The Washington, DC region is great >> and it can be greater.


As Arlington booms, traffic drops

Traffic on several Arlington roads is lower today than decades ago, despite huge increases in density and activity.

The Orange Line corridor, where new high-rises lower traffic counts. Photo by Arlington.

Since 1996, Arlington has boomed. It's added millions of square feet of new development, some of the tallest high-rises in Virginia, and about 50,000 new residents. And in that time, traffic counts have declined.

The explanation: Virtually all the growth has happened in Arlington's Metrorail corridors, where using transit, biking, and walking are the norm. As mixed-use high-rises have replaced the previous generation's car-oriented retail, the new residents don't have to drive as much.

Traffic goes down

Street SegmentStreet Type19962011/2012% Change
Lee Hwy - RosslynEW 6-lane arterial37,77031,951-15.4%
Wash. Blvd. - VA SquareEW 4-lane arterial20,46917,500-14.5%
Clarendon Blvd.EW 2-lane 1-way arterial13,98013,292-5.0%
Wilson Blvd. - ClarendonEW 2-lane 1-way arterial16,36812,603-23.0%
Arlington Blvd.EW 6-lane arterial55,86565,25916.8%
Glebe Road - BallstonNS 6-lane arterial35,23031,000-12.0%
Glebe Road - S. of Col. PikeNS 4-lane arterial29,00027,000-6.0%
George Mason DriveNS 4-lane arterial20,00220,5182.3%
Jefferson Davis Hwy - N. of GlebeNS 6-lane arterial52,00044,000-15.4%

Traffic declined most dramatically on the most urban and high-density streets. Wilson Boulevard, the main street through the Rosslyn-Ballston corridor, saw the steepest reduction, 23%.

The next steepest drops were on Route 1 through Crystal City, and on Lee Highway in Rosslyn, which each fell 15.4%.

Why these streets? They've got the best transit, but that's only part of the story. Thanks to high density and mixed-use, many trips that once required a car now happen on foot. Why drive to the store and fight parking when it's only a block away, and walking there only takes 2 minutes?

Other roads that don't mirror Metro lines saw reductions as well. For example the north-south Glebe Road, which saw 6-12% less traffic.

Traffic did rise on some roads. George Mason Drive traffic increased 2% over the period, and Arlington Boulevard (Route 50) went up 16%.

But George Mason is in the western, more suburban part of Arlington, where there's been less growth and less of a shift to the car-free diet. And Route 50 is a major commuting route for traffic from the outer suburbs, where smart growth is less prevalent, and more growth still means more cars.

Transit ridership goes up

During the same time period, Arlington's transit ridership is way up.

FY1996 ActualFY2013 Actual% Growth
Metrorail Arlington Stations45,335,00059,528,74431.3%
Metrobus Arlington Routes12,049,00014,848,03623.2%
VRE—Crystal City567,0001,102,07694.4%
Arlington Transit (ART)105,0002,644,0002,518%
Total Annual Ridership58,076,00078,122,85634.5%

Arlington's local bus operation, ART, went from a very small system to a major countywide network. The Crystal City VRE stop saw its ridership double (VRE service began in 1992). Metrorail and Metrobus grew by 31% and 23%, respectively.

Put it all together and you get one staggering statistic: Fully 40% of all Virginia statewide transit trips either begin or end in Arlington.

It almost didn't happen this way

Arlington has embraced transit-oriented development and walkability for a long time, but in the 1970s and '80s when the county was originally debating its plans, some of Arlington's choices seemed like risky moves.

Building the Metro through the heart of Arlington's business districts rather than in highway medians added huge expense to the project. But it also made possible places like Clarendon and Ballston as we know them today. Without that big initial investment, they'd likely look more like Seven Corners or Bailey's Crossroads.

For the next generation, Arlington hopes to add to its transit-oriented successes with the Columbia Pike Streetcar, the Crystal City-Potomac Yard Transitway, and new Metrorail station entrances, confident that these will put more people on transit and take more cars off the streets.

Canaan Merchant was born and raised in Powhatan, Virginia and attended George Mason University where he studied English. He became interested in urban design and transportation issues when listening to a presentation by Jeff Speck while attending GMU. He has lived all over Northern Virginia and now lives in Burke.  


Add a comment »

Arlington discussed where to locate Metro in the 60s and 70s. Density around the Metro stations was planned long before the 80s.

by JP on Jun 30, 2014 12:28 pm • linkreport

Declines in traffic on Lee HIghway and Crystal City probably have more to do with their continuing stangantion than anything transit related. Unless there is a secret metro stop on Lee Highway that nobody told me about.

Ancedotaly I don't remember being stuck in traffic on Wilson 15 years ago, and now that is a semi-regular occurence. So I'd be a bit concered on where the data comes in.

In fact, that congestion at key times probably is moving traffic to other streets.

by charlie on Jun 30, 2014 12:28 pm • linkreport

Money for the Ballston Second Entrance should be allocated to the Columbia Pike Streetcar.

by mcs on Jun 30, 2014 12:39 pm • linkreport

Should be interesting to see the traffic volumes after the Silver Line opens.

by mcs on Jun 30, 2014 12:46 pm • linkreport

That's silly, charlie. There is a building boom in the Rosslyn - Ballston corridor. Are you telling me that is due to it's stagnation? Crystal City is more mixed but again why would transit ridership go up if its stagnating. I mean it's not even logical.

by BTA on Jun 30, 2014 12:57 pm • linkreport

Cool! Numbers!

by Jasper on Jun 30, 2014 1:08 pm • linkreport


Actually I know a lot of people who live on the 29 corridor, especially around Veatch/CherryDale, that use Metro. Its a 8 minute walk. Not to mention Route 29 does have a metro stop next to it, its called East Falls Church, which has seen usership go up dramatically as well the past 10 years, without an increase in parking mind you... so what does that tell you? Ah yes, people living in the 29 corridor are using it via walking/bus/bikes.

by Navid Roshan on Jun 30, 2014 1:15 pm • linkreport

"Fully 40% of all Virginia statewide transit trips either begin or end in Arlington"

Not really surprising when you consider that Arlington is the only jurisdiction in Virginia that even remotely cares about transit. Outside of Northern VA "transit" is a foreign word.

by K Street on Jun 30, 2014 1:21 pm • linkreport

@BTA; Lee highway is most certainly NOT the R-B corridor.

@Navid; true enough on EFC.

That being said, the majority of Lee Highway in Arlington has long been neglected and is not transit friendly.

by charlie on Jun 30, 2014 1:21 pm • linkreport


US-29 is very much transit accessible. A lot of it is within a 10-20 minute walk to the metro depending on where you are, and outside of there the bus can take you to the metro. I've known several people to live along Lee Highway and they've all made the trek to Metro.

by KingmanPark on Jun 30, 2014 1:57 pm • linkreport

What would make this analysis more powerful is to compare Arlington's reduced traffic volumes with other parts of NOVA where it has skyrocketed. Then let's compare the planning and transportation strategies of those two areas to determine which strategy works at reducing traffic.

For decades, people have claimed that building more roads is the solution to reducing traffic congestion. And, if it hasn't worked, it's only because we haven't built enough roads. Now we have the data to prove that theory wrong.

by Falls Church on Jun 30, 2014 1:59 pm • linkreport

There are a lot of buses in Arlington, they carry a lot of people. A lot of them go along Lee Highway.

by BTA on Jun 30, 2014 2:39 pm • linkreport

Lee Highway also runs through Rosslyn, one or two blocks from Rosslyn Metro.

by Ian Cameron on Jun 30, 2014 3:27 pm • linkreport

It would be interesting to see the traffic numbers of the 1970s and early 1980s as a comparison of before Metro, also before I-66 and wider Arlington Blvd., I-395. This is something that has been continually stressed but many are reluctant to believe given the amount of density that has occurred in the county. Folks also forget that commuting patterns have changed significantly since with the increase in density in Tysons. I'm willing to bet that you did not have higher numbers of west bound travelers but most were heading downtown.

by Tom on Jun 30, 2014 3:54 pm • linkreport

One of the untold stories of this very good story about how one can retrofit what was an aging suburban bedroom community into a thriving more urban place through great land use planning and investments in transit-bike-walk infrastructure is that the TDM (transportation demand management) strategies that the County has simultaneously provided over the last 20 years have helped make this happen too. Its the land use, the transportation infrastructure and these programs altogether. Research shows us we get even greater utility out of the system we're building by using these TDM strategies. You can read more here: or on our research and development web site here:
Chris Hamilton
Commuter Services Bureau Chief, Arlington DOT

by Chris Hamilton on Jun 30, 2014 5:29 pm • linkreport

wrt Chris' comments, I'd argue it's the combination of multiple strategies: (1) transit; (2) densification; (3) jobs-work balance; (4) TDM.

I am planning on writing a piece about Pinellas vs. Hillsborough Counties in FL about transit planning and funding, and the Greenlight Pinellas initiative has a great great great graphic about what I call the "physics of transit" wrt the space effects.

Sure you can build more roads, but as long as most trips are SOV, it is a pretty inefficient use of space.

OF Course, wrt Arlington is the master transportation plan, which prioritizes throughput of people, not so much by mode. In most other mobility planning, throughput of motor vehicles is prioritized, and that's a losing battle.

Anyway, nice piece, great story.

by Richard Layman on Jun 30, 2014 5:42 pm • linkreport

To what extent did companies/the government fully or partially subsidize metro use in '96?

by TheCharlie on Jun 30, 2014 6:21 pm • linkreport

So wait, Arlington surveyed every major arterial in the County EXCEPT Columbia Pike? Or did they just omit Columbia Pike from the press release because they know that - without a streetcar line - traffic there is undergoing the same pattern? If you can't win the argument, spend $60K on propoganda and cook the books!

by Piker on Jun 30, 2014 6:29 pm • linkreport

I hadn't looked at it since I wrote the story but the difference in traffic at the Pike/glebe fell about 3000 cars between 1995 and 2012.

by Canaan on Jun 30, 2014 8:45 pm • linkreport

That said, the streetcar aims to reduce that VMT significantly above that.

by Canaan on Jun 30, 2014 8:52 pm • linkreport

How does the ARL ridership numbers compare to other NOVA areas like DASH and the Fairfax Connector. The ARL numbers seem lower than I would have expected. I think DASH may have a higher ridership but I did not confirm.

by novalyfer on Jul 1, 2014 12:22 am • linkreport

Where did you get the 50K population estimate from?

Census Bureau ... at least the demographic estimates that Google automatically reads in suggests a 40K increase from 1996 to 2012.

by Geof Gee on Jul 1, 2014 10:34 am • linkreport

ART is probably artificially low because there is robust Metrobus and Metrorail service in the area already. Most of WMATAs highest ridership virginia routes go through Arlington and many of the higher ridership Metro stations are there. ART is really supplemental in that sense. It looks like DASH has more service on the road as well.

by BTA on Jul 1, 2014 11:31 am • linkreport


Not only did they exclude Columbia Pike, they excluded include most of Washington Blvd, Wilson Blvd and Lee Hwy.

by Brett on Jul 1, 2014 12:15 pm • linkreport

Great article. Less traffic is great, but the broader implications of these numbers are even better. The density increases (made possible by transit) that are driving these numbers also have a huge positive impact on the tax base, less road infrastucture cost, lowered greenhouse gases, more walking (health impacts) and generally happier and healthier place to live...Bravo to the Arlington planning visionaries....its also nice to see real data that refutes the taffic Armegedon claims everytime a new development is proposed.

by CB2 on Jul 1, 2014 12:42 pm • linkreport

Do you suppose that the dramatic drop in traffic on Jeff-Davis in Crystal City has something to do with BRAC?

by Foxy on Jul 1, 2014 1:20 pm • linkreport

Some good stuff - but the data is unable to control for the population between 1996 and 2011. Alternative explanations may include such things as the original population in that area(that might have been more suburb loving) has moved out and replaced by those city lovers who self-selected into a transit rich and mixed use supported environment. Still, it it is good to see!

by Erics on Jul 3, 2014 3:57 pm • linkreport

@Erics: You're probably right that Arlington's population has changed a lot since 1996. The county is growing and has something on the order of 12% turnover annually. The biggest growth in net migration is among people in their 20s. The county is also becoming more diverse and international.

Fact sheet:

None of that conflicts with Canaan's explanation. New residents may have a pre-existing preference for a transit oriented lifestyle, but if the urban design and planning didn't support that, they'd be driving instead.

by Laurence Aurbach on Jul 4, 2014 1:23 pm • linkreport

There have been declines on many streets, there was a large increase on Arlington Boulevard:,-77.098591,3a,75y,264.49h,81.3t/data=!3m4!1e1!3m2!1sDi695rj8W1-Y6IsyAgPvbA!2e0!4m2!3m1!1s0x89b7b49324c384fb:0x7e61de89076d22c5

Washington DC is rife with highways and beltways so you can jump on quickly and get off a high capacity road quickly. Portland, OR is that way too.

Since 1996, Arlington has boomed. It’s added millions of square feet of new development, some of the tallest high-rises in Virginia, and about 50,000 new residents.

So, 50,000 residents over two decades…2,500 per year. Mmm..and yet on their figures for transit they are claiming an increase of 30,000,000 milllion? So they’ve built a commuter city. My guess would be people are reverse commuting in from Washington DC — which would then essentially give the lie to your arguements about centralized urban density.

By building up a satellite city like Arlington, you alleviate the problem of getting modern jobs into overly dense cities.

by John Bailo on Jul 6, 2014 9:57 am • linkreport

The survey did not measure traffic on the most congested roads in Arlington County: I-66 and I-395. These roads have absorbed much of the increased traffic that Arlington County's excessive development has generated. This is especially true for I-66, which parallels the Rosslyn-Ballston corridor, but is also true for I-395, which travels near Shirlington, Crystal City and Pentagon City .

This survey was fatally flawed. It is classic example of lying with statistics. The survey proves nothing.

by skeptic on Jul 7, 2014 3:46 pm • linkreport

traffic on I395 and I66 in arlington is largely folks going from FFX and points south and west to DC, so wouldnt really address the changes in how folks in Arlington get around. But does anyone really think that folks in north arlington who can take Lee or Wilson have switched to I66? Are people from Crystal City driving onto I395 to get into DC (some may be using I395 to reverse commute, I suppose)

by AWalkerInTheCity on Jul 7, 2014 4:07 pm • linkreport

It is true that much traffic on I-395 and I-66 does not enter or leave the highway in Arlington. Nevertheless, some of this traffic indeed does this.

The survey could have determined the numbers of vehicles that enter and exit these highways at interchanges within Arlington County. However, the survey failed to determine this.

The survey was therefore grossly inadequate. Nobody can draw any valid conclusions from its findings.

As an example of this inadequacy, the survey could have answered your question, "But does anyone really think that folks in north arlington who can take Lee or Wilson have switched to I66?"

To answer your question, the survey could have determined the number of vehicles that entered and exited I-66 in Rosslyn, Lyon Village and Ballston. Judging from the traffic congestion that regularly occurs during morning and evening rush hours on I-66 and nearby streets in Ballston, these numbers are not trivial.

by skeptic on Jul 7, 2014 8:16 pm • linkreport

There was no survey. It's just an examination of traffic counts on certain roads.

You could go to VDOTs website and look at counts for specific points along certain points but even if the numbers are higher there I don't know how it disproves the traffic declines elsewhere. Nor what we infer from that.

by drumz on Jul 7, 2014 8:27 pm • linkreport

OK, so it was not a survey. Whatever it was, it did not report counts of traffic entering and leaving I-66 and I-395 at interchanges in Arlington.

Of course, such data can not and would not disprove the traffic declines elsewhere. However, if the numbers of vehicles using Arlington's interstate highway interchanges have increased, the data would suggest that traffic has declined elsewhere because it has moved to the interstates, rather than for other reasons.

by skeptic on Jul 7, 2014 8:53 pm • linkreport

The survey could have determined the numbers of vehicles that enter and exit these highways at interchanges within Arlington County. However, the survey failed to determine this.

The survey was therefore grossly inadequate. Nobody can draw any valid conclusions from its findings.

Those vehicles have to go somewhere after they exit I-66 - like onto the local roads where traffic counts were conducted.

Beyond that, the traffic counts compare the present to 1996, well after I-66 was built. Traffic DECLINED on most of the roadways counted. It's silly to think that would be mostly due to traffic moving to an interstate that had been in existence for a decade prior to the first count.

by MLD on Jul 8, 2014 8:18 am • linkreport

People whose trips formerly began and ended in Arlington may now instead be using interstate highways to travel between Arlington and other jurisdictions to live, work, shop and/or engage in recreation. That thought does not seem silly.

by skeptic on Jul 9, 2014 3:51 pm • linkreport

Add a Comment

Name: (will be displayed on the comments page)

Email: (must be your real address, but will be kept private)

URL: (optional, will be displayed)

You can use some HTML, like <blockquote>quoting another comment</blockquote>, <i>italics</i>, and <a href="http://url_here">hyperlinks</a>. More here.

Your comment:

By submitting a comment, you agree to abide by our comment policy.
Notify me of followup comments via email. (You can also subscribe without commenting.)
Save my name and email address on this computer so I don't have to enter it next time, and so I don't have to answer the anti-spam map challenge question in the future.


Support Us