Arlington
County would seem to have everything going for it.
It sits at the heart of the booming Washington
metropolitan area but across the Potomac from the
frequently disordered District of Columbia, a location that
confers Virginia's favorable business climate. The county
has done a suburb job of urban planning, clustering
development around its Metro stops. County coffers
are blessed by one of the largest employment
concentrations in the metro area; businesses pay an
exceptionally high percentage of the county's taxes.
And inhabitants earn among the highest incomes in
the country: Arlington's $51,200 average per capita
income in 2001 was almost 60 percent higher than the
state average.
Why,
then, are people leaving? More to the point, why are
higher-income residents leaving?
According
to Internal Revenue Service data, 4,600 more
taxpayers and their dependents moved out of the
county between 2001 and 2002 than moved into it.
That's in a jurisdiction of 190,000 residents over
just one year! This isn't Beirut. It isn't even East
St. Louis. Even more alarming, the median income of
taxpayers who left the county was $42,200, while
that of newcomers was a mere $34,700. Add it all up
and the net loss of countywide income from out-migration over
in-migration was $214 million.(1)
Economic development practice
traditionally focuses on attracting
industrial and commercial investment, which can be
taxed in the form of real property. But in a
Knowledge Economy in which human capital is the main
engine of economic
prosperity, Virginia localities need to think more
broadly about what constitutes a healthy tax base.
High-income taxpayers generate more in sales and property taxes
than poorer people do. They also are incredibly
mobile -- roughly one in five Americans changes
residences each year -- and they can relocate with
ease. Virginia communities need to do a better job
of hanging onto these people.
Arlington is
a microcosm for Virginia's cities and urbanized
counties. According to IRS data, other Virginia cities
are leaking their productive populations, too.
In 2002, Richmond, Alexandria, Hampton, Roanoke,
Charlottesville and Newport News collectively
suffered a net loss of 11,000 taxpayers and
dependents, not to mention nearly $200 million in
gross adjusted income, through out-migration.
What's
especially worrisome is to see the travails of
Virginia's older cities oozing into the
older, urbanized counties -- the older suburbs.
Henrico and Fairfax Counties, the economic
powerhouses of their regions, have suffered
significant net losses through out-migration. The
median salary of out-migrants from Henrico County
exceeded that of in-migrants by roughly $2,600,
draining some $50 million in income from the county.
For
Fairfax County, the numbers are staggering: 12,500 more out-migrants (taxpayers and dependents)
than in-migrants. The gap in median income between
those who moved out and those who moved in was $5,400 per taxpayer. The
total loss of income to
out-migration amounted to $640 million. While
Fairfax moves from strength to strength in the field
of business recruitment, many affluent residents
are heading for the county line.
The
big winners across Virginia are the "exurban" counties, fast-developing
localities on the outer fringe of development.
Measured by gains in total gross adjusted
income attributable to migration, Loudoun, Prince
William and Stafford Counties topped the list.
Ranked by gains in median income,
Fauquier County (on the Washington metro fringe),
Powhatan County (on the Richmond metro fringe), and
Fluvanna County (on the Charlottesville metro
fringe) came out miles ahead. For those localities,
the median income of
in-migrants exceeded that of out-migrants by more
than $10,000.
All these numbers should be used
with caution. First, they reflect the performance of
a single year only. To draw meaningful conclusions
about a single locality, it would be advisable to
draw upon three to five years' of numbers. Secondly,
the IRS data is not necessarily an accurate
reflection of broader population movements: It
tracks only people who filed federal tax returns,
and their dependents.
Still,
the same pattern prevails in every metropolitan
area: Higher-income people are moving towards the
urban fringe, lower-income people are moving towards
the urban core.
The
conventional wisdom of the past 40 years suggests
that middle-class families, especially those with
children, are fleeing the crime, dysfunctional
schools and high taxes of the core cities. That
might explain what we see in Richmond, for instance,
but why, then is Henrico witnessing the same trend?
The crime rate here, where I live, is enviably low,
the schools are among the best in the state, and
taxes are the lowest of any urbanized county in the
state.
How,
also, do we explain Fairfax County, one of the
leading high-tech centers in the country and the
economic dynamo of the Washington region. As a
percentage of income, its taxes are hardly
oppressive. Its
schools are among the best in the country.
Skip
past chart
Migration
Winners and Losers
(Change
between 2001 and 2002 IRS filings) |
|
Net
Pop.
Gain
|
Gain
in
median
income ($)
|
Gain
in total income ($1,000s)
|
Loudoun |
8,815 |
7,268 |
351,088 |
Prince
William |
7,508 |
391 |
177,378 |
Stafford |
4,540 |
3,977 |
84,370 |
James
City |
1,819 |
6,906 |
79,934 |
Fauquier |
1,397 |
11,038 |
66,137 |
Hanover |
2,015 |
7,625 |
59,692 |
Suffolk |
2,422 |
5,984 |
58,590 |
Spotsylvania |
4,464 |
6,600 |
55,078 |
Chesterfield |
3,135 |
2,982 |
42,484 |
Frederick |
1,506 |
4,604 |
38,795 |
Culpeper |
1,029 |
6,574 |
31,885 |
Bedford
County |
851 |
3,132 |
20,852 |
Fluvanna |
727 |
10,643 |
19,638 |
Goochland |
317 |
1,736 |
18,975 |
Albemarle |
106 |
-2,074 |
18,868 |
Augusta |
778 |
2,993 |
18,187 |
Orange |
561 |
3,102 |
15,778 |
Warren |
552 |
4,864 |
15,397 |
Powhatan |
654 |
10.779 |
15,379 |
Franklin
County |
334 |
3,863 |
13,809 |
Botetourt |
607 |
5,083 |
13,709 |
Louisa |
507 |
4,179 |
13,289 |
Clarke |
200 |
9,589 |
12,532 |
Shenandoah |
493 |
3,204 |
12,054 |
Middlesex |
168 |
8,071 |
12,033 |
Caroline |
336 |
4,440 |
11,813 |
Mathews |
54 |
10,160 |
11,611 |
Northumberland |
235 |
6,957 |
11,019 |
York |
985 |
1,491 |
10,989 |
King
George |
396 |
-583 |
10,211 |
New
Kent |
308 |
6,535 |
9,887 |
Norfolk |
-621 |
-2,296 |
9,421 |
Radford |
616 |
-2,067 |
9,287 |
Lancaster |
69 |
4,120 |
8,104 |
Gloucester |
268 |
2,946 |
7,584 |
Virginia
Beach |
-71 |
-1,927 |
7,201 |
Chesapeake |
1,243 |
557 |
6,798 |
Dinwiddie |
287 |
936 |
6,687 |
Rockbridge |
33 |
2,662 |
6,390 |
Accomack |
158 |
2,616 |
6,370 |
Nelson |
165 |
1,423 |
5,801 |
Greene |
263 |
3,316 |
5,596 |
King
William |
228 |
3,681 |
5,585 |
Madison |
168 |
4,388 |
5,356 |
Floyd |
135 |
3,583 |
4,893 |
Mecklenburg |
72 |
3,411 |
4,655 |
Smyth |
93 |
2,028 |
4,558 |
Carroll |
-85 |
3,481 |
3,301 |
Lee |
212 |
4,074 |
3,252 |
Washington |
183 |
944 |
3,182 |
Bland |
-42 |
3,356 |
3,074 |
Winchester |
-36 |
-348 |
2,572 |
Northampton |
4 |
4,347 |
2,522 |
Amelia |
62 |
5,139 |
2,405 |
Rockingham |
134 |
365 |
2,248 |
Westmoreland |
107 |
2,539 |
2,206 |
Prince
Edward |
186 |
-1,540 |
2,053 |
Charlotte |
91 |
1,642 |
1,921 |
Rappahannock |
8 |
-1,687 |
1,910 |
Staunton |
23 |
2,349 |
1,659 |
Wythe |
136 |
1,819 |
1,597 |
Buckingham |
-80 |
-1,780 |
1,449 |
Richmond
County |
12 |
-1,467 |
1,382 |
Waynesboro |
183 |
635 |
963 |
Colonial
Heights |
55 |
-1,162 |
942 |
Cumberland |
3 |
2,554 |
880 |
Galax |
-19 |
1,771 |
667 |
Surry |
85 |
-1,947 |
493 |
Craig |
16 |
3,808 |
278 |
Grayson |
15 |
-69 |
266 |
Southampton |
-30 |
2,038 |
167 |
Bath |
16 |
1,404 |
64 |
Bedford
(city) |
-14 |
570 |
34 |
Emporia |
17 |
-4,489 |
-123 |
Highland |
-44 |
312 |
-149 |
Greensville |
-15 |
-1,364 |
-224 |
Scott |
-21 |
-236 |
-249 |
Lunenburg |
76 |
-429 |
-265 |
Brunswick |
-58 |
621 |
-268 |
King
and Queen |
0 |
2,060 |
-326 |
Buena
Vista |
-13 |
-1,932 |
-343 |
Covington |
-31 |
-1,899 |
-393 |
Sussex |
-52 |
1,504 |
-407 |
Nottoway |
-16 |
782 |
-410 |
Dickenson |
33 |
4,809 |
-543 |
Norton |
44 |
-6,143 |
-588 |
Giles |
82 |
614 |
-637 |
Poquoson |
133 |
9,159 |
-646 |
Page |
71 |
1,135 |
-657 |
Pittsylvania |
-180 |
1,294 |
-706 |
Patrick |
-8 |
572 |
-849 |
Lexington |
-9 |
-2,435 |
-867 |
Wise |
-85 |
731 |
-870 |
Alleghany |
2 |
-28 |
-1,027 |
Essex |
72 |
-562 |
-1,057 |
Russell |
-6 |
58 |
-1,124 |
Pulaski |
-88 |
1,756 |
-1,219 |
Amherst |
-50 |
2,029 |
-1,228 |
Appomattox |
-83 |
-1,875 |
-1,401 |
Hopewell |
30 |
-1,076 |
-1,558 |
Roanoke
County |
297 |
69 |
-1,600 |
Franklin
(city) |
-74 |
-1,082 |
-1,912 |
Martinsville |
-133 |
-2,834 |
-2,660 |
Falls
Church |
-64 |
-2,279 |
-2,749 |
Tazewell |
116 |
-370 |
-3,229 |
Manassas
Park |
-56 |
-2,067 |
-3,344 |
Fredericksburg |
-93 |
-4,568 |
-3,809 |
Campbell |
187 |
-314 |
-3,844 |
Halifax |
-112 |
1,139 |
-3,904 |
Williamsburg |
-153 |
-3,884 |
-4,021 |
Danville |
-149 |
-1,687 |
-4,987 |
Lynchburg |
-465 |
-811 |
-5,091 |
Manassas |
43 |
-1,653 |
-5,184 |
Prince
George |
-46 |
2,186 |
-5,305 |
Henry |
-306 |
803 |
-6,406 |
Harrisonburg |
-121 |
-2,815 |
-6,416 |
Fairfax
(city) |
-250 |
-3,701 |
-6,842 |
Montgomery |
1 |
-6,585 |
-6,958 |
Petersburg |
-292 |
-2,566 |
-7,673 |
Portsmouth |
-467 |
-650 |
-10,924 |
Buchanan |
-230 |
-4,424 |
-11,417 |
Charlottesville |
-350 |
-6,261 |
-14,195 |
Roanoke
(city) |
-995 |
-1,471 |
-17,314 |
Isle
of Wight |
495 |
9,702 |
-19,265 |
Alexandria |
-3,542 |
-4,035 |
-20,375 |
Charles
City |
151 |
4,109 |
-23,340 |
Newport
News |
-856 |
-2,692 |
-38,545 |
Hampton |
-1,879 |
-2,770 |
-48,415 |
Henrico |
-318 |
-2,577 |
-53,103 |
Richmond
(city) |
-1,704 |
-4,316 |
-55,575 |
Arlington |
-4,634 |
-6,540 |
-213,055 |
Fairfax
County |
-12,521 |
-5,400 |
-640,145 |
Source:
Internal Revenue Service
Notes...
Net
Population Gain/Loss: Difference between the
number of federal income tax exemptions filed for out-migrants and in-migrants.
Gain/Loss
in Median Income: Difference between the median
Gross Adjusted Income reported by out-migrants and
in-migrants. A positive number indicates that
in-migrants earned more than out-migrants; a
negative number indicates that out-migrants earned
more than in-migrants.
Gain/Loss
in Total Income: Difference between the
aggregate Adjusted Gross Income of all out-migrants
and in-migrants. A positive number represents a gain
in collective income for residents of the locality.
There
are many cross-cutting currents in the IRS data but
the flood of higher-income taxpayers to the suburbs
stands out as the dominant stream. In perhaps the clearest
illustration of this trend, the data demonstrate that thousands of
residents are spilling down the I-95 Corridor.
South of Washington, D.C., I-95 runs through five
counties before it reaches the metropolitan fringe: Arlington, Fairfax, Prince William,
Stafford and Spotsylvania. We shall examine each in
turn.
Arlington
County, in the urban core, lost 4,600 people in
net out-migration in 2002. Where did they go? Of the
76,000 people who either entered or left the county that
year, most shuttled back and forth between Arlington
and other urbanized locales such as Washington,
D.C., Alexandria and Montgomery County, Md. But a
significant number broke out of the urban core. Fairfax accounted for the greatest net
out-migration: 3,500 taxpayers and dependents. The
median earnings of migrants from Arlington to
Fairfax exceeded the earnings of citizens on the
reverse trek by roughly $3,000.
Fairfax
County, despite its influx from Arlington, lost
12,500 (net)
migrants in 2002. A great many headed south to
Prince William County -- 8,000 more than came the
other way. Notably, the median income of migrants
leaving Fairfax was about $6,800 more than
that of taxpayers traveling against the flow.
Prince
William County, one of Virginia's fast-growth
counties, was a big gainer in the migratory
sweepstakes, scooping up 7,500 more arrivals than
departures. But that didn't stop the torrent down I-95. Prince William suffered a net
out-migration to Stafford County of almost 2,200 people. Incredibly, the median income between people
moving to Stafford was $9,000 higher than those
heading the other way.
Stafford
County, another fast-growth county, gained 4,500
people in its migratory transactions with other
jurisdictions. Even so, it suffered a deficit with
its neighbor to the south, Spotsylvania County. More
than 700 people moved from Stafford to
Spotsylvania than the other way around. The
difference in median income: $6,200.
The
southward flow towards the metropolitan periphery
finally petered out in Spotsylvania. To the south of
Spotsylvania lies Caroline County, a
rural watershed between the Washington and
Richmond metro areas. A mere 66 more people decamped
south from Spotsylvania to Caroline than moved the
other way.
What's
going on? Taxes are one possible culprit. That's the
answer favored by my friend Tim Wise, president of
the Arlington County Taxpayers Association, who
kindly supplied the raw IRS data I used in my
analysis.
Tim's
reasoning is intuitively appealing: Arlington County has one of the
highest tax burdens in the state. (See his article, "Taxation
and Migration," page 8, in the ACTA's July
newsletter.) But I draw different conclusion from
the data. Taxes in Fairfax County, a major
depository of former Arlingtonians, don't represent
much of a bargain. A typical family of 1.3 moving from Arlington
to Fairfax could save about $350 in local taxes.
Does that justify the transaction costs associated
with the move? Perhaps, as an economist would say,
it would influence the marketplace "on the
margin."
But
a state-level analysis suggests that other factors
are at work. People aren't just leaving Arlington
County -- they're joining a migratory stream headed
toward the urban periphery.
I
would suggest a holistic approach to the subject.
People move to new locations in order to better
their lives. They seek a bundle of attributes, of
which taxes most definitely are one. Virginians
also look at crime rates, the quality of schools,
proximity to their jobs, the availability of urban
amenities and -- this is a biggie -- affordability
of housing.
One
of the main reasons that people move from one
location to another within the same metropolitan
area, in my observation, is to acquire a bigger,
better house. A middle-class lifestyle today
requires a kitchen big enough to entertain the whole
neighborhood in, a family room big enough to house a
big-screen plasma TV with surround-sound, a back
deck for the grill, a bedroom for every kid, and a
three-car garage to store all the junk. America's
need for living space, it seems, is expanding
exponentially.
Virginia's
older urban areas lack the housing stock to
accommodate the Great American Lifestyle. Homeowners
can tack on a room here, convert a duplex into a
single-family dwelling there; occasionally, someone
erects a modern house on a vacant lot. In contrast
to the constrained spaces in old neighborhoods,
developers can build entire tracts of McMansions in the wide-open
spaces of the exurbs.
Here
in western Henrico, the so-called "West End" of metro
Richmond, the housing
stock is severely constricted. There are many fine
neighborhoods in the West End, but there also are thousands
of tiny ranches and bungalows. Fifty years ago,
these dwellings
were considered respectable, middle-class homes.
Today, no one today would consider living in one
except as a starter home. As expectations ratchet up
for ever-bigger houses, an increasing proportion
of Henrico County's housing stock becomes attractive
mainly to people with below-median incomes. The
county isn't losing population through out-migration, but newcomers are making about $2,500 less
than those who are leaving.
If you want to
buy a bigger house with all the cool stuff, West End
residents have
two options: Pay really high prices for a
property close to town -- in other words, pay a premium
for a superior location -- or pay significantly lower prices
by driving 10 to 15 miles beyond the Richmond city line.
Many
people choose the second option. In
sum, the
driving force behind the ever-expanding urban
frontier is a quest for square footage.
Virginians,
take heed!
Affluent taxpayers are migrating to the urban
periphery where they can find bigger houses equipped
with 21st-century amenities. But they're losing the attributes that make a
great place. All we have to do is look at much of
the development that was “new” 50 years ago and realize
that the sheen grows dull. Rather than invest in
upgrading their houses and improving their
neighborhoods, inhabitants just move on to a newer
subdivision and bigger home. Inevitably, the
disposable neighborhoods go into decline.
In
county after county, Virginia is experiencing the
same pattern of declining neighborhoods and
emigrating citizens. Our older communities are
bleeding. Their tax bases are eroding. In many
cases, there's nothing that seems worth saving.
What
can be done? For starters, recognize that square
footage may be the main thing, but it isn't
the only thing. People also care about taxes.
They care how long it takes to drive to work. They
care how long it takes to run their errands. They
care if their neighborhoods possess a sense of
"place" that's worth preserving.
We
can retain some of our wealthier citizens by creating more
great places that offer compensations beyond the square footage of their
living space. I would urge anyone interested in the
characteristics of such places and what it takes to
create them to read Ed Risse's companion
column in this edition of Bacon's Rebellion. (See
"Wild Abandonment,"
September 8, 2003.) The high and rising prices that
homeowners are willing to pay to live in these great
places -- from Georgetown in D.C., to Old Town
Alexandria, from historic Fredericksburg to
Richmond's Fan -- are proof that people crave more
than raw square footage.
But
in our society, intangibles are not enough. People
want both: big houses and the classic
characteristics of great places. We must find a way
to provide both.
--
September 8, 2003
1.
2002 was no fluke. The numbers are
almost as serious for 2001: the loss of 4,100
taxpayers and their dependents. The median income of
out-migrants was $41,300, that of in-migrants only
$34,200. According
to U.S. Census Bureau estimates, Arlington's
total population did not decline nearly as severely
as the IRS data would indicate. The IRS tracks only
taxpayers and their dependents, not all residents.
Census estimated a drop of only 165 inhabitants in
2002.
|