Fee.

 11

Potential Impact on Future Office Development
 The future impact of the proposed measure on office development
depends on future demand for office space in the city, the availability
of sites for new office development, and city’s future level of affordable
housing production, among other factors.
 To illustrate the measure’s potential impact, we can estimate what
would have happened if the proposed measure had been in place for
the past 10 years, from 2009 to 2018. During that time, developers
built 1.2 million square feet of large office space per year, and the city
produced an average of 712 affordable units, equal to 35% of the
target in the proposed measure.
 If these trends continued for the next 20 years, then the proposed
measure would cause the city to have 560,000 fewer square feet of
office space per year, or 10.6 million less after 20 years.
 This level of office development could support approximately 2,375
fewer office jobs per year, or 47,500 fewer jobs after 20 years.
 It also represents a loss of 64% of the 875,000 square foot allocation
limit imposed by the original Prop M.

 12

Potential Fiscal Impact: Revenue Loss
 A reduction in future office space will reduce future property tax
revenue. A conservative estimate of the lost revenue can be made by
multiplying the annual office space reduction by the per-square foot
assessed value of new office space, and the General Fund property tax
rate. Using this method, we estimate the annual revenue loss at $4.8
million per year, rising to $95 million annually after 20 years.
 The business tax impact can be estimated, based on current business
tax payments and employment in office-using industries. Again, this is
a conservative estimate, because business tax revenue has grown very
rapidly in recent years. We estimate an annual loss of $6.8 million per
year, rising to $135 million annually after 20 years.
 Thus, the combined revenue loss from business and property tax
would equal an estimated $11.5 million in year 1, rising to $231 million
by year 20.
 A 560,000 square foot reduction in office space development would
also lead to a loss of $39 million per year in Jobs-Housing Linkage Fee
revenue, which is an impact fee paid by new office development to
fund affordable housing programs.

 13

Cost Savings and Total Fiscal Impact
 A proper accounting of the fiscal impact needs to reflect the fact that a
slowdown in future office development will also lead to cost savings
for the City, as fewer employees will require services paid for by the
City’s General Fund, such as police, fire, parks, and libraries.
 A common way to approximate this cost impact is to divide taxfunded General Fund expenditures by the City’s “service population”,
often defined as the city’s daytime population, but with a reduced
weighting for in-commuters, out of a belief that they use City services
like police, fire, and parks less than residents do.
 Assuming the cost of serving an in-commuter is 50% of the cost of
serving a city resident, the City would save $4.4 million in year 1, rising
to $87 million by year 20. If, conservatively, in-commuters were
assumed to use City services to the same degree as residents, the City
would save $7.3 million in year 1, rising to $146 million by year 20.
 Either way, when the revenue losses on the previous page are
considered, the proposed measure would lead to a negative net fiscal
impact to the City, of between $4.2 and $7.2 million in year 1, rising to
between $84 - $146 million annually by year 20.

 14

REMI Model Inputs
 The Office of Economic Analysis uses the REMI model to estimate the
net economic impact of new City legislation. The estimates below are
sensitive to the assumptions made about the future demand for new
office space, and the future production of affordable housing in the
city. This forecast is based on the office and affordable housing
production trends of the last ten years. The specific impacts modelled
include:
 A reduction, relative to a baseline forecast, of 2,375 office jobs, spread
proportionally across major office-using industries in the city. This job
reduction will grow each year, as more office development is reduced, to a
loss of 47,500 jobs after 20 years. Accompanying this is a reduction in
office construction spending, estimated at $625 per square foot.
 A reduction of $5.7 million in General Fund spending, the midpoint of the
negative fiscal impact discussed on the previous page. Again, this rises
over time, to $114 million after 20 years.
 A reduction of $39 million per year in affordable housing funding through
the Jobs-Housing Linkage Fee program. Unlike property and business
taxes, this is a one-time fee paid when construction is completed, and this
economic loss does not grow over time.

 15

Economic Impact Assessment
 REMI estimates impacts relative to a baseline economic forecast of
long-term growth. In 2040, the city’s GDP would be 8.5% smaller than
it would be relative to the baseline, or $23 billion less in today’s
dollars. Disposable personal income is expected to be 5.9% less, and
the city’s population would be 5.8% smaller than under the baseline.
 Similarly, total employment in the city would decline by 7.9% after 20
years, relative to the baseline forecast, which is equivalent to 91,000
jobs in 2040.
 In terms of relative job losses by industry, large office-using industries
like professional services, technology, and financial services, would
experience the greatest job losses, about two-thirds of the total.
 In percentage terms, lower-paying office-using industries like
administrative services, real estate, and non-profits would be the
hardest hit, each suffering more than 10% job loss, relative to the
baseline.
 Because of multiplier effects, non-office industries like construction,
retail trade, hospitality, and health care would all lose more than 2,000
jobs by 2040, again relative to the baseline.

 16

City GDP: Baseline and Policy Forecasts
The chart to the left illustrates
projected GDP growth in San
Francisco, under both a baseline
forecast and the proposed
measure.

Projected San Francisco GDP, 2021-40:
Baseline Projection and Under Proposed Measure
$290

While the initial difference is
small, the city’s economy would
be 8.5% smaller by 2039, as
noted on the previous page.

$270

$ Billion

$250

In this forecast, the measure is
projected to reduce office
development each year, leading
to a widening divergence
between the two forecasts.

$230

$210

This cumulative effect is what
gives this particular policy its
unusually large negative
economic impact.

$190

$170

2021

2022

2023

2024

2025

2026

2027

2028

GDP Projection, Baseline (2019 $)

2029

2030

2031

2032

2033

2034

2035

2036

GDP Projection, with Proposed Measure (2019 $)

2037

2038

2039

2040

 17

Conclusions
 By tying future office development to an affordable housing target
that the city has never met, the proposed measure is likely to lead to
higher office rents, reduced tax revenue, reduced incomes, and
reduced employment across the city's economy.
 The city’s recent history suggests that employment growth has been
associated with rising incomes. For most income groups in the city,
median income growth since 2012 has exceeded the growth in median
housing costs.
 While the near-term impact of the proposed measure is likely to be
greatly limited by the exemption of many Central SoMa office projects
that already underway, the cumulative effect of the measure over the
long term gives it an unusually large negative fiscal and economic
impact.
 These impacts are relative to a baseline forecast of long-term growth,
and are sensitive to assumptions of future economic growth in the city.
If future demand for office space is less than it has been in the past, or
the city’s production of affordable housing is higher, then the
measure’s negative impacts would be less than forecast in this report.

 18

End Notes

1.

Large cap office projects are defined as buildings with more than 50,000 square feet.

2.

Low- and moderate-income households in San Francisco were identified using ACS PUMS data from
ipums.org, and annual unadjusted income limits from the Mayor’s Office of Housing and Community
Development.

 19

Staff Contacts
Ted Egan, Ph.D., Chief Economist
ted.egan@sfgov.org
For press inquiries: CON.Media@sfgov.org