HomeMy WebLinkAboutUnderstanding the MarketUC Berkeley
IURD Working Paper Series
Title
Understanding the Market for Secondary Units in the East Bay
Permalink
https://escholarship.org/uc/item/9932417c
Authors
Wegmann, Jake
Chapple, Karen
Publication Date
2012-10-01
eScholarship.org Powered by the California Digital Library
University of California
UNIVERSITY OF CALIFORNIA
Understanding the Market for Secondary
Units in the East Bay
Working Paper 2012-03
Jake Wegmann and Karen Chapple
October 2012
IURD
Working
Paper
WP-‐2012-‐03:
Understanding
the
Market
for
Secondary
Units
in
the
East
Bay
Jake
Wegmann
and
Karen
Chapple
INTRODUCTION
AND
SUMMARY
In
order
to
understand
the
potential
market
for
secondary
units,
it
is
important
to
gain
some
insight
into
the
functioning
of
the
existing
market
for
such
dwellings.
Examining
this
market
in
detail
is
made
challenging,
as
compared
to
other
residential
submarkets,
by
the
widespread
lack
of
building
and/or
zoning
permits
for
secondary
units.
Because
of
these
informal
characteristics,
market
information
that
is
routinely
published
for
other
types
of
residential
dwellings
by
third-‐party
private
entities,
scholars,
cities,
the
US
Census
and
other
sources
is
essentially
non-‐existent
for
secondary
units.
We
therefore
must
find
ways
to
study
what
essentially
amounts
to
a
black
market
for
secondary
unit
housing
without
relying
on
the
data
that
is
typically
available
for
housing
in
general.
This
working
paper
describes
in
detail
the
methods
and
findings
from
two
separate
techniques
that
we
used
to
investigate
the
market
for
secondary
unit
housing.
One
is
a
survey
of
homeowners
residing
within
the
rail
transit
station
areas
of
the
study
region.
The
other
is
an
analysis
of
rental
Internet
advertisement
data
for
units
for
lease
throughout
the
study
area.
Each
technique
should
be
viewed
as
a
separate
approximation,
or
“cut”
through
the
market
for
secondary
unit
housing
in
the
East
Bay.
Each
technique
has
a
differing
set
of
strengths
and
shortcomings.
The
summary
of
our
findings
from
the
two
methods
is
presented
below,
while
much
lengthier
descriptions
of
each
method
are
given
afterwards.
Together,
our
two
methods
begin
to
paint
a
picture
of
the
operation
of
a
market
that
functions
largely
hidden
in
plain
sight.
Summary
of
rental
market
analysis
results
Reviewing
the
descriptive
statistics
and
hedonic
modeling
results
presented
below,
evidence
for
several
important
characteristics
of
secondary
units
as
a
rental
submarket
distinct
from
the
overall,
mainstream
rental
market
emerges.
The
most
notable
results
from
the
two
analyses
are
as
follows:
• Approximately
16
percent%
of
single-‐family
residential
properties
in
the
study
area
have
at
least
one
secondary
unit.
This
result
provides
evidence
that
secondary
units,
despite
being
frequently
unpermitted,
are
not
a
marginal
or
aberrational
phenomenon,
but
rather
are
widespread
throughout
East
Bay
station
areas
and
almost
certainly
throughout
the
East
Bay
flatlands.
• Secondary
units
most
commonly
take
the
physical
form
of
a
detached
rear
yard
structure,
a
converted
garage,
or
converted
rooms
in
the
main
house.
A
large
majority
were
installed
five
or
more
years
ago.
The
vast
majority,
though
not
all,
are
used
as
housing
units.
• While
not,
on
average,
cheaper
in
terms
of
price
per
square
foot,
secondary
units
are
more
likely
to
represent
smaller,
and
therefore
cheaper,
housing
packages
than
the
typical
rental
unit
in
the
East
Bay.
This
result
suggests
that
secondary
units
are
particularly
important
to
small
households
of
one
to
two
people
seeking
affordable
housing.
Secondary
units
rented
to
strangers
are
at
least
6
percent%
cheaper
than
non-‐secondary
units
when
their
affordability
is
expressed
in
terms
of
a
proportion
of
Area
Median
Income
(AMI),
as
is
common
practice
in
the
affordable
housing
industry.
• Ownership
and
operation
by
amateurs
is
much
more
common
(though
far
from
ubiquitous)
in
secondary
units
than
in
the
general
rental
market.
This
has
important
consequences
in
the
functioning
of
the
secondary
unit
submarket.
For
instance,
there
is
at
least
modest
evidence
that
on-‐site
secondary
unit
landlords
appear
to
offer
rents
at
a
discount
to
their
tenants
vis-‐
à-‐vis
their
absentee
landlord
peers.
In
addition,
secondary
unit
rents,
controlled
for
amenities,
appear
to
exhibit
a
wider
degree
of
variation
in
rents
than
typical
rental
units,
suggesting
the
importance
to
secondary
unit
tenants
of
engaging
in
extensive
searches
to
find
an
optimal
housing
package
when
navigating
this
idiosyncratic
market
if
they
do
not
happen
to
have
a
personal
relationship
with
a
secondary
unit
landlord
in
a
desired
location.
Indeed,
our
results
show
that
the
most
common
methods
by
which
secondary
unit
landlords
find
tenants
are
by
already
knowing
the
person,
Craigslist,
and
personal
referrals
(in
descending
order
of
frequency).
Finally,
we
observe
that
26%
of
secondary
units
are
occupied
by
the
homeowning
household
itself,
while
an
additional
29%
are
occupied
by
friends,
family
or
acquaintances
of
the
homeowner
household,
some
of
whom
are
living
rent-‐free
and
others
of
whom
are
likely
paying
reduced
rents.
• Secondary
units
are
much
less
likely
to
offer
off-‐street
parking
than
standard
rental
units.
Indeed,
single-‐family
residential
(SFR)
properties
with
secondary
units
do
not
appear
to
have
significantly
more
off-‐street
parking
than
those
that
lack
secondary
units.
This
suggests
that,
regardless
of
off-‐street
parking
requirements
imposed
by
cities,
secondary
units
are
often
offered
to
tenants
who
either
do
not
have
a
car
or
else
park
their
car(s)
on
the
street.
Furthermore,
in
the
relatively
infrequent
cases
in
which
secondary
units
do
offer
off-‐street
parking,
landlords
very
seldom
charge
for
this
amenity.
This
implies
that
off-‐
street
parking
requirements,
if
followed,
would
often
impose
a
cost
on
secondary
unit
landlords
that
they
would
not
or
could
not
pass
on
to
tenants.
• While
most
homeowners
are
aware
of
secondary
units
on
their
streets,
and
while
most
of
these
homeowners
do
not
perceive
there
to
be
a
negative
impact
from
secondary
units,
those
that
do
perceive
a
negative
impact
are
by
far
most
likely
to
cite
increased
pressure
on
on-‐street
parking
as
the
reason.
Indeed,
our
results
suggest
that
SFR
properties
with
one
secondary
unit
generate,
on
average,
0.9
more
total
cars
and
consume
0.7
more
on-‐street
parking
spaces
than
SFR
properties
lacking
secondary
units.
At
the
same
time,
SFR
properties
with
secondary
units
exhibit
wider
variability
in
car
generation
than
those
that
do
not.
• Secondary
units
appear
to
be
much
likelier
than
other
types
of
rental
units
to
offer
substandard
cooking
facilities
(though
it
should
be
noted
that
the
vast
majority
do
appear
to
offer
full
kitchens).
This
is
likely
a
consequence
of
both
the
amateur
operation
of
secondary
units
and
the
widespread
existence
of
unpermitted
secondary
units.
Substandard
cooking
facilities
are
likely
associated
with
other
substandard
conditions,
although
this
cannot
be
directly
gleaned
from
the
data
presented
herein.
• While
we
found
no
overwhelming
demographic
differences
between
households
that
have
secondary
units
and
those
that
do
not,
the
former
have
a
somewhat
smaller
number
of
people,
slightly
lower
incomes,
and
somewhat
fewer
white
adults.
Adults
living
in
secondary
units
are
considerably
younger
than
adults
in
homeowner
households,
though
relatively
few
are
under
25,
and
very
few
are
elderly.
There
are
few
children
residing
in
secondary
units.
• Secondary
units
are
much
more
likely
to
share
utility
costs
with
other
units
or
to
have
them
included
in
the
rent.
This
is
likely
associated
with
widespread
non-‐professional
management
as
well
as
a
dearth
of
fully
independent
gas,
electric,
water
and
sewer
connections
to
utility
main
lines
for
secondary
units.
• The
potential
market
for
secondary
units
amongst
homeowners
that
do
not
already
have
them
is
about
30%
of
the
total
universe
of
such
homeowners.
HOMEOWNER
SURVEY
The
homeowner
survey
was
intended
to
sample
the
existing
population
of
homeowners
living
within
the
study
corridor
in
order
to
ascertain
i)
the
prevalence
rate
of
secondary
units
for
SFR
properties
within
the
corridor;
ii)
the
attitudes
towards
secondary
units
of
homeowners
residing
on
properties
that
lack
them;
iii)
physical
characteristics
of
extant
secondary
units;
iv)
demographic
characteristics
of
homeowners
(with
and
without
secondary
units)
and
secondary
unit
occupants;
v)
details
concerning
the
economic
relationship
between
secondary
unit
homeowners
and
their
tenants;
and
vi)
the
parking
habits
of
homeowner
households
(with
and
without
secondary
units)
and
secondary
unit
occupant
households.
Because
none
of
the
cities
lying
within
the
study
corridor
had
reliable
records
of
secondary
units
within
their
boundaries
–
whether
permitted
or
otherwise
–
it
was
necessary
to
survey
a
sample
of
randomly-‐selected
SFR
properties
in
order
to
find
ones
with
secondary
units,
even
though
(as
expected)
a
large
majority
of
such
properties
lacked
secondary
units.
However,
receiving
survey
data
from
a
large
number
of
SFR
properties
lacking
secondary
units,
in
addition
to
making
it
possible
to
derive
a
secondary
unit
prevalence
estimate,
also
made
it
possible
to
contrast
non-‐secondary
unit
homeowner
households
with
households
that
have
such
units,
and
to
gauge
attitudes
towards
secondary
units
amongst
the
households
that
lack
them.
The
homeowner
survey
instrument
(Appendix
1),
in
addition
to
asking
respondent
homeowners
to
answer
specific
questions,
also
allowed
them
to
express
any
opinions
they
chose
to
share
with
the
researchers
about
secondary
units,
whether
or
not
they
had
any
on
their
own
properties.
Some
of
the
resulting
responses
(edited
for
spelling
and
length)
are
quoted
throughout
this
Working
Paper.
While
each
of
these
quotes
only
represents
the
viewpoint
of
one
particular
homeowner
that
responded
to
the
survey,
it
nevertheless
offers
a
glimpse
into
the
perspective
of
one
among
the
individuals
whose
responses
to
specific
questions
are
quantified
and
reported
in
this
document.
Design
of
homeowner
survey
Selection
of
sample:
The
properties
selected
for
inclusion
in
the
survey
sample
were
restricted
to
those
lying
within
the
half-‐mile
station
areas
described
in
Yes,
but
will
they
let
us
build?
The
feasibility
of
secondary
units
in
the
east
bay
(IURD
WP-‐2012-‐02).
Only
properties
classified
as
SFR
by
the
Alameda
and
Contra
Costa
tax
assessors
were
included
in
the
pool
of
properties
that
could
be
selected.1
Properties
were
then
selected
for
inclusion
in
the
sample
on
the
basis
of
random
numerical
rankings.
We
dropped
properties
that
could
be
readily
seen
to
be
non-‐owner
occupied
(because
of
either
a
discrepancy
between
the
property
address
and
the
mailing
address
for
the
owner,
or
because
the
owner
was
listed
as
a
firm
or
trust
rather
than
a
person)
from
the
sample.2
This
was
because
i)
we
presumed
that
we
were
much
less
likely
to
receive
a
response
from
non-‐
owner
occupied
properties
and
ii)
in
this
study
we
are
primarily
concerned
with
the
actions
of
owner-‐occupants
and
their
decisions
regarding
whether
or
not
to
install
secondary
units
on
their
properties.
Finally,
properties
were
selected
in
the
order
of
the
random
rankings
until
a
sample
of
the
desired
size
had
been
assembled.
Survey
modes:
We
deployed
a
mixed-‐mode
approach
that
used
three
methodologies
for
contacting
and
gathering
survey
responses
from
homeowners
whose
properties
had
been
selected
to
be
in
the
sample.
The
methodologies
were
as
follows:
1) Every
homeowner
in
the
sample
selected
from
parcel
data
received
a
postcard
(as
shown
in
Appendix
2),
sent
via
U.S.
Mail,
with
a
very
brief
description
of
this
study,
an
invitation
to
participate,
and
a
web
link
to
an
online
survey
instrument.
Every
postcard
was
printed
in
English
on
one
side
and
in
Spanish
on
the
other
side
(with
the
web
link
on
the
Spanish
side
pointing
to
a
Spanish-‐language
version
of
the
survey
instrument).
In
rough
concordance
with
the
Dillman
method,3
all
homeowners
who
had
not
replied
to
the
first
postcard
within
one
week
of
its
mailing
were
mailed
a
second
postcard.
Some
homeowners
who
had
not
1
Discussions
with
staff
from
the
various
cities
revealed
that
city
and
county
tax
assessor
records
of
current
land
use
for
residential
(and
other
properties)
can
sometimes
conflict.
We
relied
on
the
county
tax
assessor
records
for
building
the
samples,
since
these
were
the
comprehensive
data
sets
to
which
we
had
access.
It
is
also
worth
mentioning
that
the
classification
of
a
property
as
SFR
in
no
way
precluded
it
from
having
a
secondary
unit,
as
the
survey
results
presented
in
this
Working
Paper
indicate;
indeed,
it
is
likely
that
a
majority
of
secondary
units
are
not
included
in
the
assessments
of
their
host
properties
for
the
purposes
of
property
taxation.
The
implications
of
this
are
discussed
further
in
Scaling
up
secondary
unit
production
in
the
East
Bay:
Impacts
and
policy
implications
(WP-‐2012-‐5).
2
Because
California
estate
laws
encourage
homeowners
to
place
their
property
in
a
trust,
the
assessor’s
databases
include
many
resident
homeowners
listed
as
trusts.
We
conservatively
exclude
these
properties
because
of
the
possibility
that
the
homeowner
is
non-‐resident.
However,
this
could
introduce
a
bias
by
also
excluding
some
resident
homeowners,
who
are
likely
to
be
longer-‐term
and
more
educated
residents.
3
See
Dillman,
Don,
Jolene
Smyth
and
Leah
Christian.
2009.
Internet,
Mail,
and
Mixed-‐Mode
Surveys:
the
Tailored
Design
Method.
New
York:
Wiley
&
Sons.
replied
to
the
survey
one
week
after
being
mailed
the
second
postcard
were
mailed
a
third
postcard.4
2) Each
postcard
gave
a
phone
number
where
a
homeowner
could
leave
a
voice
mail
message
(whether
in
English
or
Spanish)
requesting
that
a
paper
version
of
the
survey
(in
the
caller’s
chosen
language),
along
with
a
pre-‐addressed
and
pre-‐stamped
return
envelope,
be
mailed
to
her
home.
We
included
this
option
in
order
to
be
sure
that
potential
survey
respondents
lacking
the
knowledge
or
the
means
to
access
the
Internet
would
have
a
way
to
respond
to
the
survey
other
than
via
the
online
version.
3) We
contacted
the
moderators
of
listservs
for
ten
organizations
representing
the
interests
of
neighborhoods
surrounding
the
station
areas
in
Oakland,
Berkeley,
and
El
Cerrito,
and
asked
them
to
send
an
e-‐mail
to
their
lists
requesting
that
single
family
house
owner-‐
occupants
respond
to
the
online
version
of
the
survey
(as
in
#1
above)
or
telephone
to
request
a
hard
copy
version
(as
in
#2).
This
e-‐mail
was
written
in
both
English
and
Spanish,
and
largely
mirrored
the
wording
of
the
postcard.
As
a
result
of
this
effort,
six
of
the
organizations
contacted
their
members
requesting
that
they
respond
to
the
survey.
In
addition,
an
online
newspaper
in
El
Cerrito
published
a
story
urging
those
of
its
readers
eligible
to
respond
to
the
survey
to
do
so.
As
an
inducement
for
their
participation,
survey
respondents
were
offered
the
opportunity
to
enter
into
a
raffle
drawing
for
three
prizes:
a
$200
gift
certificate
for
a
popular
local
electronics
store,
a
$150
BART
pass,
and
a
$100
BART
pass.
Sample
size,
response
rates
and
bias:
We
mailed
postcards
to
2,529
households.
A
first
subsample
of
1,551
households
was
split
almost
exactly
evenly
amongst
the
five
station
areas.
Following
lower
level
of
responses
from
the
Del
Norte
and
Plaza
station
areas,
we
decided
to
have
each
of
them
comprise
30%
of
the
properties
in
the
second
subsample
of
1,001
households,
with
the
other
40%
split
equally
amongst
the
three
other
station
areas.
We
received
and
acted
upon
31
separate
intelligible
requests,
received
via
telephone
voice
messages,
to
mail
written
versions
of
the
survey
instrument
(all
of
them
in
English)
to
potential
respondents,
18
of
whom
ended
up
filling
out
and
returning
the
survey.
We
used
the
minimum
sample
size
formula
nmin
=
p(1-‐p)(Zc/E)2
to
calculate
our
targets
for
the
desired
number
of
survey
responses.
For
an
estimate
of
the
prevalence
of
secondary
units
amongst
owner-‐occupied
SFR
properties
in
the
study
area,
assuming
that
p
=
0.2
(corresponding
to
the
anticipated
probability
of
a
given
SFR
having
a
secondary
unit
being
20%),
Zc
=
1.645
(the
Z
score
corresponding
to
a
10%
confidence
interval,
typically
used
in
planning
studies)
and
E
=
10%
(a
sampling
error
of
10%,
also
commonly
used
in
planning
studies)
yields
a
minimum
sample
size
of
43.3.
Therefore,
a
defensible
secondary
unit
prevalence
estimate
would
require
44
survey
responses
(i.e.
43.3
rounded
upwards
to
the
nearest
whole
number).
4
As
described
further
in
the
following
section,
the
sample
was
broken
into
two
groups.
Non-‐respondents
in
the
first,
larger
sub-‐sample
received
three
postcards,
while
non-‐respondents
in
the
second,
smaller
sub-‐
sample
received
only
two.
We
decided
to
forego
sending
the
third
postcard
to
the
second
sub-‐sample
after
we
were
satisfied
that
we
had
received
enough
responses
for
statistical
validity,
as
is
discussed
in
the
following
section.
However,
relying
upon
survey
results
to
draw
statistically
meaningful
inferences
about
the
nature
of
the
secondary
unit
population
requires
many
more
than
44
responses.
Using
the
most
conservative
value
of
p,
0.5,
yields
a
minimum
sample
size
of
68
secondary
units.
Because
most
responses
(perhaps
four
out
of
five,
depending
on
the
actual
secondary
unit
prevalence
rate)
would
be
from
properties
lacking
secondary
units,
we
would
need
on
the
order
of
350
survey
responses
to
be
able
to
make
supportable
generalizations
about
various
characteristics
of
the
secondary
units
in
our
study
area.
As
it
happened,
we
obtained
enough
survey
responses
to
meet
both
the
lower
threshold
for
the
secondary
unit
prevalence
rate
and
the
higher
threshold
for
secondary
unit
characteristics.
We
received
a
grand
total
of
515
survey
responses.
Of
these,
334
responses
originated
from
the
postcard
mailings
to
2,529
households,
corresponding
to
a
response
rate
of
13.2%.
These
334
responses
easily
exceed
the
threshold
of
44
survey
responses
needed
for
a
secondary
unit
prevalence
ratio
amongst
owner-‐occupied
properties
in
the
study
area.5
Of
the
334
responses
originating
from
the
postcards,
316
were
online
and
18
were
mailed.
The
remainder
of
the
515
survey
responses,
181,
originated
as
responses
to
the
e-‐mails
sent
out
to
neighborhood
group
listservs.
None
of
the
515
total
survey
responses
were
in
Spanish;
all
were
in
English.
Of
the
515
responses,
87
were
reported
by
respondents
to
be
from
properties
with
secondary
units.
Using
a
stricter
definition
for
secondary
unit
that
we
employed
(discussed
below),
we
determined
only
81
of
the
87
responses
to
actually
be
from
properties
with
secondary
units.
Still,
81
survey
responses
from
properties
with
secondary
units
comfortably
exceeds
the
minimum
threshold
of
68
responses
needed
for
statistically
defensible
generalizations
about
secondary
unit
characteristics,
as
discussed
two
paragraphs
above.
Leaving
aside
the
thresholds
for
statistical
significance,
we
are
still
left
with
the
question
of
whether
the
response
set
that
we
obtained
is
generally
representative
of
the
population
of
SFR
properties
in
the
study
region.
From
a
geographical
standpoint,
at
least,
the
responses
are
reasonably
well-‐balanced,
with
El
Cerrito
somewhat
over-‐represented
and
Richmond
significantly
under-‐represented
in
the
response
set
as
compared
to
the
cities’
share
of
station-‐area
SFR
lots,
but
with
no
glaring
discrepancies
otherwise
observable
(Table
1).
As
discussed
in
the
results
section,
responding
homeowners
reported
being
from
households
that
were,
on
average,
disproportionately
high-‐income,
well-‐educated
and
comprised
of
white
non-‐Hispanic
people.
Finally,
the
lack
of
a
single
Spanish
language
response
out
of
the
515
received
suggests
that
immigrant
homeowners,
at
least
Spanish-‐speaking
ones,
are
not
well-‐represented
in
the
sample.
In
summary,
while
the
responses
are
sufficiently
numerous
as
to
be
likely
to
be
broadly
representative
of
the
population
of
SFR
properties
within
the
station
areas,
the
foregoing
cautions
should
be
kept
in
mind
when
interpreting
the
results.
5
We
consider
responses
sourced
from
postcards,
rather
than
the
entire
set
of
responses,
to
be
the
appropriate
data
set
to
use
for
calculating
the
secondary
unit
prevalence
estimate,
since
the
postcards
were
mailed
to
a
truly
random
sample
of
owner-‐occupied
SFR
properties
lying
within
the
station
areas
within
the
study
corridor.
For
all
other
estimates,
we
rely
on
the
full
response
set,
not
just
the
portion
that
originated
from
postcards.
Table
1.
Geographic
distribution
of
survey
responses.
Homeowner
survey
results
While
some
of
the
returned
surveys
included
non-‐responses
to
certain
questions,
seven
out
of
the
515
returned
surveys
had
so
many
missing
responses
as
to
be
non-‐usable.
In
the
end,
therefore,
we
used
508
out
of
the
515
returned
surveys
in
the
response
set.
The
following
sections
report
survey
results
according
to
various
themes.
Note
that
while
results
for
particular
questions
for
the
entire
data
set
are,
in
many
cases,
numerous
enough
to
be
statistically
significant,
results
for
individual
cities
in
most
cases
are
not,
and
are
therefore
not
presented
here.
Secondary
unit
prevalence.
As
discussed
in
the
previous
section,
we
considered
only
81
of
the
87
reported
secondary
units
to
meet
our
minimum
standard
for
actually
qualifying
as
secondary
units:
namely,
having
at
least
one
bathroom.6
Of
the
60
survey
responses
reporting
secondary
units
that
were
sourced
from
postcard
mailings
(as
opposed
to
solicitation
from
neighborhood
listservs),
55
of
the
secondary
units
were
reported
to
have
bathrooms.
We
therefore
regard
16%,
or
55
responses
indicating
secondary
units
sourced
from
postcards
units
out
of
a
total
of
334
responses
sourced
from
postcards,
as
our
best
estimate
of
the
proportion
of
SFR
lots
in
station
areas
within
the
study
corridor
that
have
secondary
units.
Of
the
81
properties
in
the
total
response
set
(sourced
both
from
postcards
and
from
neighborhood
listservs)
reporting
at
least
one
secondary
unit,
69
properties
reported
having
one
secondary
unit,
11
properties
reported
having
two
extra
units,
and
one
reported
having
three
extra
units.
Physical
characteristics
and
construction
history.
As
can
be
seen
in
Figure
1,
the
secondary
units
in
the
sample
set
come
in
a
variety
of
physical
configurations,
with
the
most
common
being,
in
descending
order
of
frequency,
a
freestanding
structure
in
the
rear
yard
(approximately
one
third
of
the
total),
a
converted
first
floor
or
basement,
and
a
converted
garage,
with
these
three
categories
collectively
accounting
for
more
than
two
thirds
of
all
85
cases.7
6
Secondary
units
lacking
full
kitchen
facilities,
including
a
multi-‐burner
stove
and
an
oven,
can
potentially
overcome
these
deficiencies
for
the
purpose
of
functioning
as
fully
independent
units
by
relying
on
hot
plates
and/or
microwave
ovens.
By
contrast,
living
quarters
that
lack
indoor
plumbing
cannot
be
viewed
as
meeting
even
a
minimal
standard
for
an
independent
dwelling
unit.
7
Some
of
the
reported
results
have
a
number
of
responses
that
exceeds
81,
the
number
of
properties
in
the
overall
data
set
with
secondary
units
that
have
bathrooms.
This
is
because
some
properties
have
more
than
City
Postcards
mailed
(%)
Survey
responses
(%)
Station
area
SFR
lots
(%)
Albany
4.5
2.5
3.5
Berkeley
30.2
42.3
46.8
El
Cerrito
34.4
31.7
21.1
Richmond
9.0
1.6
8.6
Oakland
21.9
21.9
20.0
Total
100
100
100
Secondary
unit
homeowners
were
asked
whether
they
or
a
previous
owner
had
instigated
the
installation
of
their
unit:
23%
said
that
they
had
done
it,
while
the
remaining
77%
said
that
a
previous
homeowner
had
undertaken
the
work
(out
of
77
responses).
All
18
homeowners
that
reported
having
overseen
the
installation
of
their
secondary
unit
were
asked,
and
answered,
the
question
of
who
had
actually
done
the
physical
work
of
installing
the
unit.
The
most
common
responses
were
a
hired
contractor
(44%),
the
respondent’s
household
(22%),
and
a
combination
of
the
two
(22%).
A
friend
or
relative
had
done
the
work
in
one
case,
and
the
other
two
responses
were
categorized
as
“other.”
All
secondary
unit
homeowners
were
asked
if
their
unit
has
a
complete
kitchen
(defined
as
a
rangetop
stove,
oven
and
refrigerator):
88%
of
the
74
respondents
answered
that
they
did,
while
the
remaining
12%
answered
“no.”
Respondents
were
asked
the
number
of
bedrooms
in
their
secondary
units.
As
might
be
expected,
a
majority,
63%,
of
secondary
units
were
reported
to
be
either
studios
or
one-‐bedrooms,
but
there
were
also
significant
numbers
of
units
that
have
two
bedrooms
or
even
three
bedrooms
or
more
(Figure
2).
Figure
1.
The
physical
configuration
of
reported
secondary
units.
one
secondary
unit,
each
of
which
is
counted
separately
unless
otherwise
stated.
Thus,
the
data
in
Figure
2
is
derived
from
a
response
set
of
84
secondary
units,
but
these
units
are
associated
with
a
subset
of
the
81
properties
that
contain
at
least
one
secondary
unit.
0%
10%
20%
30%
40%
Apt
over
garage
A1ached
apt
behind
main
house
Detached
apt
behind
main
house
Freestanding
garage
w/
apt
Converted
garage
Built
as
mulBunit
structure
Converted
basement/1st
floor
Converted
aDc
Converted
rooms
in
main
house
Two
side-‐by-‐side
houses
on
one
lot
Other
Figure
2.
The
number
of
bedrooms
reported
for
the
secondary
units.
As
can
be
seen
in
Figure
3,
close
to
a
majority
of
homeowners
with
secondary
units
do
not
know
when
their
units
were
installed,
which
suggests
that
they
were
installed
at
least
prior
to
when
the
respondents
purchased
their
home.
Homeowners
representing
15
percent
of
the
84
units
for
which
responses
to
this
question
were
received
reported
that
their
secondary
units
had
been
installed
within
the
last
five
years.
For
secondary
units
with
particular
reported
ages
(i.e.
where
the
respondents
did
not
answer
“I
don’t
know”
to
this
question),
the
average
age
of
installation
for
the
unit
was
38
years.
Studio
27%
1
BR
36%
2
BRs
27%
3+
BRs
10%
Figure
3.
The
reported
age
of
responding
homeowners’
secondary
units.
Details
of
secondary
use,
occupancy,
and
rental.
The
majority
(85%
of
84
units)
of
secondary
units
are
reported
to
be
occupied
by
at
least
one
person.
Twelve
of
the
13
unoccupied
secondary
units
(92%)
are
reported
to
be
used
for
a
non-‐residential
purpose,
such
as
a
home
office,
workshop
or
studio
space,
while
the
owner
of
the
remaining
unit
is
seeking
a
tenant.
Of
the
71
secondary
units
that
are
occupied,
the
relationship
of
the
occupant(s)
to
the
homeowner
household
was
reported
for
70
of
them.
The
largest
number
(46%)
of
secondary
unit
occupant
households
consist
of
people
that
the
homeowner
did
not
previously
know,
26%
of
the
units
are
occupied
by
all
or
part
of
the
homeowner
household,
and
29%
are
occupied
by
relatives,
friends,
or
acquaintances
of
the
homeowner
household.
By
far
the
most
common
methods
that
homeowners
used
to
find
occupants,
besides
having
a
part
or
all
of
their
own
household
occupying
the
unit
(26%
of
69
responses),
are
by
already
knowing
the
occupants
(25%)8,
Craigslist
(25%),
and
referrals
from
a
known
person
(7%).
Other
methods
include
the
tenants
already
being
in
place
at
the
time
the
property
was
purchased
(4%),
the
University
of
California-‐Berkeley
Housing
Office
(3%),
and
a
combination
of
Craigslist
and
the
UC-‐
Berkeley
Housing
Office
(6%).
In
4%
of
the
responses,
the
secondary
unit
is
separately
owned
by
its
occupant
as
a
condominium
or
tenancy-‐in-‐common
(TIC).
8
There
is
an
inconsistency
between
the
reported
figures
of
29%
of
units
being
occupied
by
friends,
family
or
acquaintances
that
are
not
part
of
the
homeowner
household,
and
25%
of
unit
occupant
households
having
been
sourced
via
the
homeowner
household
having
previously
known
the
occupants.
This
discrepancy
arises
because
there
are
two
cases
in
which
the
respondent
reports
that
the
unit
is
occupied
by
one
or
more
acquaintances,
but
that
the
tenants
were
sourced
via
a
personal
referral.
In
these
cases,
presumably,
the
tenants
have
become
acquaintances
after
moving
in,
or
else
the
homeowner
knew
them
by
reputation
beforehand
but
had
not
actually
met
them.
Thus,
in
short,
the
inconsistency
between
the
two
figures
arises
from
an
ambiguity
in
the
precise
nature
of
the
term
“acquaintance,”
which
was
not
rigorously
defined
in
the
survey
instrument.
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
50%
As
can
be
seen
in
Table
2,
34
out
of
41
respondents
reported
that
the
occupants
of
their
secondary
units
pay
rent,
while
the
remaining
seven
stay
for
free
or
perform
in-‐kind
work
in
exchange
for
living
in
the
secondary
unit.
Rents
charged
vary
widely.
(Please
refer
to
the
discussion
of
the
results
from
a
separate
analytical
technique,
the
rental
advertisement
data
study
later
in
this
document,
for
a
much
more
detailed
examination
of
market
rents
in
secondary
dwelling
units.)
Of
the
tenants
paying
rent,
75%
are
people
that
the
homeowner
did
not
previously
know,
while
the
rest
are
friends,
acquaintances,
or
relatives.
All
seven
of
the
tenants
staying
for
free
or
in
exchange
for
in-‐
kind
rent
are,
not
surprisingly,
friends
or
relatives
of
the
homeowner.
Table
2.
Rent
paid
by
unit
size
(number
of
bedrooms).
Studio
1
BR
2
BRs
3+
BRs
Total
Free
or
in-‐kind
rent
2
3
1
1
7
Paid
rent
7
12
10
5
34
Minimum
paid
rent
($)
650
550
650
1,250
-‐-‐
Maximum
paid
rent($)
1,750
1,500
2,975
2,200
-‐-‐
Average
paid
rent($)
948
931
1,486
1,483
-‐-‐
Demographics
of
homeowner
households.
As
is
evident
in
Table
3,
households
with
or
without
secondary
units
differ
little
demographically,
though
they
are
quite
different
from
residents
of
the
five
East
Bay
cities
as
a
whole
(including
renter
households).
The
results
indicate
that,
as
might
be
expected,
households
with
secondary
units
have
somewhat
smaller
household
sizes
(both
with
respect
to
adults
and
children)
than
non-‐secondary
unit-‐owning
homeowner
households.
In
addition,
adults
in
secondary
unit-‐owning
households
are
slightly
older,
earn
slightly
less,
are
slightly
more
likely
to
be
college
educated,
and
are
somewhat
less
likely
to
be
non-‐Hispanic
whites
than
their
non-‐secondary
unit-‐owning
peers.
The
differences,
however,
are
quite
small.
Secondary
unit-‐owning
households
are
notable
more
than
anything
for
their
broad
demographic
similarity
to
their
non-‐secondary
unit-‐owning
peers.
Note
that
these
results
indicate,
as
mentioned
in
an
earlier
section,
that
the
respondent
households
are,
in
general,
highly
affluent,
comprised
of
highly
educated
adults,
and
comprised
of
relatively
few
adults
of
color.
While
the
income
and
education
results
can
be
regarded
with
some
skepticism,
due
to
a
possible
tendency
for
people
to
exaggerate
their
income
and
education
status,
we
should
nevertheless
bear
in
mind
that
the
survey
results
herein
appear
to
be
skewed
towards
a
well-‐off
group
of
homeowners,
and
calibrate
our
interpretation
of
the
results
accordingly.
Table
3.
Demographics
of
homeowner
households
with
and
without
secondary
units,
relative
to
East
Bay
study
area.
Households
East
Bay*
With
Secondary
Units
Without
Secondary
Units
Demographic
characteristic
Average
Response
set
size
Average
Response
set
size
Adults
in
household
1.77
1.73
77
HHs
1.95
404
HHs
Children
in
household
0.71
0.37
68
HHs
0.43
405
HHs
Avg.
age
of
adults
in
household
35.4
49.6
117
adults
48.9
771
adults
Household
income
($)
$77,775
$105,000
61
HHs
$109,000
382
HHs
White
non-‐Latino
adults
(%)
34.2
70.8
120
adults
79.7
784
adults
African
American
adults
(%)
22.9
4.2
120
adults
5.2
784
adults
Latino
adults
(%)
20.7
6.7
120
adults
4.2
784
adults
Asian/Pacific
Islander
adults
(%)
18.8
12.5
120
adults
8.0
784
adults
Adults
with
high
school
or
less
(%)
36.7
5.8
120
adults
4.7
783
adults
College
graduates
(%)
37.1
82.5
120
adults
78.4
783
adults
Male
adults
(%)
47.9
42.5
120
adults
46.2
783
adults
*
Includes
data
from
Albany,
Berkeley,
El
Cerrito,
Oakland,
and
Richmond
from
either
Census
2010
or
the
2005-‐2009
American
Community
Survey.
Income
is
adjusted
to
2010
dollars.
Demographics
of
secondary
unit
occupant
households.
Survey
respondents
reported
that
73
adults
were
living
in
48
secondary
units,
for
an
average
of
1.52
adults
per
secondary
unit.9
The
average
age
for
the
68
adults
residing
in
the
secondary
units
with
reported
ages
is
38.6
years,
with
17%
of
the
adults
ranging
in
age
from
18
to
25
and
another
4%
with
ages
65
and
greater.
One
or
more
children
were
reported
to
be
living
in
only
four
out
of
the
50
secondary
units
(8%)
for
which
this
question
was
answered.
The
racial/ethnic
breakdown
of
the
62
adult
secondary
unit
occupants
for
whom
this
information
was
provided
is
as
follows:
80%
white
non-‐Latino,
8%
Latino,
8%
Asian/Pacific
Islander,
and
5%
mixed
race.
Of
the
63
adult
secondary
unit
occupants
with
reported
gender,
37%
of
them
are
male.
Perceptions
of
secondary
units
by
non-‐secondary
unit
homeowners.
The
421
survey
respondents
that
reported
not
having
at
least
one
secondary
unit
were
asked
a
series
of
questions
regarding
their
attitudes
towards
secondary
units.
One
question,
which
drew
404
responses,
asked
these
homeowners
why
they
do
not
already
have
a
secondary
unit.
As
can
be
seen
in
Figure
4,
the
potential
market
for
future
secondary
unit
installations
–
that
subset
of
homeowners
that
either
has
already
attempted,
is
actively
planning
to
or
might
consider
installing
a
secondary
unit
–
is
about
30%
of
homeowners
that
do
not
already
have
such
an
extra
unit
on
their
properties.
9
Note
that
homeowner
households
were
asked
to
answer
questions
about
the
occupants,
if
any,
of
the
secondary
units
on
their
properties,
if
any.
While
we
considered
attempting
to
ask
homeowners
to
pass
along
a
separate
survey
to
the
occupants
of
their
secondary
units
for
them
to
fill
out
directly,
we
concluded
that
this
was
unlikely
to
result
in
a
high
response
rate,
as
well
as
being
cost-‐prohibitive.
Consequently,
figures
on
secondary
unit
occupants
should
be
interpreted
with
the
caution
that
they
originate
from
information
provided
by
a
member
of
the
homeowner
household
rather
than
directly
from
the
occupants
themselves.
Also
note
that
figures
for
secondary
unit
occupant
demographics
exclude
instances
in
which
the
entire
homeowner
household
occupies
the
secondary
unit
–
those
are
included
in
the
previous
section
summarizing
homeowner
household
demographics.
Figure
4.
Stated
reasons
that
homeowners
lacking
a
secondary
unit
do
not
already
have
one.
Homeowners
that
reported
having
tried
but
failed
to
install
secondary
units
were
asked
for
the
reason
for
the
failure.
Their
responses
can
be
seen
in
Figure
5.
The
leading
reason
is
an
inability
to
fit
the
required
amount
of
off-‐street
parking,
but
other
prominent
answers
are
excessive
cost
and
difficulty
of
the
city’s
regulatory
process,
among
others.
One
Berkeley
homeowner
exemplified
all
three
of
these
leading
reasons
for
not
having
installed
a
secondary
unit:
“To
add
a
unit,
one
must
have
noncontiguous
parking
spaces.
I
would
have
to
put
a
car
in
my
front
yard,
which
is
probably
illegal,
or
give
up
the
back
yard.
I
have
an
old
garage/studio
I
would
love
to
turn
into
an
additional
unit
but
[cost]
and
permits
and
parking
make
it
difficult
[or]
impossible.”
Homeowners
lacking
secondary
units
on
their
properties
were
asked
if
they
knew
of
secondary
units
on
other
properties
on
their
block.
Out
of
407
responses,
61.9%
were
in
the
affirmative,
8.6%
of
homeowners
stated
that
there
were
no
secondary
units
on
their
street,
and
29.5%
were
not
sure.
Homeowners
indicating
that
they
knew
of
secondary
units
on
their
street
were
then
asked
whether
or
not
they
felt
that
these
units
had
a
negative
impact
on
their
quality
of
life,
and
if
so,
what
the
leading
negative
impact
was.
Of
the
255
responses
to
this
question,
by
far
the
most
common
(62.0%)
was
that
the
nearby
secondary
units
have
no
negative
impact
on
the
respondent’s
quality
of
life.
The
leading
negative
impact
reported
(24.7%)
was
an
increased
number
of
cars
competing
for
on-‐street
parking.
In
the
words
of
an
El
Cerrito
homeowner:
“Neighbors
who
have
built
extra
units
and
rented
their
property
have
created
tensions
and
problems
because
now
there
are
six
cars
vying
for
one
off-‐street
shared
parking
spot.”
Don't
want
one
38%
Never
considered
it
31%
Tried
but
failed
7%
Might
want
to
21%
Planning
on
it
3%
Figure
5.
Reasons
for
failure
of
attempted
secondary
unit
installations.
Other
negative
impacts
were
reported
in
much
smaller
numbers,
including
the
presence
of
renters
on
a
street
mostly
occupied
by
homeowner
households
(5.1%),
density-‐related
impacts
(3.1%),
unruly
tenants
(2.0%),
aesthetic
impacts
(1.2%),
and
reduced
property
values
(0.8%).
(Figure
6.)
An
Oakland
homeowner
disturbed
by
the
prevalence
of
secondary
units
on
the
local
block
expressed
some
of
the
concerns
about
increased
density
that
arose
with
a
minority
of
homeowners:
“The
extra
units,
which
are
rentals,
make
it
difficult
to
know
all
the
neighbors.
People
seem
to
come
and
go.
It
makes
the
block
seem
less
friendly
[than]
when
there
was
one
family
per
property.
I
used
to
know
all
my
neighbors,
now
I
know
only
a
few.
I've
also
seen
a
big
increase
in
the
number
of
cars
on
my
block
since
these
additional
units
were
added.
With
all
the
additional
cars,
if
I
didn't
have
my
one
parking
spot
on
my
property,
I
would
often
have
to
park
a
block
or
so
away.
That
would
make
me
think
twice
about
going
out
at
night.
It
also
makes
it
difficult
when
friends
visit.”
0
2
4
6
8
10
12
City
process
DisrupBon
Too
expensive
Couldn't
fit
parking
Lot
too
small
Other
Figure
6.
Perceived
impact
of
secondary
units
on
quality
of
life.
Parking.
All
homeowner
households
were
asked
to
report
the
number
of
off-‐street
parking
spaces
on
their
property
(whether
in
the
form
of
driveway
spaces,
spaces
in
a
garage,
or
others),
the
number
of
cars
their
household
parks
off-‐street
on
their
property,
and
the
number
of
cars
their
household
parks
on-‐street.
Households
reporting
having
secondary
units
with
occupants
were
asked
how
many
cars
those
occupants
park
off-‐street
on
the
parcel
and
how
many
cars
they
park
on-‐street.
While
the
majority
of
households
occupying
secondary
units
have
at
least
one
car,
22%
(out
of
responses
from
37
properties)
have
no
car
at
all
(Figure
7).
0
50
100
150
200
No
negaBve
impact
Yes
-‐-‐
aestheBc
impacts
Yes
-‐-‐
presence
of
renters
Yes
-‐-‐
reduced
property
values
Yes
-‐-‐
unruly
tenants,
safety
and/or
noise
Yes
–
parking
Yes
-‐-‐
density-‐related
impacts
Figure
7.
Number
of
cars
parked
by
households
in
secondary
units.
Note:
If
a
given
property
has
more
than
one
secondary
unit,
all
cars
reported
as
having
been
generated
by
those
units
are
included
in
the
total
for
that
property.
Meanwhile,
households
with
at
least
one
secondary
unit
provide
off-‐street
parking
spaces
at
a
rate
that
is
statistically
indistinguishable
from
houses
lacking
secondary
units
(Table
4).
Table
4.
Average
number
of
parking
spaces
reported.
With
secondary
unit
Without
secondary
unit
mean
#
of
off-‐street
spaces
1.86
1.67
standard
deviation
1.23
1.16
#
of
observations
66
406
significantly
different
@
95%
level?
No
Note:
Statistical
significance
for
the
difference
of
means
is
determined
via
a
standard
two-‐tailed
T-‐test
with
a
95%
significance
threshold.
Properties
with
secondary
units
generate
a
greater
number
of
cars
from
the
combined
homeowner
household
and
secondary
unit
occupant
household(s),
as
can
be
seen
in
Table
5.
Properties
with
secondary
units
generate,
on
average,
3.35
cars,
while
properties
without
them
generate
an
average
of
two
cars,
a
difference
of
almost
1.4
cars
in
the
average
case.
However,
when
properties
with
no
more
than
one
secondary
unit
are
excluded,
the
average
number
of
cars
generated
drops
to
2.91,
or
a
(statistically
significant)
increase
of
only
about
0.9
cars
above
properties
that
lack
secondary
units.10
As
seen
in
the
standard
deviation
figures,
properties
with
secondary
units
exhibit
more
10
In
our
opinion,
properties
with
only
one
secondary
unit
are
the
relevant
analytic
category.
While
many
cities
might
be
expected
to
ease
the
regulatory
burdens
that
currently
prevent
SFR
properties
from
legally
adding
one
secondary
unit,
it
is
difficult
to
imagine
many
cases
in
which
cities
would
be
willing
to
consider
allowing
two
secondary
units
in
areas
zoned
for
single-‐family
houses.
For
that
reason,
examining
data
sets
that
exclude
properties
with
more
than
one
secondary
unit
allows
us
to
gain
insight
into
the
parking
impacts
no
cars,
22%
1
car,
40%
2
cars,
19%
3
cars,
8%
3+
cars,
8%
5
cars,
3%
variability
than
those
without
them,
implying
that
a
wider
range
of
parking
behaviors
is
seen
on
such
properties
than
in
properties
that
lack
secondary
units
(something
that
is
presumably
related
to
the
large
variation
in
car
ownership
habits
of
secondary
unit
households
seen
previously
in
Figure
7).11
Table
5.
Number
of
cars
generated
by
units.
Secondary
Units
1
or
more
0
1
mean
#
of
total
cars
3.35
2.00
2.91
standard
deviation
1.92
1.05
1.60
#
of
observations
43*
405
35*
significantly
different
@
95%
level?
Yes
significantly
different
@
95%
level?
Yes
Note:
The
difference
of
means
test
is
done
in
the
same
way
described
in
Table
4.
*8
were
imputed.
The
imputation
technique
is
described
in
footnote
#9.
Table
6
shows
the
number
of
cars
parked
on-‐street
reported
to
be
generated
by
properties
with
and
without
secondary
units,
since
on-‐street
parking
is
of
concern
to
some
neighbors
of
secondary
unit
properties,
as
evidenced
earlier
by
Figure
6.
Properties
with
secondary
units
generate
an
extra
1.1
cars
parked
on-‐street,
on
average,
vis-‐à-‐vis
properties
without
secondary
units.
This
discrepancy,
however,
drops
to
only
0.7
cars
when
properties
with
more
than
one
secondary
unit
are
excluded
from
the
analysis.
While
interpreting
these
figures
in
tandem
with
the
results
shown
in
Table
6
should
be
done
with
caution,
particularly
since
we
did
not
impute
on-‐street
car
generation
for
some
of
the
incomplete
records
as
we
did
with
the
number
of
total
cars,12
nevertheless
an
intuitive
picture
begins
to
emerge.
It
would
appear
that
people
residing
on
properties
with
secondary
units
“pick
up
more
of
the
slack”
of
their
available
off-‐street
parking
spaces
by
putting
more
of
them
to
use
than
their
peers
living
on
properties
lacking
secondary
units.
This
might
explain
why
properties
with
one
secondary
unit
generate
0.9
more
cars
than
properties
without
secondary
units,
but
result
in
only
of
the
house-‐plus-‐secondary
unit
configuration
that
would
become
much
more
common
under
the
policy
outcome
that
cities
would
be
most
likely
to
seek.
11
Note
that
some
of
the
data
records
for
total
and
overspill
cars
used
for
Tables
5
and
7,
respectively,
are
imputed,
as
is
indicated
in
the
charts.
We
undertook
the
imputation
in
an
attempt
to
boost
the
sample
size
for
the
difference
of
means
computations,
since
many
of
the
records
for
secondary
unit
properties
lack
all
of
the
data
needed
to
compute
total
and
overspill
cars
generated.
Where
enough
data
was
available,
imputations
were
computed
by
selecting
a
random
value
from
an
assumed
normal
distribution
of
the
number
of
cars
per
person
calculated
from
the
set
of
non-‐secondary
unit
homeowner
households,
and
using
this
to
impute
the
number
of
cars
generated
by
a
secondary
unit
households
from
the
number
of
adults
reported
to
comprise
the
household.
12
Intuitively,
one
would
expect
that
on-‐street
parking
behavior
for
secondary
unit
occupants
might
differ
markedly
from
that
of
homeowner
households.
Thus
we
decided
that
it
would
not
be
helpful
to
attempt
to
impute
the
former
from
the
latter.
0.7
additional
cars
parked
on-‐street,
despite
having
on
average
essentially
no
extra
off-‐street
parking
available
(as
seen
in
Table
4).13
Table
6.
The
number
of
cars
parked
on-‐street
generated
by
properties
with
and
without
secondary
units.
Number
of
Secondary
Units
1
or
more
0
1
mean
#
of
cars
parked
on-‐street
1.97
0.90
1.61
standard
deviation
1.61
0.85
1.32
#
of
observations
35
406
27
significantly
different
@
95%
level?
Yes
significantly
different
@
95%
level?
Yes
If
all
of
“the
slack”
of
off-‐street
parking
spaces
were
to
be
taken
up
(which
one
might
envision
occurring
under
conditions
in
which
all
available
on-‐street
parking
spaces
are
in
use),
how
many
cars
would
households
living
on
properties
with
secondary
units
push
out
onto
the
street
as
compared
to
their
peers
living
on
parcels
without
such
units?
One
metric
that
can
shed
some
light
on
the
matter
is
“overspill
cars,”
which
we
define
to
be
the
difference
between
the
number
of
cars
generated
by
the
household(s)
residing
on
the
parcel
(as
summarized
in
Table
5)
and
the
number
of
off-‐street
parking
spaces
available
on
the
property
(as
summarized
in
Table
4).
Intuitively,
a
measure
of
“overspill
cars”
is
equivalent
to
the
number
of
cars
that
are
pushed
into
on-‐street
parking
once
all
available
off-‐street
parking
spaces
on
the
parcel
have
been
used.
(A
negative
value
for
the
overspill
cars
metric
indicates
that
the
property
has
more
than
enough
off-‐street
parking
spaces
for
the
cars
owned
by
the
household
or
households
living
on
the
parcel.)
Table
7
shows
the
average
number
of
overspill
cars
generated
by
properties
with
and
without
secondary
units.
Again,
while
the
discrepancy
between
the
two
categories
is
large
(1.1
cars)
and
statistically
significant,
it
drops
to
a
much
lower
level
(0.7
cars),
albeit
one
that
is
still
statistically
significant,
when
properties
with
more
than
one
secondary
unit
are
excluded.
Properties
with
only
one
secondary
unit
do,
however,
exhibit
an
appreciably
higher
standard
deviation
in
the
number
of
overspill
cars
that
they
generate
compared
to
properties
without
secondary
units
(2.23
versus
1.40).
This
suggests
the
need
for
flexible
policy
tools
that
can
span
the
wide
variety
of
car
ownership
propensities
exhibited
by
households
living
on
properties
with
secondary
units.
13
In
the
words
of
a
Berkeley
homeowner,
“No
one
around
here
parks
in
their
garage,
for
various
reasons
I
guess,
thus
the
garage
parking
spaces
are
only
available
in
‘theory’.“
There
are
several
reasons
why
this
might
be
so.
For
instance,
an
off-‐street
space
may
be
arranged
in
tandem
with
another
off-‐street
space
that
is
used
frequently,
requiring
the
owner
of
one
of
the
cars
to
move
her
automobile
to
make
way
for
the
other.
A
driveway
or
garage
entrance
may
be
narrow
and
require
that
a
driver
back
into
it
slowly
and
carefully.
In
many
neighborhoods,
on-‐street
parking
directly
in
front
of
the
main
house
is
frequently
available
at
least
some
of
the
time,
and
could
be
perceived
as
the
most
convenient
option.
Table
7.
Overspill
cars
per
unit.
Number
of
Secondary
Units
1
or
more
0
1
mean
#
of
cars
parked
on-‐street
1.47
0.35
1.09
standard
deviation
2.32
1.40
2.23
#
of
observations
43*
406
35*
significantly
different
@
95%
level?
Yes
significantly
different
@
95%
level?
Yes
Note:
We
define
the
measure
of
“overspill
cars”
as
the
difference
between
the
number
of
cars
generated
by
a
property
(as
summarized
in
Table
5)
and
the
number
of
off-‐street
parking
spaces
available
on
that
property
(as
summarized
in
Table
4).
Intuitively,
it
measures
the
parking
impact
of
a
parcel
on
its
street
and
(possibly
on
nearby
streets)
if
all
of
its
off-‐street
parking
spaces
are
being
fully
used.
*8
were
imputed.
The
imputation
technique
is
described
in
footnote
#9.
RENTAL
INTERNET
ADVERTISEMENT
STUDY
Purpose
and
data
collection
methodology
While
the
survey
of
homeowners
described
above
provides
many
insights
into
the
characteristics,
behaviors
and
attitudes
of
homeowners,
the
survey
is
of
limited
usefulness
for
systematically
analyzing
the
open
rental
market
for
secondary
units,
and
the
ways
in
which
it
differs
from
the
corresponding
market
for
rental
dwellings
that
are
not
secondary
units.
The
survey
did
not
collect
a
sufficient
quantity
of
rental
market
data
on
secondary
units
to
make
it
possible
for
us
to
relate
these
units’
characteristics
to
the
level
of
rent
charged
via
a
hedonic
model
or
other
quantitative
technique.
In
addition,
the
questions
that
we
would
have
needed
to
add
to
the
survey
instrument
in
order
to
capture
some
of
the
information
important
to
a
hedonic
analysis
might
have
risked
making
the
survey
instrument
too
long,
thus
reducing
response
rates.
We
therefore
pursued
a
separate
analysis
in
which
we
collected
data
about
rental
units
–
both
secondary
units
and
other
types
of
dwellings
–
from
advertisements
placed
on
Craigslist.com,
a
website
widely
used
by
San
Francisco
Bay
Area
tenants
and
landlords.
Craigslist
ads
offered
several
advantages.
First,
we
could
assume
that
landlords
placing
advertisements
seeking
tenants
on
Craigslist
were
not
seeking
to
rent
to
people
they
already
knew,
and
thus
were
not
likely
planning
to
offer
“friends
and
family”
discounts.
By
contrast,
as
previously
described,
about
26%
of
the
secondary
units
in
the
survey
response
set
are
occupied
by
the
homeowner
household,
while
a
further
25%
are
occupied
by
at
least
one
person
known
to
the
homeowning
household
prior
to
move-‐in,
whether
as
an
acquaintance,
friend,
or
relative.
Second,
we
could
readily
collect
data
on
secondary
and
non-‐secondary
rental
units
simultaneously,
thus
making
it
easy
to
compare
the
two.
Finally,
landlords
advertising
on
Craigslist
presumably
have
a
direct
financial
incentive
(one
not
present
for
survey
respondents)
to
list
all
pertinent
features
of
the
rental
unit
that
they
are
offering
in
order
to
attract
prospective
tenants.
During
the
three-‐month
period
of
May,
June,
and
July
of
2011,
we
collected
unit
characteristics
from
all
Craigslist
advertisements
for
secondary
units14
listed
for
rent
and
located
in
the
general
corridor15
under
study.
We
also
collected
information
from
a
similar
number
of
ads
for
non-‐
secondary
rental
units.16
After
discarding
records
that
were
unusable
for
lack
of
essential
information
(such
as
rent
charged,
number
of
bedrooms,
and
number
of
bathrooms),
we
ended
up
with
a
sample
of
174
secondary
units
and
164
non-‐secondary
units.
In
usable
ads,
amenities
such
as
off-‐street
parking,
dishwashers
and
on-‐site
laundry
were
assumed
to
not
be
in
place
unless
mentioned,
due
to
the
direct
incentive
for
landlords
to
trumpet
all
attractive
features
about
the
apartments
they
were
seeking
to
rent.
One
exception
to
this
general
rule
was
for
cooking
facilities,
which
were
assumed
to
be
fully
provided
(i.e.,
to
include
the
typical
minimal
configuration
of
a
four-‐burner
range
top
stove
and
oven)
unless
otherwise
mentioned,
due
to
typical
market
expectations
concerning
kitchens.
Locational
characteristics
were
captured
via
three
dimensions:
a
measure
of
the
unit’s
walkability
(from
the
publicly
available
website
www.walkscore.com),
which
we
took
to
serve
as
a
proxy
for
the
general
level
of
neighborhood
amenity,
an
index
of
the
crime
rate
for
the
local
area
(from
the
proprietary
website
www.neighborhoodscout.com),
and
the
unit’s
proximity
to
a
freeway,
which
we
assumed
to
coincide
with
environmental
blight
due
to
noise,
fumes,
visual
impacts
and
so
forth.17
In
a
few
cases,
we
needed
to
impute
the
values
for
certain
variables
that
could
not
be
recorded
for
every
record
in
the
data
set.
Some
independent
variables
that
we
originally
intended
to
collect,
and
that
indeed
might
have
proved
to
be
useful,
such
as
the
number
of
units
in
the
apartment’s
building,
had
to
be
dropped
from
our
model,
because
they
could
not
be
inferred
or
plausibly
imputed
from
every
single
advertisement.
Details
about
all
of
the
variables
included
in
the
model
are
described
in
Appendix
3.
14
A
unit
was
deemed
to
be
a
“secondary
unit”
for
these
purposes
if
this
characteristic
could
be
ascertained
from
the
advertisement,
such
as
a
picture
depicting
a
secondary
unit,
or
text
giving
some
tell-‐tale
description
such
as
“a
cottage
behind
the
main
house.”
Note
that
for
these
purposes
the
analytic
distinction
between
a
“secondary
unit”
and
a
unit
in
a
duplex
or
triplex
is
bound
to
be
blurry.
The
inherent
characteristic
of
a
secondary
unit
of
interest
for
this
analysis
is
that
it
is
located
on
a
property
with
only
two
or
three
units,
possibly
(though
not
necessarily)
with
an
owner-‐operator
living
on-‐site
or
in
the
immediate
vicinity.
15
This
corridor
is
roughly
defined
as
that
portion
of
the
urban
East
Bay
that
extends
from
the
bay
shore
on
the
west
to
the
topographic
break
in
the
hills
to
the
east
(corresponding
more
or
less
to
the
“Flatlands,”
as
described
in
WP-‐2012-‐01),
from
approximately
one
mile
north
of
El
Cerrito
Del
Norte
BART
on
the
north
end
to
approximately
one
mile
south
of
MacArthur
BART
on
the
south
end.
While
this
region
generally
coincides
with
the
areas
considered
in
the
rest
of
this
study,
it
is
not
limited
to
only
the
BART
station
areas
within
this
corridor.
16
When
collecting
data
for
non-‐secondary
units,
we
kept
gathering
these
advertisements
until
their
numbers
roughly
matched
those
of
the
secondary
unit
advertisements.
Once
the
latter
exceeded
the
former,
we
resumed
gathering
the
latter.
Thus,
while
the
non-‐secondary
unit
advertisements
are
a
subset
of
the
entire
universe
of
ads
placed
for
units
lying
within
the
corridor,
one
can
assume
that
the
sample
was
randomly
drawn.
17
Because
of
the
high
density
of
freeway
coverage
in
the
study
area
and
the
consequent
ease
of
accessing
a
freeway
from
anywhere
within
it
(no
location
in
the
study
area
lies
more
than
three
miles
from
an
on-‐ramp),
we
follow
the
lead
of
Cervero
and
Landis
in
treating
close
freeway
proximity
in
the
urban
East
Bay
as
a
disamenity
with
no
countervailing
proximity-‐based
advantages
for
slightly
greater
distances.
See
Cervero,
R.
and
Landis,
J.
1997.
“Twenty
Years
of
the
Bay
Area
Rapid
Transit
System:
Land
Use
and
Development
Impacts.”
Transportation
Research
A,
Vol
31,
No.
4,
pp.
309-‐333.
The
dependent
variable
used
for
the
hedonic
analysis
is
the
adjusted
monthly
rent.
To
facilitate
a
true
“apples
to
apples”
comparison,
the
rent
for
each
unit
was
adjusted
to
account
for
the
availability
or
lack
thereof
of
off-‐street
parking
(whether
free
or
paid)
and
the
partial
or
total
inclusion
of
three
utility
costs
(gas/electric,
water/sewer,
and
cable/Internet)
with
monthly
rent.
While
four
of
the
secondary
units
offered
opportunities
for
tenants
to
offset
their
rent
through
on-‐
site
work
(such
as
childcare,
yard
maintenance,
etc.),
this
type
of
in-‐kind
work
was
considered
to
be
above
and
beyond
the
rental
transaction
(since
it
required
labor
to
be
performed)
and
thus
was
not
factored
into
adjusted
rent.
Note
that
for
the
full
data
set,
we
chose
to
not
normalize
rent
by
dwelling
unit
floor
area,
since
square
footages
were
only
directly
provided
in
24%
of
the
real
estate
ads
that
were
recorded
in
the
data
set
(the
rest
were
imputed).
(We
conducted
hedonic
model
runs
using
both
the
full
data
set
and
the
subset
that
includes
only
records
with
directly
observable
square
footages,
as
is
described
below
in
the
discussion
of
the
hedonic
model
results.)
With
the
variables
constructed
as
described
above,
we
were
able
to
conduct
two
types
of
analysis
on
the
data
set:
1)
a
series
of
difference-‐of-‐means
tests
among
the
various
variables,
in
order
to
ascertain
the
ways
in
which
secondary
unit
characteristics
differ
from
those
of
non-‐secondary
units,
and
to
gather
summary
statistics
on
these
two
submarkets;
and
2)
a
hedonic
model,
which
allowed
us
to
compare
how
various
unit
amenities
and
locational
characteristics
are
differentially
priced
into
rents
in
the
secondary
unit
and
non-‐secondary
unit
rental
markets.
The
results
of
these
analyses
are
treated
in
turn
in
the
following
two
sections.
Table
8
shows
the
distribution
of
dwellings
in
the
data
set
by
municipal
jurisdiction.
This
distribution
suggests
that
a
cautionary
note
be
kept
in
mind
when
interpreting
these
results:
60%
(secondary
units)
to
65%
(non-‐secondary
units)
of
the
apartments
from
each
category
are
located
in
Berkeley
and
the
adjacent
unincorporated
community
of
Kensington.
By
comparison,
the
Berkeley
flatlands
and
Kensington
account
for
only
38%
of
the
SFR
parcels
of
the
study
area
(Table
8).
The
concentration
of
rental
units
in
and
around
Berkeley
may
reflect
the
heavy
influence
that
the
local
University
of
California
campus
exerts
on
the
rental
market
of
the
corridor
under
study,
and
perhaps
of
the
particular
ubiquity
of
Craigslist
as
a
means
of
locating
rental
housing
opportunities
among
the
highly
educated
population
of
Berkeley.
The
first
concern
is
perhaps
tempered
by
the
timing
of
the
data
collection,
which
occurred
in
the
late
spring
and
early
summer,
a
period
during
which
university
students
are
unlikely
to
seek
housing
(although
it
may
be
a
time
during
which
students
are
vacating
housing
and
leaving
the
region,
and
landlords
are
seeking
to
fill
the
resulting
vacancies).
As
can
be
seen,
Oakland
and
Emeryville
are
also
somewhat
over-‐
represented
(though
less
so
than
Berkeley/Kensington).
El
Cerrito
and
Albany
are
mildly
under-‐
represented,
while
Piedmont
and
the
relevant
portion
of
Richmond
are
highly
under-‐represented.
Table
8.
Distribution
of
units
and
SFR
parcels
by
jurisdiction.
Location
of
non-‐secondary
and
secondary
units
in
data
set
(left
and
central
columns,
respectively)
compared
against
proportion
of
total
study
region
SFR
parcels
(right
column).
Results
in
the
right
column
are
estimated
from
the
following
sources:
Center
for
Community
Innovation,
and
2009
Housing
Elements
for
the
cities
of
Albany,
Emeryville,
and
Piedmont
and
for
the
County
of
Contra
Costa
(which
contains
unincorporated
Kensington).
Table
9
gives
summary
statistics
for
secondary
and
non-‐secondary
units
on
a
variety
of
variables,
along
with
the
results
of
a
difference-‐of-‐means
test
(T
test)
performed
on
each
variable.
The
statistics
reveal
a
number
of
intuitive,
yet
notable,
differences
between
rental
units
that
are
secondary
units
versus
rental
units
that
are
not,
suggesting
that
secondary
units
occupy
a
distinctive
niche
within
the
overall
rental
market.
Jurisdiction
Non-‐secondary
units
(%)
Secondary
units
(%)
SFR
parcels
in
jurisdiction
(%)
Berkeley
flatlands
65
60
38
Oakland
flatlands
21
24
19
City
of
Albany
4
6
13
El
Cerrito
flatlands
8
9
13
S
Richmond
flatlands
2
1
7
City
of
Piedmont
<
1
<
1
12
Total
100%
100%
100%
non-‐secondary
units*
secondary
units**
significantly
different***
physical
characteristics
#
of
bedrooms
1.52
0.99
yes
#
of
bathrooms
1.11
1.00
yes
floor
area
(sf)
703
559
yes
secure
entrance
(%)
12.8
2.9
yes
economic
characteristics
adjusted
monthly
gross
rent
($)
1,358
1,106
yes
adjusted
monthly
gross
rent
per
($/sf)
2.17
2.30
no
adjusted
monthly
gross
rent
as
%
of
AMI
68.4
62.4
yes
owner
on
site
(%)
2.4
52.9
yes
Parking
no
off-‐street
parking
(%)
45.7
87.4
yes
free
off-‐street
parking
(%)
31.7
10.9
yes
paid
off-‐street
parking
available
(%)
22.6
1.7
yes
Amenities
coin-‐operated
laundry
(%)
46.3
0.0
yes
free
laundry
(%)
16.5
45.1
yes
full
kitchen
facilities
(%)
99.4
89.1
yes
Dishwasher
(%)
14.0
9.2
no
Microwave
(%)
15.1
26.8
yes
Utilities
included
gas/electric
(worst
=
0,
best
=
1)
0.052
0.454
yes
included
water/sewer
fees
(worst
=
0,
best
=
1)
0.377
0.583
yes
included
cable/Internet
(worst
=
0,
best
=
1)
0.040
0.419
yes
locational
attributes
Walkscore
(worst
=
0,
best
=
100)
84.5
72.9
yes
crime
index
(most
dangerous
=
0,
safest
=
100)
18.3
40.6
yes
freeway
proximity
(furthest
=
0,
closest
=
1)
0.241
0.1865
no
Table
9.
The
mean
value
of
various
characteristics
of
units,
divided
into
the
categories
of
non-‐secondary
units
(left
column,
N=164)
and
secondary
units
(central
column;
N=174).
The
right-‐hand
column
indicates
whether
each
variable
is
or
is
not
significantly
different,
between
the
non-‐secondary
and
secondary
unit
subsets,
at
the
95%
level
of
confidence.
First,
secondary
units
are
smaller,
with
on
average
one
half
of
a
bedroom
less
than
non-‐secondary
units,
one
tenth
of
a
bathroom
less,
and
about
140
sf
less
floor
area.
Not
surprisingly,
the
adjusted
rent
of
the
average
secondary
units
is
considerably
less
than
that
of
the
average
non-‐secondary
unit,
by
about
$250
per
month.
When
expressed
in
terms
of
Area
Median
Income,
in
the
manner
typical
of
subsidized
affordable
housing,
the
adjusted
rents
of
secondary
units
are
more
affordable
to
a
modest
though
significant
degree:
the
average
secondary
unit
is
affordable
to
a
household
earning
62%
of
AMI,
versus
about
68%
of
AMI
for
non-‐secondary
units.18
While
adjusted
rent
normalized
by
floor
area
is
slightly
higher
for
secondary
units
(albeit
not
to
a
statistically
significant
degree),
we
are
nevertheless
left
with
a
distinct
impression
that
secondary
units
are
more
likely
than
units
in
the
general
rental
market
to
provide
smaller,
more
affordable
housing
packages
for
the
households
that
are
seeking
them.
If
anything,
the
estimated
6%
difference
in
affordability
as
a
minimum
AMI
percentage
is
a
lower-‐bound
estimate,
since
non-‐secondary
units
advertised
on
Craisglist
are
likely
to
be
less
expensive,
on
average,
than
apartments
rented
via
other,
more
costly
means
(such
as
freeway
billboards,
posters
in
train
stations,
listings
in
glossy
apartment
magazines,
etc.).
In
addition,
the
secondary
units
advertised
on
Craigslist
were
being
rented
to
strangers,
and
therefore
excluded
the
share
of
secondary
units
(which
we
estimated
in
the
homeowner
survey
to
be
51%)
that
are
rented
to
family,
friends
or
acquaintances
for
free
or
reduced
rents
in
many
cases.
If
it
were
possible
to
include
the
secondary
units
occupied
by
people
already
known
to
the
homeowner
household
in
the
analysis,
doubtless
the
disparity
in
affordability
between
secondary
and
non-‐secondary
units
would
be
even
greater.
It
is
also
worth
noting
that
secondary
units
in
the
dataset
are
far
more
likely
(53%
versus
2%)
to
have
owners
residing
on-‐site
than
their
non-‐secondary
counterparts.
This
is
not
surprising
when
one
considers
that
owner-‐occupancy
of
one
of
the
two
units
on
a
property
consisting
of
a
main
house
and
a
secondary
unit
is
mandated
by
law
in
the
cities
of
Albany,
Berkeley,
El
Cerrito,
Piedmont,
and
Oakland.19
The
strong
contrast
in
owner-‐occupancy
between
non-‐secondary
and
secondary
units
likely
accounts
for
many
of
the
differences
between
these
two
housing
submarkets.
It
also
provides
indirect
but
strong
evidence
that
secondary
units
are
much
more
likely
to
be
owned
and
operated
by
amateur
landlords
who
earn
a
portion
or
a
majority
of
their
living
from
activities
other
than
real
estate,
whereas
non-‐secondary
units
are
much
more
likely
to
be
owned
and
operated
by
real
estate
professionals
or
firms.
18
Area
Median
Income
is
taken
from
the
2011
income
limit
data
published
by
the
US
Department
of
Housing
and
Urban
Development
for
the
Oakland-‐Fremont
HMFA,
which
includes
all
of
Alameda
and
Contra
Costa
Counties,
which
in
turn
include
the
entirety
of
this
report’s
study
area.
“Affordability”
is
taken
here
to
correspond
to
the
HUD
standard
wherein
a
household
expends
no
more
than
30%
of
its
gross
income
on
housing
rent
plus
utility
costs.
The
Area
Median
Income,
again
following
standard
practice,
is
computed
for
a
household
size
that
equates
to
1.5
persons
per
bedroom
in
the
unit
(or
1
person
per
studio
apartment).
Our
method
of
adjusting
rent
to
account
for
utility
costs
roughly
mirrors
standard
practice
in
affordable
housing,
since
utility
costs,
when
not
covered
by
the
landlord,
are
effectively
deducted
from
rent
for
affordability
ratio
computation
purposes.
19
The
reason
that
owner-‐occupancy
amongst
what
we
are
classifying
as
“secondary
units”
for
the
purposes
of
the
rental
Internet
advertisement
study
falls
short
of
100%
is
some
combination
of
1)
non-‐compliance
with
the
law
by
property
owners
and
2)
the
inclusion
of
units
in
the
“secondary
unit”
data
set
that,
legally
speaking,
are
actually
duplexes
or
triplexes
and
thus
not
subject
to
the
owner-‐occupancy
requirement.
Possibly
as
a
consequence
of
much
higher
levels
of
non-‐professional
management,
secondary
units
appear
to
handle
parking
for
their
tenants
in
a
markedly
different
manner
from
non-‐secondary
rental
units.
This
can
be
seen
from
the
distribution
of
the
three
mutually
exclusive
categories
of
parking
leasing
for
these
two
submarkets:
no
parking
provided,
free
off-‐street
parking,
and
off-‐
street
parking
offered
for
a
fee.
Secondary
units
are
far
more
likely
(87%)
to
offer
no
off-‐street
parking
whatsoever
(whether
free
or
otherwise)
than
non-‐secondary
units
(46%).
In
the
relatively
rare
cases
where
off-‐street
parking
is
offered
for
secondary
units,
it
is
much
likelier
to
be
offered
for
free
(11%)
than
for
a
fee
(2%).
A
comment
from
a
Berkeley
homeowner
typifies
the
most
common
scenario
with
secondary
units:
“There
are
three
households
on
our
block
with
extra
housing
units
…
No
one
on
our
block
has
a
usable
garage
and
most
do
not
or
cannot
use
their
driveways
for
parking.”
Differences
in
amenities
between
secondary
and
non-‐secondary
units
are
less
dramatic
than
is
the
case
with
parking,
but
a
couple
of
comments
concerning
these
variables
are
nonetheless
in
order.
First,
the
likelihood
of
secondary
units
offering
a
fully-‐equipped
kitchen
is
lower
than
for
their
non-‐
secondary
peer
units
to
a
modest
but
statistically
significant
degree.
This
could
well
be
reflective
of
the
prevalence
of
unpermitted
secondary
units
(as
discussed
in
WP-‐2012-‐02).
This
could
also
explain
the
somewhat
higher
incidence
of
microwave
oven
provision
in
secondary
units,
i.e.,
microwave
ovens
could
be
viewed
as
a
means
by
which
landlords
attempt
to
partially
compensate
tenants
for
offering
substandard
cooking
facilities.
Indeed,
13
of
the
18
secondary
units
lacking
cooking
facilities
(72%)
offer
microwave
ovens,
whereas
the
one
non-‐secondary
unit
that
does
not
have
full
cooking
appliances
does
have
a
microwave
oven.
Some
portion
or
all
of
three
utility
bills
(gas/electric,
water/sewer,
and
cable/Internet)
are
markedly
and
statistically
significantly
more
likely
to
be
included
as
part
of
rent
in
secondary
units
than
in
non-‐secondary
units.
Note
that
this
should
not
be
taken
to
imply
that
secondary
units
are
therefore
a
better
deal
for
their
tenants
than
non-‐secondary
units,
because
these
utility
bills
have
been
factored
into
the
adjusted
rent
computed
for
each
unit
(as
described
in
the
previous
section).
It
does,
however,
mean
that
housing
packages
with
utility
payments
included
are
more
common
for
secondary
unit
tenants
than
for
their
peers
not
living
in
secondary
units.
This
discrepancy
is
likely
attributable
to
1)
the
higher
likelihood
of
sharing
utility
costs
with
an
on-‐site,
amateur
landlord
and
2)
the
greater
likelihood
of
single
connections
to
electric
lines
and
gas,
water
and
sewer
mains
being
shared
between
a
secondary
unit
and
other
units
on
the
property.
That
locational
attributes
are
not
dramatically
dissimilar
between
secondary
and
non-‐secondary
units
in
the
dataset
is
not
terribly
surprising,
given
that
the
corridor
from
which
the
units
were
selected
is
a
reasonably
homogeneous
environment
in
terms
of
its
urban
development
patterns.
Nonetheless,
we
see
that
secondary
units
are,
to
a
modest
degree,
located
in
less
walkable
and
lower-‐crime
neighborhoods,
and
are
less
likely
to
be
located
within
close
proximity
to
a
freeway,
than
is
the
case
for
non-‐secondary
units.
This
is
to
be
expected,
as
secondary
units
are
presumably
more
likely
to
be
located
in
predominantly
residential
and
lower-‐density
neighborhoods.
Hedonic
model
methodology
and
results
A
hedonic
model,
by
regressing
the
dependent
variable
(adjusted
monthly
rent)
against
the
characteristics
described
in
the
previous
section,
can
quantify
how
the
average
actor
in
the
housing
market
values
certain
unit
and
locational
attributes.
By
modeling
the
entire
dataset,
as
well
as
separately
modeling
the
roughly
equal
portions
comprised
of
secondary
and
non-‐secondary
units,
we
can
infer
whether
participants
in
these
two
submarkets
value
attributes
differently.
The
hedonic
model
is
specified
as
follows:
ln(adjusted _rent)=num_BRs +num _BAs +sqfootage +SECURE +OWNONSITE
+ON _STREET +COINOP _LAUND +FREE _LAUND +MICROW +DISHWASHER
+FULL _KITCHEN +Walkscore +crime_index +freeway +ε
Here
adjusted_rent
(rent
adjusted
according
to
the
method
described
in
the
first
section
of
this
chapter)
is
the
dependent
variable;
num_BRs
and
num_BAs
indicate
the
number
of
bedrooms
and
bathrooms,
respectively;
sqfootage
is
the
unit’s
interior
square
footage;
SECURE
is
a
dummy
variable
indicating
whether
or
not
the
unit
has
a
secure
entry
(i.e.
an
entry
that
lies
behind
a
locked
gate
or
door
with
limited
access);
OWNONSITE
is
a
dummy
variable
indicating
whether
or
not
the
owner
of
the
rental
unit
resides
on-‐site
in
another
unit;
ON_STREET
is
a
dummy
variable
indicating
whether
or
not
the
tenant
must
find
parking
on
the
street
if
she
has
a
car;
COINOP_LAUND
is
a
dummy
indicating
whether
or
not
the
unit
has
coin-‐operated
laundry
machines
on
the
property;
FREE_LAUND
is
a
dummy
set
to
1
only
if
the
unit
has
access
to
laundry
machines
on
the
property
for
no
charge;
MICROW
is
a
dummy
variable
indicating
whether
or
not
a
microwave
oven
is
included
in
the
unit;
DISHWASHER
is
a
dummy
variable
indicating
whether
or
not
the
unit
includes
a
dishwasher;
FULL_KITCHEN
is
a
dummy
variable
indicating
whether
or
not
the
unit
has
appliances
befitting
a
full
kitchen
(four-‐burner
stove
and
full-‐sized
oven);
Walkscore
is
a
freely
available
index
of
the
unit’s
address’s
“walkability”
from
0
(worst)
to
100
(best);
crime_index
is
a
proprietary
index
of
crime
for
a
given
location
based
on
published
crime
statistics
ranging
from
0
(worst)
to
100
(best);
and
freeway
is
an
index
of
proximity
to
the
nearest
freeway
that
ranges
from
0
(more
than
1000’
from
a
freeway)
to
2
(within
500’
of
a
freeway).20
(As
previously
mentioned,
a
fuller
description
of
the
construction
of
the
variables,
along
with
the
imputation
procedures
used
to
populate
variables
with
sparse
coverage,
is
provided
in
Appendix
3.)
The
semi-‐log
form
of
the
model
allows,
in
the
case
of
a
coefficient
computed
by
the
model
for
a
particular
independent
variable
,
for
to
be
interpreted
as
the
proportional
change
in
the
dependent
variable
resulting
from
a
unit
change
in
X
.
Thus,
if
were
computed
by
the
model
to
be
0.2,
or
20%,
for
example,
then
that
would
signify
a
20%
modeled
increase
in
adjusted_rent
20
In
the
previous
section,
this
variable
was
normalized
to
run
from
a
minimum
of
0
to
a
maximum
of
1,
rather
than
from
0
to
2,
for
ease
of
interpretation.
k
X
ek −1
ek −1
with
an
increase
in
X
by
one
unit.
This
form
therefore
allows
for
an
intuitive
interpretation
of
the
monetary
contribution
to
rents
from
the
various
independent
variables
that
are
modeled.21
Results
from
full
data
set.
Table
10
allows
a
comparison
of
the
results
from
three
model
runs:
one
performed
on
the
full
data
set,
and
the
other
two
performed
on
two
subsets,
one
including
only
the
non-‐secondary
units
and
the
other
including
only
the
secondary
units.
As
might
be
expected,
the
increase
to
adjusted
rent
resulting
from
the
addition
of
a
bedroom
is
significant
and
large
in
each
case.
The
same
is
largely
true
for
bathrooms
(although
the
magnitude
of
the
increase
is
smaller),
except
that
this
variable
is
omitted
from
the
secondary
unit-‐only
model
run
as
a
result
of
every
single
secondary
unit
in
the
data
set
(all
174
of
them)
having
one
bathroom.
These
results
are
intuitive,
as
is
the
positive
and
significant
effect
of
square
footage
(above
and
beyond
what
is
captured
in
the
bedroom
and
bathroom
variables)
in
the
full
data
set.
It
is
interesting
to
note
that
the
presence
of
an
owner
on-‐site,
all
else
being
equal,
results
in
a
statistically
significant
reduction
in
adjusted
rent
of
almost
9%
in
the
case
of
secondary
units,
and
none
at
all
in
non-‐secondary
units.
This
result
gives
support
to
the
supposition
that
amateur,
on-‐
site
property
owners
are
less
concerned
about
maximizing
rental
revenues
from
tenants,
and
are
perhaps
more
concerned
with
maintaining
long-‐term
amity
with
their
renters.
From
the
tenant’s
point
of
view,
this
result
suggests
that
secondary
units
residing
on
owner-‐occupied
properties
are
often
a
housing
bargain.
21
See,
for
example,
Sirmans,
G.
Stacy
and
David
A.
Macpherson.
2003.
“The
Value
of
Housing
Characteristics.”
Washington,
DC:
National
Association
of
Realtors,
December.
Table
10.
Hedonic
model
results
(excluding
imputed
square
footages).
Results
are
shown
for
hedonic
model
runs
performed
with
the
full
data
set,
non-‐secondary
units
only,
and
secondary
units
only.
Note
that
because
the
model
was
run
in
the
semi-‐log
form,
and
because
what
is
displayed
here
represents
the
exponentiation
of
the
model
coefficients,
each
independent
variable’s
value
shown
here
can
be
interpreted
as
the
percentage
change
in
the
dependent
variable
(adjusted
rent)
that
would
result
from
an
increase
of
one
unit
of
that
independent
variable.
Units
All
Non-‐secondary
Secondary
units
regression
statistics
#
of
data
points
336
164
172
r-‐squared
of
model
run
0.51
0.64
0.41
modeled
variable
coefficients
(%)
#
of
bedrooms
21.0
21.5
20.0
#
of
bathrooms
13.7
15.9
-‐-‐
floor
area
(per
100
sf)
1.6
1.5
1.7
secure
entry
-‐0.8
4.6
-‐12.0
owner
on-‐site
-‐6.4
-‐1.2
-‐8.7
no
off-‐street
parking
-‐4.8
-‐10.0
2.6
free
laundry
11.9
24.1
6.5
coin-‐operated
laundry
3.1
6.9
-‐-‐
microwave
oven
5.0
16.8
0.6
dishwasher
11.0
1.1
20.8
full
kitchen
facilities
43.2
24.2
43.1
Walkscore
(per
10
points)
0.7
2.6
0.2
crime
index
(per
10
points)
0.4
-‐0.3
0.0
adjacent
to
freeway
-‐10.0
-‐12.0
-‐3.3
Coefficients
that
are
in
boldface
are
statistically
significant
at
the
95%
level.
Failing
to
offer
on-‐street
parking
depresses
rents
in
non-‐secondary
units
by
about
9%,
while
the
comparable
coefficient
for
secondary
units
is
not
statistically
significant.
The
9%
decrease
in
rent
for
non-‐secondary
units
compares
to
similar
decreases,
found
in
a
hedonic
study
of
for-‐sale
housing
San
Francisco
in
the
mid-‐1990s,
of
12%
for
SFR
and
13%
for
condominium
properties.22
The
9%
reduction
found
for
East
Bay
non-‐secondary
rental
units
suggests
that
while
provision
of
off-‐street
parking,
or
lack
thereof,
is
priced
into
transactions
in
mainstream
rental
units,
this
is
not
the
case
in
the
market
for
East
Bay
secondary
units.
Because
the
vast
majority
of
secondary
units
in
this
region
appear
to
offer
no
off-‐street
parking
at
all
(as
shown
in
the
previous
section),
whether
paid
or
free,
it
could
be
that
tenants
seeking
secondary
units
have
little
to
no
expectation
of
off-‐street
parking
being
offered,
and
therefore
do
not
take
this
factor
into
account
when
formulating
their
bids
for
housing.
Amenities
are
reflected
in
rents
to
varying
degrees.
For
instance,
the
availability
of
free
laundry
machines
on
the
premises
shows
up
as
a
strong
influence
on
rent
in
the
overall
data
set
and
for
non-‐secondary
units,
but
not
in
secondary
units.
Perhaps
this
is
because
secondary
unit
tenants
are
so
often
able
to
share
the
use
of
laundry
machines
with
the
main
house
that
on-‐site
laundry
availability
is
less
of
a
prized
feature
than
in
the
general
rental
market,
where
a
lack
of
laundry
machines
usually
means
relying
on
a
commercial
laundromat.
Meanwhile,
it
is
difficult
to
understand
the
large
magnitudes
of
the
coefficients
in
favor
of
microwave
ovens
(for
non-‐
secondary
units)
and
dishwashers
(for
secondary
units),
particularly
the
former.
Perhaps
the
offering
of
microwave
ovens
in
mainstream
apartments,
which
are
much
less
likely
to
have
substandard
kitchens
(as
discussed
in
the
previous
section),
is
associated
with
other
attractive
kitchen
amenities,
while
the
presence
of
microwaves
in
secondary
units
could
more
often
be
a
partial
recompense
for
the
substandard
kitchens
more
likely
to
be
seen
in
that
submarket
(evidence
for
which
was
also
presented
in
the
previous
section).
That
a
large
and
statistically
significant
effect
of
kitchen
appliances
on
rents
is
only
seen
for
secondary
units
can
be
interpreted
as
the
consequence
of
market
participants
taking
into
account
the
wide
variation
of
kitchen
quality
in
secondary
units
(whereas
mainstream
rental
units,
as
previously
discussed,
tend
to
have
more
uniformly
up-‐to-‐standard
kitchens).
It
is
interesting
that
none
of
the
three
locational
effects
have
significant
effects
on
rent,
with
the
exception
of
freeway
proximity,
which
depresses
the
rents
offered
for
non-‐secondary
units.
Perhaps
the
perceived
benefit
of
proximity
to
commercial
and
other
neighborhood
amenities,
prized
by
some
tenants,
that
is
reflected
in
a
high
Walkscore
is
negated
by
decreased
tranquility
for
others.
Finally,
East
Bay
renters
in
general
appear
to
not
have
a
great
deal
of
price
sensitivity
to
crime
rates.
22
Jia,
W.
and
M.
Wachs.
1998.
Parking
and
Affordable
Housing.
Access
13.
See
also
Litman,
T.
2009.
Parking
Requirement
Impacts
on
Housing
Affordability.
Victoria,
CA:
Victoria
Transportation
Policy
Institute,
in
which
it
was
argued
that
subsidized
rental
housing
developments
could
reduce
development
costs
–
and
thus,
presumably,
rents
–
by
10%
if
off-‐street
parking
did
not
have
to
be
provided.
Lastly,
it
is
worth
noting
that
the
model
run
for
the
non-‐secondary
units
has
a
considerably
higher
value
than
that
for
the
secondary
units
(0.64
versus
0.41).
This
can
be
interpreted
as
evidence
that
the
rental
market
for
mainstream
rental
units
is
more
transparent
(given
the
usage
of
real
estate
professionals
of
rental
brokers,
apartment
magazines,
etc.)
and
thickly
traded.
On
the
flip
side,
one
could
surmise
that
the
secondary
unit
rental
market,
in
its
much
heavier
reliance
on
amateur
actors,
exhibits
more
unpredictability
in
rent
levels
as
a
result
of
varying
experience
levels
amongst
its
landlords,
informal
renting
practices,
less
sophisticated
tenant
sourcing
(such
as
via
Craigslist),
and
the
like.
Results
using
only
non-‐imputed
square
footages
Because
conventional
hedonic
studies
almost
always
include
actual,
rather
than
imputed,
unit
floor
areas,
we
re-‐ran
the
hedonic
model
described
above
for
a
restricted
data
set
that
only
included
units
for
which
actual
unit
square
footage
amounts
had
been
specified
as
a
check
on
the
validity
of
the
results
from
the
full
data
set.
The
results
from
the
reduced
data
set
are
shown
in
Table
11.
The
results
highlight
the
tension
between
using
data
points
(i.e.,
properties)
that
are
complete
versus
using
a
larger
number
of
data
points.
They
also
illustrate
the
difficulties
of
collecting
data
on
a
black
market
phenomenon
–
in
a
typical
hedonic
analysis
of
rental
apartments,
having
an
actual
floor
area
value
for
each
dwelling
unit
in
the
model
would
not
be
an
issue.
Restricting
the
data
set
to
only
those
units
with
known
square
footages
reduces
the
size
of
the
data
set
by
more
than
three-‐quarters
overall
(73%
for
the
non-‐secondary
units
and
79%
for
the
secondary
units).
On
the
other
hand,
the
values
increase
appreciably:
from
0.51
to
0.64
for
the
full
data
set,
from
0.64
to
0.88
for
the
non-‐secondary
units,
and
from
0.41
to
0.60
for
the
secondary
units.
It
would
appear
that
having
actual,
as
opposed
to
imputed,
square
footages
for
the
units
has
a
major
impact
on
the
predictive
power
of
the
hedonic
model.
r2
r2
Table
11.
Hedonic
model
results
(excluding
imputed
square
footages).
Hedonic
model
results
are
shown
for
a
restricted
data
set
in
which
only
units
with
known
floor
areas
are
included.
(By
contrast,
Table
10
shows
the
results
from
model
runs
in
which
units
with
imputed
floor
areas
are
also
included.)
As
in
Table
10,
each
independent
variable’s
value
shown
here
can
be
interpreted
as
the
percentage
change
in
the
dependent
variable
(adjusted
rent)
that
would
result
from
an
increase
of
one
unit
of
that
independent
variable.
Units
All
Non-‐secondary
Secondary
units
regression
statistics
#
of
data
points
80
44
36
r-‐squared
of
model
run
0.64
0.88
0.60
modeled
variable
coefficients
(%)
#
of
bedrooms
13.0
37.1
1.5
#
of
bathrooms
16.7
3.0
-‐-‐
floor
area
(per
100
sf)
5.3
2.1
5.2
secure
entry
5.3
22.5
-‐26.3
owner
on-‐site
-‐2.9
-‐6.0
-‐3.6
no
off-‐street
parking
-‐4.2
-‐1.5
-‐0.3
free
laundry
-‐5.4
13.6
-‐9.3
coin-‐operated
laundry
-‐1.9
-‐0.1
-‐-‐
microwave
oven
2.3
10.8
-‐1.3
dishwasher
18.5
28.5
31.0
full
kitchen
facilities
51.5
5.6
70.4
Walkscore
(per
10
points)
-‐1.1
10.3
-‐3.2
crime
index
(per
10
points)
0.9
-‐1.4
0.9
adjacent
to
freeway
-‐11.8
-‐12.9
-‐8.6
Coefficients
that
are
in
boldface
are
statistically
significant
at
the
95%
level.
In
the
model
runs
on
the
restricted
data
set,
the
number
of
bedrooms
has
a
large
and
significant
effect
in
the
expected
direction
on
adjusted
rents
for
non-‐secondary
units,
although
none
for
secondary
units.
Number
of
bathrooms
ceases
to
have
a
significant
effect
–
presumably
this
variable
has
been
absorbed
by
square
footage,
whose
coefficient
becomes
large
and
significant
in
the
case
of
secondary
units.
In
the
restricted
data
set,
a
secure
entry
emerges
as
a
highly
valued
and
significant
amenity.
Meanwhile,
off-‐street
parking
recedes
as
a
significant
amenity,
as
do
free
laundry
and
microwave
ovens.
Dishwashers
and
full
kitchens
remain
very
important
and
significant
for
secondary
units.
Owner-‐occupancy
ceases
to
be
significant
for
secondary
units.
Finally,
the
locational
factors
remain
virtually
unchanged
in
the
reduced
data
set.
Thus,
in
interpreting
the
results
of
the
hedonic
data
study,
we
must
exercise
some
caution
with
the
results
presented
from
the
full
data
set
with
respect
to
the
importance
of
owner-‐occupancy
in
secondary
units,
and
the
importance
of
off-‐street
parking
provision,
free
laundry,
and
microwave
ovens
in
non-‐secondary
units,
which
cease
to
be
significant
in
the
reduced
data
set.
Secondary
units
in
the
reduced
data
set
remain
highly
sensitive
to
the
quality
of
kitchen
facilities
(as
measured
by
the
presence
of
a
dishwasher
and
of
a
full
kitchen).
Adjusted
rents
for
secondary
units
also
remain
much
more
unpredictable
(as
measured
by
value),
thereby
confirming
that
the
rental
market
for
secondary
units
is
more
idiosyncratic
than
that
for
non-‐secondary
units.
CONCLUSION
Using
both
the
homeowner
survey
and
the
rental
Internet
advertisement
study,
we
are
able
to
overcome
at
least
some
of
the
limitations
posed
by
the
opacity
of
the
market
for
secondary
unit
housing
and
to
gain
some
insights
into
the
nature
of
this
market.
From
the
homeowner
survey,
we
learn,
first
of
all,
that
secondary
units
are
a
widespread
and
prevalent,
rather
than
an
aberrational
and
exceptional,
phenomenon.
This
suggests
that
scholars
interested
in
the
functioning
of
the
rental
housing
market
in
the
San
Francisco
Bay
Area
(and
probably
other
high-‐cost
regions
like
it)
would
be
well-‐advised
to
take
account
of
the
existing
stock
of
secondary
units.
From
the
homeowner
survey
we
learn
about
the
physical
nature
of
secondary
units
in
the
urban
core
of
the
East
Bay.
For
instance,
most
units
are
freestanding
structures
or
converted
garages.
In
addition,
we
have
survey
evidence
that
suggests
that
homeowners
with
secondary
units
are
mainstream
insofar
as
they
appear
to
differ
little,
from
a
demographic
standpoint,
from
homeowners
who
lack
such
units.
The
survey
also
gives
us
an
indication
of
what
homeowners
who
lack
secondary
units
at
present
think
about
them.
Awareness
of
the
existence
of
secondary
units
is
high,
and
in
most
cases
the
units
themselves
are
seen
as
relatively
benign.
Where
secondary
units
are
viewed
negatively,
impacts
on
on-‐street
parking
resulting
from
their
tenants
are
the
most
common
reason.
Nonetheless,
the
potential
market
for
secondary
unit
installation
is
quite
sizable,
with
30%
of
homeowners
expressing
at
least
a
potential
interest
in
installing
them.
This
highlights
an
unusual
aspect
of
r2
secondary
units
in
relation
to
other
types
of
rental
housing:
while
they
are
sometimes
seen
to
pose
impacts
to
local
homeowners’
quality
of
life,
as
is
typical
with
rental
housing,
they
also,
quite
unusually,
are
often
seen
as
something
from
which
homeowners
could
personally
benefit.
From
the
rental
Internet
advertisement
study
we
are
able
to
infer
that
even
most
of
the
secondary
units
rented
to
strangers
are
owned
and
operated
by
homeowners
who
live
on-‐site.
In
addition,
the
survey
results
tell
us
that
a
sizable
fraction
of
secondary
units
are
rented
to
acquaintances,
friends
or
family,
in
some
cases
for
free
and
in
other
cases,
we
presume,
for
reduced
rents.
Thus,
the
two
methods
combine
to
give
us
a
picture
of
secondary
units
as
frequently
(though
not
always)
forming
part
of
an
amateur-‐operated
rental
market,
with
some
informal
characteristics.
These
features
are
consistent
with
our
view
of
this
type
of
housing
as
a
black
market,
or
one
that
operates
at
least
partially
outside
of
governmental
taxation
and
regulation.
An
Albany
homeowner
provides
a
perfect
example
of
this
type
of
informality:
“We
are
on
a
corner
and
have
five
street
parking
places
as
well
as
a
driveway
and
a
garage
(three
places),
yet
we
were
forced
to
add
two
parking
spaces
when
we
added
our
[secondary]
unit.
We
are
now
in
the
process
of
illegally
removing
those
two
[newly-‐added
parking
spaces]
as
they
take
up
virtually
the
entire
back
yard.”
Whereas
one
would
generally
assume
compliance
with
off-‐street
parking
or
other
regulations
imposed
on
professionally-‐managed
rental
housing,
the
picture
differs
considerably
for
secondary
unit
housing.
The
rental
Internet
advertisement
study,
finally,
provides
fairly
strong
evidence
that
secondary
units
comprise
cheaper
housing
packages
to
small
households
of
one
or
two
people
than
what
is
commonly
available
in
the
unsubsidized
overall
rental
housing
market.
We
calculate,
as
a
lower-‐
bound
estimate,
that
this
discrepancy
corresponds
to
a
savings,
for
tenants
in
secondary
units,
of
6%
of
AMI
on
average.
Secondary
units
are
cheaper
for
a
variety
of
reasons:
they
tend
to
be
smaller,
they
are
less
likely
to
offer
off-‐street
parking,
they
are
less
likely
to
provide
interior
amenities
such
as
dishwashers,
and
they
are
more
likely
to
share
utility
costs
with
the
homeowner
or
with
other
units,
for
example.
The
hedonic
model,
using
data
collected
in
the
rental
Internet
advertisement
study,
provides
at
least
some
evidence
that
amateur
ownership
and
operation
has
a
positive
influence
on
secondary
unit
affordability,
although
more
research
will
be
needed
to
fully
test
this
claim.
We
are
therefore
left
with
a
striking
picture
of
an
important,
and
often
overlooked,
segment
of
the
existing
market
for
rental
housing
in
the
East
Bay,
and
one
that
appears
to
be
particularly
important
for
low-‐moderate
to
moderate
income
rental
housing.
While
our
two
means
of
probing
the
rental
market
for
secondary
units,
the
homeowner
survey
and
the
rental
Internet
advertisement
study,
have
provided
what
we
believe
to
be
important
insights
into
this
market,
much
more
work
will
be
needed
–
both
using
other
methods
and
also
studying
other
high-‐cost
housing
markets
–
to
paint
a
full
picture
of
the
elusive
phenomenon
of
existing
secondary
units.
Appendix
1:
Homeowner
survey
instrument
(hard
copy,
English-‐
language
version)
Thank you for participating in our study on housing in the East Bay. This survey
should be filled out by the owner (or by one of the owners if there is more than
one) of the property to which this survey was mailed. It should take about 15
minutes of your time to fill it out.
PLEASE NOTE THAT WE WILL MAINTAIN STRICT ANONYMITY FOR ALL
ANSWERS THAT YOU GIVE. YOUR IDENTITY AND INDIVIDUAL ANSWERS WILL
NOT BE SHARED WITH ANYONE OUTSIDE THE RESEARCH TEAM.
Extra Housing Units question
Many residential properties in this area include one or more Extra Housing Units in
addition to the main dwelling. Extra Housing Units are often rented out to tenants. Extra
Housing Units can take different forms, including a first floor or basement that has been
converted to a unit, a freestanding backyard cottage, a garage that has been turned into
an apartment, and others. (Please see images of several examples of Extra Housing
Units below.)
1. How many Extra Housing Units (see above for definition) are on your property?
(Circle one.)
a) I don’t have any extra housing units on my property.
b) 1
c) 2
d) 3
e) 4 or more
If your answer to question 1 was a (i.e., that you have no Extra Housing Units on
your property), then skip ahead to question 56. Otherwise, please answer the
following questions about each Extra Housing Unit that is on your property.
WE WILL ASK YOU A SERIES OF QUESTIONS ABOUT EACH EXTRA HOUSING
UNIT ON YOUR PROPERTY. THE FOLLOWING QUESTIONS ARE ABOUT THE
FIRST ONE, WHICH WE WILL CALL EXTRA HOUSING UNIT #1
2. When was the decision made to install Extra Housing Unit #1 on your property?
(Circle one.)
a) It happened while my household owned the property
b) It happened before my household owned the property
If you answered “It happened before my household owned the property” to
question 2, then skip to question 4. Otherwise, proceed to question 3.
3. Who did the actual construction work of installing Extra Housing Unit #1 on your
property? (Circle all that apply.)
a) I did it.
b) A hired contractor did the work.
c) A friend or relative did the work.
d) Other __________________________________________.
4. How would you best describe the physical layout of Extra Housing Unit #1? (Circle
one.)
a) Part or all of basement or first floor converted to an apartment
b) Garage converted to an apartment
c) Apartment above a garage
d) Attic converted to an apartment
e) Rooms inside main part of house converted to an apartment
f) Apartment behind main house and attached to it
g) Apartment behind main house and in its own separate structure
h) Other: Please describe _________________________
5. Does Extra Housing Unit #1 have its own complete kitchen (sink, range, and
refrigerator)?
Yes No (circle one)
6. Does Extra Housing Unit #1 have at least one complete bathroom (toilet, sink, and
shower/bath)?
Yes No (circle one)
7. How many bedrooms does Extra Housing Unit #1 have? (Circle one.)
a) None (it is a studio or efficiency unit)
b) 1
c) 2
d) 3 or more
8. About how many years ago was Extra Housing Unit #1 installed? (If less than one
year ago, just mark “0.” If you don’t know, check “I don’t know.”)
________ years ago _____ I don’t know
9. Is Extra Housing Unit #1 occupied by at least one person?
Yes No (circle one)
If you answered Yes to question 9, then skip ahead to question 11. Otherwise,
proceed to question 10.
10. Why is Extra Housing Unit #1 currently unoccupied? (Circle one.)
a) It needs physical work to be rentable.
b) It is vacant, but I am looking for a tenant.
c) It is being used as something other than an apartment (home office,
workshop, studio, etc).
d) Other reason: __________________________________________
If you answered question 10, then skip ahead to question 20 (if you have a
second Extra Housing Unit) or proceed to question 60 (if you don’t).
11. Which of the following best describes the relationship of the occupant(s) in Extra
Housing Unit #1 to you at the time of move-in? (Circle one)
a) My household lives in Extra Housing Unit #1.
b) Relative(s)
c) Friend(s)
d) Acquaintance(s)
e) I did not know the occupant(s) before move-in.
If your answer to question 11 was a), then skip to question 20 (if you have a
second Extra Housing Unit) or question 60 (if you don’t). Otherwise, please
proceed to question 12.
12. How did you find the occupant(s) in Extra Housing Unit #1? (Circle all that apply.)
a) I already knew the occupant(s)
b) Craigslist
c) Other Internet source
d) Newspaper classified ad
e) Referred by someone I know
f) Other: ____________________________
13. How much is the rent paid to your household by the occupant(s) in Extra Housing
Unit #1? If the occupant(s) are staying in Extra Housing Unit #1 for free, then mark “$0.”
$__________ per month
14. Please circle all of the following utilities that are included in the rent, if any, paid to
your household by the occupant(s) in Extra Housing Unit #1.
Water/Sewer Gas/Electricity Trash/Recycling Pickup
Telephone Cable TV Internet
15. How long has the person who has been living in Extra Housing Unit #1 for the
longest time lived there?
_________ years _____ less than one year (check if applicable)
_________ not sure
16. How many total cars do the people (if any) who are living on your property in Extra
Housing Unit #1 (if any) normally park on your property (i.e. not on the street)?
(Please choose one.)
a) none
b) 1
c) 2
d) 3 or more
17. How many total cars do the people (if any) who are living on your property in Extra
Housing Unit #1 (if any) normally park on the street within a five-minute walk of where
you live? (Circle one.)
a) none
b) 1
c) 2
d) 3 or more
18. How many children (i.e. people under age 18) live in Extra Housing Unit #1?
a) none
b) 1
c) 2
d) 3 or more
19. For each adult (i.e. person of age 18 or more) living in Extra Housing Unit #1, please
answer the following questions.
Question Adult #1 Adult #2 Adult #3 Adult #4
What is your
best
estimate of
this person’s
age?
_____ years
_____ years
_____ years
_____ years
How would
you best
describe the
ethnicity or
race of this
person?
(Circle all
that apply.)
White/Caucasian
Latino/Hispanic
Black/African
American
Asian/Pacific
Islander
American
Indian/Alaska
Native
Other:
_______________
Mixed Race
White/Caucasian
Latino/Hispanic
Black/African
American
Asian/Pacific
Islander
American
Indian/Alaska
Native
Other:
_______________
Mixed Race
White/Caucasian
Latino/Hispanic
Black/African
American
Asian/Pacific
Islander
American
Indian/Alaska
Native
Other:
_______________
Mixed Race
White/Caucasian
Latino/Hispanic
Black/African
American
Asian/Pacific
Islander
American
Indian/Alaska
Native
Other:
_______________
Mixed Race
What is this
person’s
gender?
(Circle one.)
Male
Female
Male
Female
Male
Female
Male
Female
20. THE FOLLOWING QUESTIONS ARE ABOUT EXTRA HOUSING UNIT #2. IF YOU
DON’T HAVE A SECOND EXTRA HOUSING UNIT, PLEASE SKIP TO QUESTION
#60. OTHERWISE, PROCEED.
When was the decision made to install Extra Housing Unit #2 on your property? (Circle
one.)
a) It happened while my household owned the property
b) It happened before my household owned the property
If you answered “It happened before my household owned the property” to
question 20, then skip to question 22. Otherwise, proceed to question 21.
21. Who did the actual construction work of installing Extra Housing Unit #2 on your
property? (Circle all that apply.)
a) I did it.
b) A hired contractor did the work.
c) A friend or relative did the work.
d) Other __________________________________________.
22. How would you best describe the physical layout of Extra Housing Unit #2? (Circle
one.)
a) Part or all of basement or first floor converted to an apartment
b) Garage converted to an apartment
c) Apartment above a garage
d) Attic converted to an apartment
e) Rooms inside main part of house converted to an apartment
f) Apartment behind main house and attached to it
g) Apartment behind main house and in its own separate structure
h) Other: Please describe _________________________
23. Does Extra Housing Unit #2 have its own complete kitchen (sink, range, and
refrigerator)?
Yes No (circle one)
24. Does Extra Housing Unit #2 have at least one complete bathroom (toilet, sink, and
shower/bath)?
Yes No (circle one)
25. How many bedrooms does Extra Housing Unit #2 have? (Circle one.)
a) None (it is a studio or efficiency unit)
b) 1
c) 2
d) 3 or more
26. About how many years ago was Extra Housing Unit #2 installed? (If less than one
year ago, just mark “0.” If you don’t know, check “I don’t know.”)
________ years ago _____ I don’t know
27. Is Extra Housing Unit #2 occupied by at least one person?
Yes No (circle one)
If you answered Yes to question 27, then skip ahead to question 29. Otherwise,
proceed to question 28.
28. Why is Extra Housing Unit #2 currently unoccupied? (Circle one.)
a) It needs physical work to be rentable.
b) It is vacant, but I am looking for a tenant.
c) It is being used as something other than an apartment (home office,
workshop, studio, etc).
d) Other reason: __________________________________________
If you just answered question 28, then skip ahead to question 34 (if you have a
third Extra Housing Unit) or proceed to question 60 (if you don’t).
29. Which of the following best describes the relationship of the occupants(s) in Extra
Housing Unit #2 to you at the time of move-in? (Circle one.
a) My household lives in Extra Housing Unit #2.
b) Relative(s)
c) Friend(s)
d) Acquaintance(s)
e) I did not know the tenant(s) before move-in.
If your answer to question 29 was a), then skip to question 38 (if you have a third
Extra Housing Unit) or question 60 (if you don’t). Otherwise, please proceed to
question 30.
30. How did you find the occupant(s) in Extra Housing Unit #2? (Circle all that apply.)
a) I already knew the occupant(s)
b) Craigslist
c) Other Internet source
d) Newspaper classified ad
e) Referred by someone I know
f) Other: ____________________________
31. How much is the rent paid to your household by the occupant(s) in Extra Housing
Unit #2? If the occupant(s) are staying in Extra Housing Unit #2 for free, then mark “$0.”
$__________ per month
32. Please circle all of the following utilities that are included in the rent, if any, paid to
your household by the occupant(s) in Extra Housing Unit #2.
Water/Sewer Gas/Electricity Trash/Recycling Pickup
Telephone Cable TV Internet
33. How long has the person who has been living in Extra Housing Unit #2 for the
longest time lived there?
_________ years _____ less than one year (check if applicable)
_________ not sure
34. How many total cars do the people (if any) who are living on your property in Extra
Housing Unit #2 (if any) normally park on your property (i.e. not on the street)?
(Please choose one.)
a) none
b) 1
c) 2
d) 3 or more
35. How many total cars do the people (if any) who are living on your property in Extra
Housing Unit #2 (if any) normally park on the street within a five-minute walk of where
you live? (Circle one.)
a) none
b) 1
c) 2
d) 3 or more
36. How many children (i.e. people under age 18) live in Extra Housing Unit #2?
a) none
b) 1
c) 2
d) 3 or more
37. For each adult (i.e. person of age 18 or more) living in Extra Housing Unit #2, please
answer the following questions.
Question Adult #1 Adult #2 Adult #3 Adult #4
What is your
best
estimate of
this person’s
age?
_____ years
_____ years
_____ years
_____ years
How would
you best
describe the
ethnicity or
race of this
person?
(Circle all
that apply.)
White/Caucasian
Latino/Hispanic
Black/African
American
Asian/Pacific
Islander
American
Indian/Alaska
Native
Other:
_______________
Mixed Race
White/Caucasian
Latino/Hispanic
Black/African
American
Asian/Pacific
Islander
American
Indian/Alaska
Native
Other:
_______________
Mixed Race
White/Caucasian
Latino/Hispanic
Black/African
American
Asian/Pacific
Islander
American
Indian/Alaska
Native
Other:
_______________
Mixed Race
White/Caucasian
Latino/Hispanic
Black/African
American
Asian/Pacific
Islander
American
Indian/Alaska
Native
Other:
_______________
Mixed Race
What is this
person’s
gender?
(Circle one.)
Male
Female
Male
Female
Male
Female
Male
Female
38. THE FOLLOWING QUESTIONS ARE ABOUT EXTRA HOUSING UNIT #3. IF YOU
DON’T HAVE A THIRD EXTRA HOUSING UNIT, PLEASE SKIP TO QUESTION #60.
OTHERWISE, PROCEED.
When was the decision made to install Extra Housing Unit #3 on your property? (Circle
one.)
a) It happened while my household owned the property
b) It happened before my household owned the property
If you answered “It happened before my household owned the property” to
question 38, then skip to question 40. Otherwise, proceed to question 39.
39. Who did the actual construction work of installing Extra Housing Unit #3 on your
property? (Circle all that apply.)
a) I did it.
b) A hired contractor did the work.
c) A friend or relative did the work.
d) Other __________________________________________.
40. How would you best describe the physical layout of Extra Housing Unit #3? (Circle
one.)
a) Part or all of basement or first floor converted to an apartment
b) Garage converted to an apartment
c) Apartment above a garage
d) Attic converted to an apartment
e) Rooms inside main part of house converted to an apartment
f) Apartment behind main house and attached to it
g) Apartment behind main house and in its own separate structure
h) Other: Please describe _________________________
41. Does Extra Housing Unit #3 have its own complete kitchen (sink, range, and
refrigerator)?
Yes No (circle one)
42. Does Extra Housing Unit #3 have at least one complete bathroom (toilet, sink, and
shower/bath)?
Yes No (circle one)
43. How many bedrooms does Extra Housing Unit #3 have? (Circle one.)
a) None (it is a studio or efficiency unit)
b) 1
c) 2
d) 3 or more
44. About how many years ago was Extra Housing Unit #3 installed? (If less than one
year ago, just mark “0.” If you don’t know, check “I don’t know.”)
________ years ago _____ I don’t know
45. Is Extra Housing Unit #3 occupied by at least one person?
Yes No (circle one)
If you answered Yes to question 45, then skip ahead to question 47. Otherwise,
proceed to question 46.
46. Why is Extra Housing Unit #3 currently unoccupied? (Circle one.)
a) It needs physical work to be rentable.
b) It is vacant, but I am looking for a tenant.
c) It is being used as something other than an apartment (home office,
workshop, studio, etc).
d) Other reason: __________________________________________
Please skip ahead to question 60.
47. Which of the following best describes the relationship of the occupant(s) in Extra
Housing Unit #3 to you at the time of move-in? (Circle one.)
a) My household lives in Extra Housing Unit #1.
b) Relative(s)
c) Friend(s)
d) Acquaintance(s)
e) I did not know the occupant(s) before move-in.
If your answer to question 47 was a), then skip to question 60. Otherwise, please
proceed to question 48.
48. How did you find the occupant(s) in Extra Housing Unit #3? (Circle all that apply.)
a) I already knew the occupant(s)
b) Craigslist
c) Other Internet source
d) Newspaper classified ad
e) Referred by someone I know
f) Other: ____________________________
49. How much is the rent paid to your household by the occupant(s) in Extra Housing
Unit #3? If the occupant(s) are staying in Extra Housing Unit #3 for free, then mark “$0.”
$__________ per month
50. Please circle all of the following utilities that are included in the rent, if any, paid to
your household by the occupant(s) in Extra Housing Unit #3.
Water/Sewer Gas/Electricity Trash/Recycling Pickup
Telephone Cable TV Internet
51. How long has the person who has been living in Extra Housing Unit #3 for the
longest time lived there?
_________ years _____ less than one year (check if applicable)
_________ not sure
52. How many total cars do the people (if any) who are living on your property in Extra
Housing Unit #3 (if any) normally park on your property (i.e. not on the street)?
(Please choose one.)
a) none
b) 1
c) 2
d) 3 or more
53. How many total cars do the people (if any) who are living on your property in Extra
Housing Unit #3 (if any) normally park on the street within a five-minute walk of where
you live? (Circle one.)
a) none
b) 1
c) 2
d) 3 or more
54. How many children (i.e. people under age 18) live in Extra Housing Unit #3?
a) none
b) 1
c) 2
d) 3 or more
55. For each adult (i.e. person of age 18 or more) living in Extra Housing Unit #3, please
answer the following questions.
Question Adult #1 Adult #2 Adult #3 Adult #4
What is
your best
estimate of
this
person’s
age?
_____ years
_____ years
_____ years
_____ years
How would
you best
describe the
ethnicity or
race of this
person?
(Circle all
that apply.)
White/Caucasian
Latino/Hispanic
Black/African
American
Asian/Pacific
Islander
American
White/Caucasian
Latino/Hispanic
Black/African
American
Asian/Pacific
Islander
American
White/Caucasian
Latino/Hispanic
Black/African
American
Asian/Pacific
Islander
American
White/Caucasian
Latino/Hispanic
Black/African
American
Asian/Pacific
Islander
American
Indian/Alaska
Native
Other:
_______________
Mixed Race
Indian/Alaska
Native
Other:
_______________
Mixed Race
Indian/Alaska
Native
Other:
_______________
Mixed Race
Indian/Alaska
Native
Other:
_______________
Mixed Race
What is this
person’s
gender?
(Circle one.)
Male
Female
Male
Female
Male
Female
Male
Female
Please skip ahead to question 60.
Extra Housing Unit questions
56. Which of the following best describes why you do not have any Extra Housing Units
on your property?
a) I have never given any thought to installing Extra Housing Units on my
property.
b) I don’t want any Extra Housing Units on my property.
c) I might like to have one or more Extra Housing Units on my property, but I
haven’t gotten around to installing any.
d) I looked into installing one or more Extra Housing Units on my property, but
it didn’t work out.
e) I am planning on installing one or more Extra Housing Units on my
property.
57. If your answer to the previous question was d (“I looked into installing one or more
Extra Housing Units on my property, but it didn’t work out,”) please answer this
question: what was the most important reason that you did not end up installing one or
more Extra Housing Units on your property?
a) It would have been too expensive.
b) I couldn’t get financing.
c) One or more of my neighbors would have been unhappy with it.
d) My property couldn’t fit the amount of off-street parking required by the city.
e) Other city requirement: ________________________
f) Other reason: _______________________________
58. Is there at least one residential property on your street that has one or more Extra
Housing Units on it?
a) Yes
b) No
c) I’m not sure
59. If you answered a (“Yes”) to the previous question, do you think that the Extra
Housing Units in one or more residential properties on your street have a negative
impact on the neighborhood? (Please circle the best choice.)
a) Yes – too many cars trying to park on my street.
b) Yes – too much noise.
c) Yes – unruly tenants.
d) Yes – other impact: _________________________________________
e) No – there is no negative impact.
Transportation questions
60. How many cars (if any) do you and the other people living in your unit, and NOT
including any people living in any other units on your property (if any), normally park on
your property (i.e. not on the street)?
a) none
b) 1
c) 2
d) 3 or more
61. How many cars (if any) do you and the people living in your unit, and NOT
including any other people living in your property's other units (if any), normally park on
the street within a five-minute walk of where you live? (Please choose one.)
a) none
b) 1
c) 2
d) 3 or more
62. How many currently usable parking spaces are on your property, including garage
and driveway spaces?
a) none
b) 1
c) 2
d) 3
e) 4
f) 5
g) more than 5
Questions about you and your household
63. How many children (people under the age of 18) live on your property in your
household (not including children who are living on your property but are members of
other households)?
a) 0
b) 1
c) 2
d) 3 or more
64. For each adult (over the age of 18) living in your household (but NOT including
people living in Extra Housing Units, if any), please answer the following questions.
Question You (person
filling out this
survey)
Adult #2 Adult #3 Adult #4
What is your
best estimate
of this
person’s
age?
_____ years
_____ years
_____ years
_____ years
How would
you best
describe this
person’s race
or ethnicity?
(Circle all
that apply.)
White/Caucasian
Latino/Hispanic
Black/African
American
Asian/Pacific
Islander
American
Indian/Alaska
Native
Other:
_______________
Mixed Race
White/Caucasian
Latino/Hispanic
Black/African
American
Asian/Pacific
Islander
American
Indian/Alaska
Native
Other:
_______________
Mixed Race
White/Caucasian
Latino/Hispanic
Black/African
American
Asian/Pacific
Islander
American
Indian/Alaska
Native
Other:
_______________
Mixed Race
White/Caucasian
Latino/Hispanic
Black/African
American
Asian/Pacific
Islander
American
Indian/Alaska
Native
Other:
_______________
Mixed Race
What is this
person’s
gender?
(Circle one.)
Male
Female
Male
Female
Male
Female
Male
Female
Which of the
following best
describes
this person’s
educational
attainment?
(Circle one.)
Didn’t finish high
school
High school grad
Some college
College grad
Graduate degree
Didn’t finish high
school
High school grad
Some college
College grad
Graduate degree
Didn’t finish high
school
High school grad
Some college
College grad
Graduate degree
Didn’t finish high
school
High school grad
Some college
College grad
Graduate degree
65. What was your household’s before-tax income in the last 12 months? Please include
all income, including salaries, wages, investments, government benefits, etc. Please do
not include people living in units on your property (if any) other than the one you live in
as members of your household for the purposes of this question. (Please choose one.)
a) Less than $10,000
b) $10,000 to $14,999
c) $15,000 to 24,999
d) $25,000 to $34,999
e) $35,000 to $49,999
f) $50,000 to $74,999
g) $75,000 to $99,999
h) $100,000 to $149,999
i) $150,000 to $199,999
j) $200,000 or more
Final questions
In the next two questions, we will ask you if you would be willing to be
interviewed, and if you would like to be part of a raffle. If you do not want to be
either interviewed at a later time or included in the raffle, we will replace your
name in our records with a number so that your anonymity will be completely
protected. If you agree to be interviewed, we will not publish your name. If you
elect to be part of our raffle prize drawing but do not agree to be interviewed, we
will replace your name with an identifying number once the prizes have been
mailed out. We are committed to protecting your anonymity.
66. While we expect that we will learn a lot from the results of this survey, there is also a
lot that we can only learn by doing in-person interviews. Would you be willing to have us
contact you to schedule an interview at a later date? As is the case with this survey,
strict anonymity would be maintained.
Yes No (Circle one)
67. As a thank-you to people who have taken the time to fill out this survey, we are
conducting a raffle drawing. The prizes will be 1) a $200 Apple Store certificate; 2) a
$150 BART card; and 3) a $100 BART card. Would you like to take part in the raffle
drawing?
Yes No (Circle one)
68. While we tried to be comprehensive in selecting the questions to include in this
survey, there surely are issues that we have not considered. If there is anything else
about your experience with housing on your property or in your neighborhood that you
would like to tell us about, please write it below.
___________________________________________________________________
___________________________________________________________________
___________________________________________________________________
___________________________________________________________________
___________________________________________________________________
THANK YOU FOR TAKING THE TIME TO FILL OUT THIS SURVEY. PLEASE PUT
THE COMPLETED SURVEY INTO THE SELF-ADDRESSED, PRE-STAMPED
ENVELOPE, SEAL IT, AND PUT IT IN THE MAIL TO RETURN IT TO THE
RESEARCHERS. WE APPRECIATE YOUR TIME.
Appendix
2:
Homeowner
respondent
solicitation
postcard
(English-‐
language
version)
Appendix
3:
Summary
of
variables
and
imputation
methods
used
in
rental
market
analysis
of
online
apartment
advertisements
This
section
gives
a
fuller
description
of
how
each
variable
harvested
from
Craigslist
advertisements
for
rental
apartments
and
used
in
the
hedonic
model
and
for
descriptive
statistics
was
computed.
In
cases
where
variables
were
partially
populated
via
imputation,
the
methodology
by
which
this
was
done
is
described.
adjusted_rent
is
the
dependent
variable
in
the
hedonic
analysis.
The
baseline
rent
was
ascertained
from
the
advertisement
(all
advertisements
that
failed
to
report
a
monthly
rent
were
discarded),
and
then
adjusted
according
to
whether
or
not
off-‐street
parking
was
offered,
and
whether
or
not
all
of
part
of
gas/electric,
water/sewer,
and
cable/Internet
utility
costs
were
included
in
the
rent.
Where
an
off-‐street
parking
space
was
offered
for
a
given
monthly
cost,
this
cost
was
added
to
the
rent.
Where
an
off-‐street
parking
space
was
offered
for
an
unspecified
price,
a
price
randomly
selected
from
the
units
with
specified
parking
costs
was
used.
Where
an
off-‐street
parking
space
was
offered
for
free,
the
rent
was
decreased
by
a
price
randomly
selected
from
the
set
of
units
with
specified
parking
costs.
(In
cases
where
no
off-‐street
parking
was
offered
at
all,
and
where
any
tenant
parking
therefore
occurred
on
the
public
street,
adjusted_rent
remained
unchanged
but
the
ON_STREET
dummy,
described
below,
was
flagged.)
Upward
or
downward
adjustments
were
done
on
the
basis
of
imputation
in
76
records
(out
of
a
total
of
338),
while
upward
adjustments
on
the
basis
of
directly
reported
off-‐street
parking
charges
were
done
in
33
cases.
In
cases
where
gas/electric
and
water/sewer
charges
were
fully
included
with
rent,
monthly
rent
was
adjusted
downward
by
an
amount
selected
from
the
table
of
utility
allowances
(as
of
12/1/2010)
published
by
the
Oakland
Housing
Authority
for
use
in
subsidized
housing
rent
calculations
(electric
heating
and
cooking
were
assumed,
although
the
amounts
were
little
different
than
had
gas
usage
been
assumed).
These
allowances
vary
with
the
number
of
persons
assumed
to
be
in
the
household,
which
in
turn
vary
with
the
number
of
bedrooms
reported
for
the
unit
(1
person
is
assumed
to
occupy
a
studio,
and
1.5
people
are
assumed
to
occupy
each
bedroom
in
a
unit
that
has
one
or
more
bedrooms).
In
cases
where
one
or
both
of
these
charges
were
shared
with
the
landlord,
one
or
both
of
the
amounts
of
the
downward
revision(s)
were
halved.
Cable
and
Internet
packages
were
assumed
to
be
worth
$60
per
month
based
on
a
perusal
of
offerings
by
local
companies
serving
the
East
Bay.
In
cases
where
only
electric
but
not
gas,
or
only
cable
but
not
Internet,
or
vice
versa,
were
offered,
the
adjustment
was
also
cut
in
half.
num_BRs
and
num_BAs
were
straightforward:
the
number
of
bedrooms
and
bathrooms
indicated
in
the
ad
were
recorded
in
the
database.
Studio
units
were
recorded
as
having
zero
bedrooms.
Every
advertisement
that
failed
to
indicate
a
number
of
bedrooms
and
bathrooms
for
the
unit
was
excluded
from
the
database.
sq_footage
was
recorded
from
the
advertisement
in
cases
where
it
was
specified.
Since
square
footage
was
specified
in
only
80
out
of
the
338
records
that
were
recorded,
it
was
necessary
to
impute
square
footage
for
the
majority
of
units.
This
was
done
by
randomly
selecting
a
square
footage
amount
from
the
set
of
units
with
the
same
bedroom/bathroom
configuration
with
known
square
footages.
In
cases
where
there
were
not
enough
units
for
this
to
be
possible
(as
was
the
case
generally
with
units
with
larger
numbers
of
bedrooms),
a
smaller
unit
configuration
was
randomly
selected
and
then
square
footages
were
adjusted
upward
by
an
assumed
100
sf
per
bedroom
and
50
sf
per
bathroom
(or
25
sf
per
half
bathroom).
SECONDUNIT
was
recorded
as
a
dummy
variable
to
indicate
whether
or
not
a
unit
is
a
secondary
unit.
Note
that
this
variable
was
not
actually
included
in
the
hedonic
model
runs,
but
rather
was
used
to
split
the
data
set
into
portions
of
roughly
equivalent
size
exclusively
composed
of
either
secondary
or
non-‐secondary
units.
SECURE
is
a
dummy
variable
that
indicates
whether
or
not
a
unit
has
a
secure
entry
(i.e.
an
entry
that
lies
behind
a
locked
gate
or
door
with
access
limited
to
those
residing
on
the
property).
ON_STREET
is
a
dummy
variable
that
takes
on
the
value
of
0
if
off-‐street
parking,
whether
free
or
paid,
is
provided
on
the
property
(in
which
case
adjusted_rent
is
adjusted
either
upward
or
downward,
as
described
above),
and
1
if
the
only
parking
to
be
found
is
on-‐street
(in
which
case
there
is
no
adjustment
to
adjusted_rent).
Free
off-‐street
parking,
paid
off-‐street
parking,
and
on-‐
street
parking
are
considered
to
be
mutually
exclusive
categories
for
each
unit.
OWNONSITE
is
a
dummy
variable
that
indicates
whether
or
not
the
owner
of
the
rental
unit
resides
on
the
same
property
in
another
unit.
It
was
only
set
in
the
affirmative
if
the
text
of
the
ad
made
it
clear
that
the
owner
was
present.
Because
absentee
landlordism
is
often
seen
as
a
negative
by
renters,
it
was
assumed
that
this
characteristic
was
not
in
place
unless
it
was
specifically
mentioned
(i.e.,
since
landlords
would
have
an
incentive
to
mention
the
presence
of
a
landlord
in
an
ad
if
on-‐
site
landlord
presence
were
indeed
seen
as
a
good
thing).
COINOP_LAUND
is
a
dummy
variable
that
was
set
to
1
if
the
ad
mentioned
the
presence
of
coin-‐
operated
laundry
machines
(washer
and
dryer)
on
the
same
property
as
the
unit.
FREE_LAUND
is
a
(mostly)
dummy
variable
that
was
set
to
1
if
the
ad
mentioned
the
presence
of
laundry
machines
(washer
and
dryer)
on
the
same
property
as
the
unit
(whether
in
the
unit
or
elsewhere
on
the
premises)
that
the
tenant
could
use
for
free.
In
one
instance,
the
ad
mentioned
that
laundry
machines
could
be
used
only
on
a
Friday
or
Saturday;
in
this
case,
this
variable
was
assigned
a
value
of
0.5.
MICROW
is
a
dummy
variable
indicating
whether
or
not
the
unit
was
equipped
with
a
microwave
oven.
A
microwave
oven
was
assumed
to
not
be
provided
unless
affirmatively
specified.
DISHWASHER
is
a
dummy
variable
indicating
whether
or
not
the
unit
included
a
dishwasher
(assumed
to
not
be
provided
unless
specifically
mentioned).
FULL_KITCHEN
is
a
dummy
variable
that
was
set
to
1
if
the
unit
had
up-‐to-‐standard
kitchen
appliances,
including
a
four-‐burner
stove
(whether
gas
or
electric)
and
a
full-‐sized
oven.
An
up-‐to-‐
standard
kitchen
was
assumed
unless
the
advertisement
mentioned
otherwise
(due
to
our
presumptions
about
tenants’
expectations
of
rental
properties).
If
an
ad
made
any
mention
of
elements
of
a
substandard
kitchen
–
for
example,
the
presence
of
a
hotplate,
two-‐burner
stove,
convection
oven,
or
half-‐sized
refrigerator
–
then
a
value
of
0
was
assigned
to
this
variable.
Walkscore
is
a
percentile
index
ranging
from
0
to
100
indicating
the
“walkability”
of
any
address
in
the
United
States.
It
is
measured
via
a
free
web
tool
at
www.walkscore.com.
Walkability
is
construed
to
correspond
to
the
number
and
variety
of
businesses
and
convenience
services
that
can
be
reached
within
a
short
walk
of
a
given
location.
A
value
of
0
means
that
a
location
is
equivalent
in
walkability
to
the
least
walkable
locations
in
the
United
States
and
a
value
of
100
means
that
it
is
equivalent
to
the
most
walkable
locations.
crime_index
is
a
percentile
index
ranging
from
0
to
100
indicating
the
blended
level
of
violent
and
property
crime
rates
(from
publicly
available
data)
for
a
given
address
in
the
United
States.
It
can
be
determined
via
the
use
of
a
proprietary
web-‐based
tool
at
www.neighborhoodscout.com.
An
index
of
0
indicates
that
a
location
is
as
dangerous
as
the
least
safe
neighborhoods
in
the
United
States,
whereas
an
index
of
100
means
that
the
location
is
as
safe
as
the
safest
locations
in
the
nation.
freeway
is
an
index
of
proximity
to
a
freeway.
It
takes
on
a
value
of
0
if
a
location
is
at
least
1,000’
from
the
nearest
freeway,
1
if
it
is
between
500’
and
1,000’
from
the
nearest
freeway,
and
2
if
it
is
within
500’
of
the
nearest
freeway.