M C P D: Memorandum
M C P D: Memorandum
Item 10
MEMORANDUM
In February, the Planning Board approved an approach for the Housing Policy Element of
the General Plan whereby the Board engages in regular discussions of housing policy issues
throughout the spring. These discussions occur in at least two forums: public work-sessions
where the conversation is sparked by a staff memorandum and discussions that begin with
presentations by invited speakers. Through these discussions, we will identify housing policy
issues that we may decide to address through amending the Housing Policy Element of the
General Plan or through other means.
INTRODUCTION
Although suburban communities have existed in the United States for more than a
century, the growth of America’s suburbs increased significantly after World War II. The
1
Washington metropolitan area was no exception to this trend. From 1950 to 2000, the region’s
population increased approximately 150 percent. Montgomery County’s population trended
upward as well, showing a 431 percent increase during the same time period. According to
Montgomery County housing inventory data, the number of housing units in the County has
increased from 47,199 units to 334,632 units between 1950 and 2000, representing an increase of
approximately 609 percent.
400,000
350,000
300,000
No. of Housing Units
250,000
200,000
150,000
100,000
50,000
0
1950 1960 1970 1980 1990 2000
Source: U.S. Bureau of the Census; Montgomery County Department of Park and Planning, Research
and Technology Center, revised June 2001
While the County’s and region’s population and housing supply continued to grow at a
rapid pace through the last several decades, some suburban communities began to show signs of
decline and are susceptible to adversities that have plagued inner-city neighborhoods since the
late 1960s and on, including, disinvestment, increased poverty rates, infrastructure deterioration,
and an increase in crime.
This report sets out to examine indicators of change that could be used to measure change
and stability of Montgomery County’s communities. First, a brief overview of literature
addressing neighborhood decline1 in suburban communities in the U.S. will be presented.
Second, the report includes a description of the research methodology used and indicator’s
identified to analyze patterns of neighborhood change. Finally, findings on the regional, County
and neighborhood level are presented. For the purposes of this study, regional analysis includes
1
While neighborhood decline is just one side of the neighborhood change issue, it has been the subject of
considerable scholarly attention in recent years.
2
an examination of County-level for Montgomery County, Prince George’s County, Fairfax
County, Virginia, Arlington County, Virginia and the City of Alexandria, Virginia. A more
detailed analysis will follow, comparing neighborhood change County-wide against the sub-
jurisdiction level. To compare municipalities within the region, we are restricted to Census data.
For Montgomery County, however we have more detailed and recent information including the
Census Update Survey (specifically 1997 and 2005) and the recently compiled departmental
housing inventory.
The report will conclude with a discussion of implications and possible next steps,
including the establishment of regular periodic review of change and stability in Montgomery
County’s communities. A data compendium (Appendix I) will be provided in order to provide
additional information regarding this very important issue facing the County and the region.
A literature scan performed by Staff revealed that there are many definitions of
neighborhood decline. In the broadest context possible, decline is defined by adverse conditions
being present or increasing in a particular area. Such conditions include declining population,
poverty increase, eroding employment opportunities and fiscal strain. Social conditions that
indicate neighborhood decline include increase in crime rate, quality of schools, significant
amount of female-headed households and community transience.2 Neighborhood decline can
also be linked to the availability or lack of capital investment in the built environment, which is
shown in the form of new construction and renovation of housing and public infrastructure.3
Due to the downturn in the housing market and the uptick in home foreclosures throughout the
region and the Country, home vacancy rates are also a sign of neighborhood decline. This factor
was recently explored in the March 2008 edition of the Atlantic Monthly article titled The Next
Slum, by Christopher Leinberger4. Mr. Leinberger discussed the devastating impact of the
foreclosure situation is having on suburban communities.
For the purposes of this report, neighborhood decline is primarily defined by population
decrease, a decrease median family income, increase of poverty, and declining property values.
2
Carter, T. and C. Polevychok. (2006). Understanding Disinvestment and Decline.
http://ius.uwinnipeg.ca/CRC/Understanding%20Disinvestment%20and%20Decline.pdf
3
Smith, N., P. Caris and E. Wyly. (2001). The “Camden Syndrome” and the Menace of Suburban Decline –
Residential Disinvestment and Its Discontents in Camden County, New Jersey. Urban Affairs Review, 36 (497-531).
4
Mr. Leinberger will be speaking later tonight (May 29th, 2008) as part of our Excellence in Planning Speaker
Series.
3
Similarly, neighborhood change is defined by changes in population, income, poverty, and
property values.
Over the past 40 years, there has been a great deal of examination regarding
neighborhood decline in America’s central cities or ‘inner-city’ communities. From the 1950s to
the late 1990s, many neighborhoods in urban America began to suffer in terms of adequate
housing and related infrastructure, stalled economic development and a cohesive sense of
community.
During this time period, suburban communities grew across the Country at an
exponential rate. Due to several factors, including the building of interstate highways, mortgage
financing subsidies, land-use controls via subdivision regulation and deindustrialization of center
cities5,
Considered the ‘second urban crisis’6, suburban decline has quickly become a major issue
in metropolitan areas. According to the 2000 article Suburban Decline: The Next Urban Crisis7:
From 1960 to 1990, among 554 established U.S. suburbs located in the 24 of the
County’s highest populated metropolitan areas, 405 of the suburbs declined in median
family income;
20 percent of the suburbs declined in relative income at a faster rate than the metropolitan
area’s central cities during the same time period;
From 1980 to 1990, approximately 33 percent of the suburbs declined in relative income
at a faster rate than the metropolitan area’s central cities – signaling that the rate of
decline quickened during later decades.
4
suburbs, particularly in first-ring suburbs have high poverty rates. By 2000, approximately nine
percent of first-ring suburbs showed having at least 20 percent of its population living below the
poverty line.
Taking the relationship of disinvestment and neighborhood decline into account, the
importance of community reinvestment will be further explored in the policy implications
section of this report.
9
Boise City Disinvestment Monitoring Report
http://www.cityofboise.org/Departments/PDS/PDF/CompPlanning/Disinvestment%20Study/2_Disinvestment_Stud
y_Intro.pdf
10
Oriefield, M. (1997). A Need for Coalition. http://bostonreview.net/BR22.1/orfield.html
11
Davis, M. (2001). Ozzie and Harriet in Hell.
http://www.gsd.harvard.edu/research/publications/hdm/back/1davis.pdf
5
METHODOLOGY
1. Identifying Indicators: Staff determined12 that the use of the following indicators would
be crucial to determining change on the metropolitan and local levels:
Income Change – Median household income figures are used to determine if incomes
have declined in specific areas of the County;
Poverty Increase or Decrease – Poverty rate trends are used to determine if poverty is
on the rise in the County’s neighborhoods;
Housing Property Values – Steady or rising property values are one determinate of
neighborhood stability;
Age of Housing Stock – The percentage of housing stock built during a certain time
period allows for determination if a neighborhood can be determined ‘older’ or
‘newer.’ Housing stock age has also been shown to be a factor in neighborhood
decline. Older housing stock is more expensive to maintain, thought it may initially
be a cost-effective purchase in some cases. Deterioration of housing stock due to
prohibitive maintenance costs impacts neighborhood desirability
12
Hanlon, B. (2007) The Decline of Older Suburbs: A New Reality in the U.S., (Unpublished Dissertation). With
respect to establishing indicators, according to Hanlon, when defining suburban decline, three elements ought to be
taken into account: (1) It should identify primary indicators that measure or describe decline in the suburbs; (2)
These primary indicators of decline should be relative in nature and therefore should provide a standard against
which to compare declining from non-declining suburbs”; and (3) “Measures of decline should, in theory be linked
to the effective functioning of suburbs. In other words, how well the suburb is performing socially and
economically should be clearly understood by the measures used.” (p.40)
6
County Household Income;
2. Determination of what Counties and places would be examined to analyze change on the
metropolitan level will involve comparison of change in Montgomery County progress
against change in Prince George’s County, Arlington County, Virginia, Fairfax County,
Virginia and the City of Alexandria, Virginia. Local level data will be examined using
indicator findings of Montgomery County’s Planning Areas.
3. Comparison of indicator findings: Comparisons are made to show trends using the
indicators. Trend data will be shown in chart/graph form and GIS.
Data Sources
The primary data sources used for the portions of this report that focus on metropolitan-
level analysis include census data collected from the U.S. Department of Housing and Urban
Development (HUD) The State of the Cities Data System (SOCDS). As noted before, the
timeline that will be measured is 1980 to 2000. For Montgomery County local-level data,
additional sources include the Housing Inventory data for Montgomery County and the Census
Update Survey for the years 1997 and 2005.
Finding: Population has increased for each study jurisdiction in the region.
Between 1980 and 2000, the population in the Washington DC metro area increased by
approximately 1.4 million from 3,397,935 to 4,796,183. Population in all of the area’s suburban
areas including the Counties presented had significant increases in population over the same time
period. County comparisons of population show that Fairfax County, Virginia experienced the
largest increase in population in absolute terms and relative to the other Counties. Fairfax
County’s population increased in population from 596,901 to 969,749, representing an increase
of approximately 63 percent. Montgomery County followed, increasing its population from
579,053 to 873,341 (51 percent increase).
7
Figure 2: Population Change in the Washington Metropolitan Region, 1980 to 2000
5,000,000
4,000,000
Population
3,000,000
2,000,000
1,000,000
0
1980 Washington, Suburbs* Montgomery Prince Arlington Fairfax Alexandria,
DC Metro County, MD George's County, VA County, VA VA
1990 County, MD
2000
Source: M-NCPPC Research and Technology Center; State of the Cities Data Systems
Finding: Median household incomes (in constant dollars) have increased in the region.
Figure 3: Median Household Income, Washington DC Metro – 1980 to 2000 (2005 Dollars)
$100,000
$90,000
$80,000
$70,000
$60,000
$50,000
1980 1990 2000
8
Finding: Overall, poverty rates in the Washington, DC metro decreased from 1980 to 2000.
Poverty rates in Washington’s suburban communities, including Montgomery County, trended
upwards during the same time period.
Between 1980 and 2000, the poverty rate in the Washington, DC metropolitan area fell
from 8.3 percent to 7.1 percent. Poverty rates in area suburbs increased from 5.8 to 6.4 percent.
Notable increases in poverty rates were seen in Montgomery County (4.3 percent to 5.4 percent)
and Prince George’s County (6.7 percent to 7.7 percent).
10
9
8
7
6
Percent
5
4
3
2
1
0
1980 1990 2000
Finding: From 1980 to 2000, median housing values increased by approximately 12 percent in
the Washington, DC metropolitan area and its suburbs.13
Housing values have increased at a steady pace in the region since 1980. Overall, the
median home value in the Washington region increased from $182,914 in 1980 to $205,493 in
2000. Suburban Washington’s median home value increased from $179,640 to $201,428. The
median home value in Montgomery County jumped 64 percent from 1980 to 2000, representing
the largest increase in median home value of all Counties mentioned.
13
2005 Dollars
9
Figure 5: Median Housing Values, Washington Metropolitan Area, 1980 to 2000
(2005 Dollars)
$350,000
$300,000
$250,000
$200,000
$150,000
$100,000
$50,000
$0
1980 Washington, Suburbs Montgomery Prince Arlington Fairfax Alexandria,
1990 DC Metro County, MD George's County, VA County, VA VA
2000 County, MD
Finding: A significant amount of the region’s housing stock has been built before 1970 – the
Washington metropolitan area is a relatively even mix of mature and emerging communities.
Table 1: Housing Stock in the Study Jurisdictions (Percent) The majority of the
housing stock in Montgomery, Prince George’s and Fairfax Counties was built after 1970
(Montgomery County – 56 percent, Prince George’s County – 52 percent; Fairfax County – 68
percent). In both Arlington County and the City of Alexandria significant portions of the
housing stock were built before 1970 (75 percent and 59 percent, respectively).
Prince George’s
Montgomery County, MD Arlington County, Fairfax County, Alexandria, VA
County, MD VA VA
2.0 1.9 0.7 2.1 4.2
Year Built
5.6 7.4 3.6 7.4 6.6
1999 -2000
7.2 10.4 4.4 8.8 3.3
1995-1998
24.7 16.1 10.3 27.2 8.7
1990-1994
16.5 16.1 6.4 22.7 18.3
1980-1989
16.5 22.1 9.1 15.9 10.8
1970-1979
22.0 21.4 47.8 14.7 30.9
1960-1969
5.5 4.7 17.7 1.2 17.2
1940-1959
2.0 1.9 0.7 2.1 4.2
1939 and earlier
Source: US Census
10
ANALYSIS RESULTS: CHANGE IN MONTGOMERY COUNTY COMMUNITIES
Initially, staff applied Dr. Hanlon’s research methodology to Montgomery County using
Census data. However, the 2000 Census data are the most recent detailed Census information
and marked changes in the housing market and economy have occurred since 2000. The County
has much more recent data from other sources which will be discussed later in the report.
One very useful element of Dr. Hanlon’s research is comparing mature and emerging
suburbs. Using 2000 Census data, fifty-one communities (Census Designated Places) were
identified for measurement according to U.S. Census designated place boundaries. After further
research it was determined that 30 communities were mature suburban communities and the
balance being emerging suburban communities (using Dr. Hanlon’s criteria). Table 2 provides a
list of the communities in their respective age categories.
11
Rockville
Silver Spring
Somerset
South Kensington
Takoma Park
Washington Grove
Wheaton/Glenmont
White Oak
After the 2010 Census Data, a more relevant analysis of these Census Designated Places
will be possible, using Dr. Hanlon’s methodology.
12
The CUS data is a rich source of information about the residents of Montgomery County: how
their households are composed, their education levels, where they work and how they get there,
what languages they speak, how much they pay for housing, etc. The following analysis is only
an example of what can be analyzed using five housing related fields. The analysis conducted by
staff yielded relative differences among the planning areas and shows trends in the indicators of
population, median household income, average monthly household costs and number of rental
units. Table 3 provides a glance the 2005 results. Mapping these data will allow staff to easily
analyze changes that may be occurring in different sections of the county.
13
Table 4 provides the percent change between 1997 and 2005 using these indicators.
1997 and 2005 % Change
Average Monthly
Housing Costs
Household Household
Population Income** Owners Renters
Aspen Hill 7.8% ‐11.0% 11.6% 12.9%
Bethesda/Chevy Chase 8.9% 0.8% 8.8% 26.2%
Clarksburg & Vicinity 215.5% 26.4% 21.8% *
Cloverly/Norwood 18.0% 17.0% 7.9% ‐8.6%
Colesville/White Oak 4.8% ‐3.3% 11.0% 18.2%
Damascus & Vicinity 19.0% 10.5% 10.8% 1.1%
Darnestown 14.4% 2.4% 2.5% 14.3%
Fairland 15.1% ‐10.1% 6.5% 8.2%
Gaithersburg & Vicinity 11.2% ‐4.4% 4.7% 17.5%
Germantown 37.5% 2.9% 9.2% 7.4%
Kemp Mill/Four Corners 0.9% 9.8% 9.8% 6.2%
Kensington/Wheaton 4.1% 6.0% 14.6% 14.7%
North Bethesda/Garrett Park 7.5% 4.0% 17.1% 26.0%
Olney & Vicinity 16.0% 2.0% 11.2% *
Poolesville & Vicinity 10.6% 9.2% 5.9% 17.3%
Potomac/Cabin John 8.1% 5.9% ‐0.2% ‐3.7%
Rockville 23.5% 5.2% 11.6% 13.1%
Silver Spring 12.7% 1.5% 1.2% 25.6%
Takoma Park ‐9.9% ‐15.1% 28.2% ‐1.7%
Travilah 35.4% 9.2% 11.9% ‐14.7%
Upper Rock Creek 37.5% ‐13.6% 6.1% 20.7%
County 13.1% 2.4% 9.4% 17.1%
* INSUFFICIENT DATA FOR CALCULATION
** REPORTED IN 2004 CONSTANT DOLLARS
The data from the CUS can be combined with the County’s new housing inventory
(which contains data by individual addresses) to provide an even richer analysis. This will allow
us to not only look at trends over time but combine the following CUS field with housing data
14
such as assessed housing values, housing sales, foreclosure information, MPDU’s, as well as
social data (crime, health, educational, etc.).
Using only the CUS data and these 5 characteristics, staff found the following:
The greatest rate of population increase occurred in Clarksburg (215% but only an actual
increase from 2,060 to 6,500. This increase is to be expected with a significant amount of the
county’s growth occurring there), followed by Germantown, Upper Rock Creek and Travilah.
Takoma Park was the only planning area that experienced a decrease in population between 1997
and 2005.
100,000
80,000
60,000
40,000
20,000
0
Da ville rwo ty
Fa sto Vic k
G e ersb
Ta p r
Tr a P
As
Be Hi
Cl sda
Cl b u e v y
Da scu hit
Ga nd
Ke ant &
Ke Mi
No gt ur
Ol Be Wh rner
Po & sda on
Ro ac Vi
Si
Up ah
lv le
ar
ov rg
av
ne t h e a s
irl w in
ko in
pe
m ow Vic
ns ll/
to le
th ll
le No in
ol Vic /G
m / W od
ck /Ca cin
rn s & e O
r m ur
ith
pe
er
ks /Ch
a
e & Ch
il ark
s
e
p
a
m &
e
y
in Fo
m g
vi
e
n
r R
S
1997 l
oc
k
n
i
g
Cr
bin
ee
Jo
k
2005
i
hn
i
a
ty
i
a
in
ty
i ty
tt
P
ar
k
Finding: On the County level, median household income increased 2.4 percent. Increases in
median household income were evident in most planning areas.
Median household income increased in 14 of 21 planning areas between 1997 and 2005.
The largest increase was seen in Clarksburg & Vicinity and the smallest increase in
Bethesda/Chevy Chase. Six planning areas experienced a decrease in median household income
during the same time period. Takoma Park experienced the most significant decrease in income.
Notable decreases in income also occurred in Upper Rock Creek and Aspen Hill. As an example
15
of the fact that this must be construed as an indicator of change rather than decline, Takoma Park
experienced the greatest percentage increase in monthly housing costs for homeowners over this
period of time (28.2%).
$140,000
$120,000
$100,000
$80,000
$60,000
$40,000
$20,000
$0
Cloverly/Norwood
Germantown
Kensington/Wheaton
Travilah
Damascus & Vicinity
Olney & Vicinity
Silver Spring
Montgomery County
Aspen Hill
Fairland
Potomac/Cabin John
1997
2005
By linking changes in population and household income, we can identify planning areas
that may warrant further examination and research. For example, Takoma Park showed decreases
in both population and household income. Three others planning areas had increases in
population and decreases in median household income:
Upper Rock Creek had a 37.5 percent increase in population with a decrease of 13.6
percent in household income.
Fairland’s population increased by 15.1 percent; at the same time the median income in
the area decreased by 10.1 percent.
Aspen Hill’s population increased by 7.8 percent but the median household income
decreased by 11 percent.
These changes do not necessarily indicate decline, but do indicate change. Further
investigation of these areas may show that increases in development of multi-family residential
could account for the changes, and there may be other factors that contribute to the changes in
the community.
16
Figure 8: Rate of Population Change vs. Median Household Income Change*
40%
30% Populatio
n
20%
Median
10% Househol
d Income
0% Change
‐10%
‐20%
As Be Cl Co Da Da Fa Ga Ge Ke Ke No Ol Po Po Ro Sil Ta Tr Up
pe th ov le m rn irl ith rm m ns rth ne ol to ck ve ko av pe
n es er sv as es an er an p in B y & es m vil r S m ila r R
Hi da ly/ ille cu to d s t M g t e vil ac l e p a h oc
ll /C No /W s & wn bu ow ill on th Vi le /C rin Pa
he rg n / Fo /W es cin & ab g rk k C
rw h V & ur d i Vi in re
vy oo ite ici he a/ ty c ek
C d O ni V i Co at Ga in Joh
ha ak t y c in rn on rre ity n
se ity er tt
s Pa
rk
Finding: In many cases, average monthly household costs for County homeowners and
renters have increased substantially while median household income has diminished.
Comparing rate of change in housing costs to median income within planning areas can
also shed light on possible issues concerning neighborhood change. According to census update
survey data, six planning areas show considerable increases in either or both owner and renter
average monthly housing costs and decreases in median household income between 1997 and
2005:
In Aspen Hill, average monthly household costs for homeowners rose 12 percent and
average monthly household costs for renters rose 13 percent; however, the median
household income decreased by 11 percent.
In Colesville/White Oak, average monthly household costs for homeowners and renters
increased by 11 and 18 percent respectively; median household income for this planning
area decreased by three percent.
Fairland’s average monthly household costs for homeowners and renters increased by 7
percent and 8 percent, respectively; median household income for the planning area fell
10 percent.
In the Gaithersburg & Vicinity planning area, average monthly housing costs for renters
significantly outpaced the same costs for homeowners (18 percent increase for renters; 5
17
percent increase for homeowners); the median household income for this planning area
decreased by 4 percent.
Of all planning areas that experienced median household income decreases, Takoma Park
experienced the largest increase in average monthly household costs for homeowners (28
percent). Average monthly household costs for renters actually decreased by 2 percent.
The median household shrunk 15 percent.
Upper Rock Creek experienced notable changes in its housing costs and median
household income. Average monthly household costs for renters rose 21 percent and
average monthly household costs for homeowners rose 6 percent. During the same time
period, the planning area’s median household income decreased by 14 percent.
Figure 9: Comparison of Average Monthly Housing Costs and Median Household Income
– Rate Change From 1997 to 2005
G nd n
K p H n ici
N ing ou
R ma &
U la h rk
C esd
C erly he
D svi orw Ch
D sc h d
Fa est & V Oa
G er
K an rg
O B /W orn
Po esv ini ar
Si il
Ta r Sp
Tr
M rR
sp
et Hi
en il
or
oc c
pp
ai
lo
ol
am lle o a
ar
em to &
er sb
ln e t h
on o
lv le in ty
av a P
ol Vi a/G n
irl ow i c k
ko ri
to ille ty ret
h
th
th
v
e N vy
e
m u
en
kv /Ca ici
e
e
a
tg ck
a / W o se
om C
us ite
a
t
/
o
ll
l
/C
er ee
/F
w V
a
n C
ng
c
y k
b
C
V
r
r
ou
e ers
Jo
in
ni
nt
i ty
hn
y
ni
ty
tP
ar
k
18
Finding: In many planning areas, the number of rental housing units declined considerably.
Between 1997 and 2005, the number of rental units decreased in 14 of the County’s 21 planning
areas. The unit decreases ranged from 1.3 percent to 7.2 percent. The decrease in rental units in
these planning areas, however, is lower than the County-level decrease (10.2 percent).
Da ille woo se
Ge rsb
Ke anto & V
No gto r C
Ro ac/ Vic
Si ille in J
Ta Spr
Tr a P
Up ah
As
Be Hil
Co rly/ evy
Da scu ite
Fa sto Vici
Ga d
Ke Hil
Ol Bet he ers
Po & V da/ n
M Ro
lve
ov
av
on ck
ne
irl wn nit
ko ing
pe
to e &
th l
ns l/F
le No Ch
ol
ck
m /W d
rn s & Oa
ith
rm ur
pe
rth n/ orn
e Ch
il
an
sv
e
a
r
e
tg Cr
y
m
in
e
n
r
e
om ee
/
he at
ar
Ca ini
er k
i
w
cin
ou
k
g
n cini
r
y C
W
b
h
ou
Ga
oh
i
nt
o
ty
y
n
r
a
y
k
ty
P
ar
k
Staff identified possible factors for decrease in units to include conversion of apartment to
condominium of sale of rental houses to owner occupied housing. The decrease of rental units,
while very small in some cases, is a cause for concern. Any decrease in housing units in the
County will impact housing choices of residents, particularly those with lower household
incomes.
NEXT STEPS
The staff analysis was only a sample of what can be done with the Census Update Survey results.
Combining CUS with the recently compiled housing inventory will yield even richer results. The
2008 Census Update Survey has just been mailed to Montgomery County residents. When the
data are available later this year, the results should be compared to 2005 as this will reflect the
recent changes in the economy not reflected in the 2005 CUS. Currently, staff should continue to
19
collect and map relevant housing information such as foreclosures, housing sales, and rental
information to combine with the CUS. If the economy continues to worsen with increasing
energy costs, the housing burden could become greater on those residents least able to afford.
20