By Yonah Freemark, February 4, 2014
Over the past 30 years, at the same time the percentage of commuters
driving has flatlined, the share of people working from home has
exploded, almost doubling since 1980. Telecommuting is controversial —
some suggest it increases productivity while others maintain the value of offices — but it is gaining popularity no matter its merits.
More than just affecting the way people work, the increasing ease of
telecommuting will dramatically affect the way we adapt our urban
transportation systems. Indeed, an increase in working from home might
suggest that we have less to fear about the future of traffic congestion
than we might have believed.
Since 1980, the share of Americans telecommuting every day has
increased from 2.3 percent to 4.4 percent in 2012. The U.S. Census
Bureau, moreover, reports that
9.4 percent of people now work from home at least one day a week, up
from 7 percent in 1994. (This trend is global; in the United Kingdom,
telework increased by 13 percent between 2007 and 2012 and now represents about a tenth of the workforce.)
These national trends mask the changing ways Americans are working from
home. The Census notes that while home-based workers were mostly
self-employed 30 years ago, they’re now mostly employed by private
corporations. That evidence is backed by a comparison between
telecommuting trends in rural areas and the nation’s largest cities.
Between 1970 and 2010, the share of workers telecommuting in the most
agriculturally dependent states declined by more than 50 percent, while
it more than doubled in the ten most populous counties. In other words,
the profile of telecommuters is becoming more urban.
As you can see in the charts above, based on data from the U.S. Census,
telecommuting is increasingly a city phenomenon. Of the 275 U.S.
counties with more than 100,000 workers in 2012, 91 percent saw an
increase in telecommuting since 1970 — and all counties with more
than 300,000 workers saw an increase during that period. This shift has
been particularly evident in places such as Boulder, Colorado, which
increased its share of working from home from 3.1 percent in 1980 to
11.1 percent in 2012; the Atlanta region’s Cobb and Fulton Counties both
increased their share from less than 2 percent to more than 7.5
percent; and exurban counties like Marin, outside of San Francisco, and
Loudoun, outside of D.C., increased their shares from less than 4
percent to more than 8 percent. Dense urban areas often saw their shares
increase by about 4 percentage points, including Manhattan, King County
(Seattle), and San Francisco.
In other words, telecommuting is occurring everywhere in metropolitan
areas, from dense cities to their far-flung suburbs. The rise of the
Internet is producing more at-home work, but not just, as once believed,
by people who want to live far from their workplaces. Many
telecommuters are likely only a few miles from their potential offices.
What's happening across the country that may explain these changes?
A look at another cross section of Census data, contrasting urban
density and the percentage of people with long commutes to work (in
large counties across the U.S.), suggests that, at least on average,
there's no relationship between those variables and the choice to
telecommute. That means that the decision to work from home is not
directly influenced by the difficult traffic conditions that workers
would otherwise face. Nor, contrary to what we might expect, are people
in less dense, exurban communities more likely to work from home than
people living in center cities with plenty of jobs nearby.
Two other variables, on the other hand, appear to correlate relatively
closely with working from home, at least at the county level. Counties
with a higher share of people holding a bachelor's degree or higher are
likelier to telecommute, and those with a higher percentage of
"professional" workers as classified by the Census
exhibit similar trends. (Professional workers include legal services,
management, business support, and scientific research, among other
occupations.) Other occupational categories have far less of a
correlation to telecommuting. It's not the availability of work nearby
or the way we get to work that's making us telecommuters — it’s the type
of work we’re doing.
Nationally, the share of people holding a bachelor's degree or higher
has increased from 10.7 percent in 1970 to 29.1 percent in 2012; the
share with some college at all has increased from 21.3 percent to 58.3
percent. Meanwhile, the share of people working in professional
occupations has increased from 4.3 percent in 1980 to 10.9 percent in
2012. So the rise in telecommuting is at least partially a consequence
of increases in educational achievement, which could imply the share of
people working at home will probably continue to rise in the decades to
come — and the consequences could be dramatic when it comes to
Take King County, Washington, which includes Seattle. In 2012, 63,000
people, or about 6 percent of the county’s workforce, worked from home,
according to the Census. Taking 63,000 people off the city’s roadways or
train system at rush hour could be making a huge difference in terms of
congestion relief. As of October 2013,
only about 30,000 riders used the region’s light rail system daily. And
the 20 highway lanes entering downtown only have the capacity for about
40,000 vehicles per hour.
Seattle, like many other cities, is now investing in the capacity of
its transportation system in order to ward off congestion; the state is funding a $2 billion highway tunnel under downtown, and the city is extending light rail in almost every direction.
These projects will help fulfill projected increases in peak-hour
demand, but could telecommuting make those expansions unnecessary? Given
current limitations on funding for transportation infrastructure,
planners and politicians have a responsibility to take the rise of
telecommuting and the potential for further declines in peak-hour
traffic into account as they identify the most effective investments for
All chart data from U.S. Census Bureau.