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


Arlington's streetcars will carry more riders than VRE or the entire Richmond bus network combined

Here's the simplest reason to build a streetcar on Columbia Pike: Absolutely tons of people will ride it. The latest ridership projections show that by 2035 there will be more streetcar riders on Columbia Pike and in Crystal City than there are on VRE or riding buses in Richmond today.

Streetcar and buses in Toronto. Photo by Sean_Marshall on flickr.

The latest ridership projections for the streetcar are huge. By 2035, 37,100 riders per day are expected to use the combined Columbia Pike / Crystal City line, which will operate as a single through route. Another 22,700 will ride buses daily, for a total of 59,800 riders in the corridor.

For comparison, VRE carries about 20,000 per day, and as of 2011 (the most recent data available), the entire Richmond metropolitan area bus system carried an average of 35,200 riders per day. That's every bus route in the whole region put together.

Granted, comparing 2035 projections to contemporary ridership is not exactly valid. Surely by 2035 VRE and Richmond's GRTC will be carrying more riders than they are now.

But these comparisons are useful nonetheless. They give us a sense of the scale of transit demand on Columbia Pike.

Let's keep going. According to the American Public Transportation Association's 4th quarter 2013 ridership report, here are more total networks that the Columbia Pike / Crystal City streetcar's 37,100 daily riders in 2035 will beat or approximately match:

  • MARC commuter rail (34,100 riders per day)
  • Regional light rail systems in Baltimore (26,800), San Jose (34,300), New Orleans (20,200), Minneapolis (30,100), Charlotte (15,400), Buffalo (17,400), Pittsburgh (28,300), Houston (38,300), Seattle (33,200), Norfolk (5,000 in 2012)
  • Regional bus networks in Indianapolis (35,000), Memphis (28,700), Nashville (31,200)
  • Fairfax Connector bus system (36,300)
  • Prince William County Omni-Ride bus system (13,400)
Of course there are plenty of bigger systems out there. Here's a sampling from the same data:
  • Subway systems such as New York (8,733,300), WMATA (855,300), Atlanta (221,200), and even Baltimore (48,500)
  • Light rail systems such as Los Angeles (200,900), Sacramento (46,400), Portland (115,300), Dallas (98,300), and Saint Louis (53,000)
  • Bus networks like Baltimore (237,600), Montgomery County Ride-On (86,600), WMATA (441,100), and Norfolk (52,800 in 2012)
Now let's compare the ridership projections for Columbia Pike / Crystal City with other rail projects near the DC region:Each of those projects will carry more riders than the Columbia Pike / Crystal City streetcars, but each of them also has a price tag in the multiple billions of dollars. Streetcars in Arlington will cost hundreds of millions, but produce great bang for the buck.

The bottom line

With 16,000 daily bus riders today, Columbia Pike is already Virginia's busiest bus corridor. By 2035 there will be nearly 60,000 combined streetcar and bus trips on the Columbia Pike / Crystal City corridor, with 42,800 of those coming on Columbia Pike.

Streetcar detractors want you to believe it's practical to move more people on Columbia Pike by bus alone than the entire Richmond or Norfolk regional bus networks move in sum total. They want you to believe it's practical to move more people on Columbia Pike with buses than MARC or VRE move on commuter rail, or that Baltimore, Minneapolis, or Houston move on light rail.

That's ridiculous. The huge transit demand on Columbia Pike easily justifies rail, and it comes at a better cost value than other rail projects around the region. To suggest otherwise ignores reality.

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 »

The question is not how many, but how many more than alternatives which are orders of magnitude less expensive. And ridership projections are just that, projections. One has to factor in the probability that the projections could be overstated. And comparing to other transit systems is pointless.

by Steve on May 14, 2014 10:26 am • linkreport

How many streetcar ridership projections have been accurate? My recollection is that nearly every single one greatly overstated actual ridership.

by michael on May 14, 2014 10:30 am • linkreport

No mention of the newly-increased projected costs?

by Fitz on May 14, 2014 10:31 am • linkreport

Surely by 2035 [...] Richmond's GRTC will be carrying more riders than they are now.

HAHAHAHAHAHA. Considering the system is 50% owned by a suburban county that outright refuses to allow any non-express service in its jurisdiction, and that almost EVERY route requires going through a single downtown transfer station (the 3.5 mile trip from my home to an old job used to take 1h15), it'll take a massive cultural shift for much transit growth there.

by NovaNoob on May 14, 2014 10:42 am • linkreport

Interesting how the projections are never based on resident/employee surveys or preferences. Simply based on planner goals.

by Ben Bradlee on May 14, 2014 10:52 am • linkreport

@ Fitz

No Joke - 40% over budget and rising. Best reason NOT to build.

by Mike on May 14, 2014 10:55 am • linkreport

Interesting how the projections are never based on resident/employee surveys or preferences. Simply based on planner goals.


by drumz on May 14, 2014 10:59 am • linkreport

I'm all for smart growth and rail projects in the area, but it's hard for me to see how rail projects without a right of way make much sense or are cause for increased ridership. Wish we could somehow get funding for heavy rail down to the pike.

by Tim on May 14, 2014 11:02 am • linkreport

@ michael

Although ridership on Norfolk's Tide is off a fraction from its opening year it's still beating projections about 70%. That's the example with which I'm most familiar.

by Peter K on May 14, 2014 11:05 am • linkreport

The streetcar will be terrible news for butthurt Arlingtonians who love getting traffic tickets on the Pike and getting stuck at endless lights. For the rest of us, it will be a convenient way to avoid the logjam going in and out of the District.

by Arlingtonian on May 14, 2014 11:06 am • linkreport


1) Salt Lake City's most recent light rail extension exceeded its two year projection on the first day.

2) Charlotte's light rail line exceeded projections when it opened in 2008.

3) Norfolk's light rail exceeded projections in 2011.

4) Phoenix's light rail line exceeded projections in 2010.

5) Dallas' DART light rail system exceeds projections when its first stub opened in 1996.

There are many more examples such as Minneapolis, New York Metro area (New Jersey), Portland, Seattle, Los Angeles, Pittsburgh, Denver, and a bunch of others I'm probably forgetting that exceeded projections on the first day they opened.

Making stuff up out of thin air isn't the same as not recalling.

by Cavan on May 14, 2014 11:35 am • linkreport

Notice, Cavan, that those are all "light rail" and not streetcar. Seriously, what a tremendously wasted opportunity to not give this it's own right of way. Shame on Virginia (and seriously NOVA, how have you not yet initiated secession/state independence hearings?)

Sounds to me like that ridership projection calls for at LEAST light rail, if not proper underground heavy rail.

by LowHeadways on May 14, 2014 11:47 am • linkreport

1. Ridership projections are partially dependent on what kind of development is encouraged to happen around the rail project. I think Northern Virginia has proven that they are pretty good at rezoning and encouraging development given the process that surrounded the Rosslyn-Ballston corridor.

2. Ridership projections have been more conservative in the past 15 years. Some studies came out that said that older predictions were off, and federal agencies started paying way more attention to the methods behind predictions.

I don't think it is at all an exaggeration to say that the entire line (Columbia Pike & Crystal City/Potomac Yard) could have 37,000 boardings a day 15 years after it opens. The Minneapolis light rail is about a decade into its existence and gets 30-33K trips per day.

by MLD on May 14, 2014 11:49 am • linkreport

I hate to say it, but people calling for light rail or heavy rail here are doing damage to the project overall. It just gives ammo to opponents who are fishing for any reason to be against it.

There's already lots of misinformation about streetcars vs. buses and its not helped when people think that the project should be some sort of other rail project.

1. It's too late in the game for it to be seriously considered.
2. And there are good reasons why those options were rejected in the first place.
- If the only difference between a streetcar and light rail is the dedicated lanes then we really have nothing more to say about that. VDOT won't allow it. It's the world we live in.
- Metro is far too expensive, complex, and requires way too much buy in from other jurisdictions. Moreover, a streetcar here doesn't preclude heavy rail in the (distant) future.

by drumz on May 14, 2014 11:52 am • linkreport

Also, I believe the 2020 forecast for the Minneapolis light rail was something like 25,000 trips so it is above that already.

by MLD on May 14, 2014 11:53 am • linkreport

I'll agree with the others here- given those numbers, it really should be light rail with a dedicated right of way. There's the space to do it, there's the demand to do it. But a streetcar is a far sight better than trying to use busses, that's for sure, and if there was a need, it probably wouldn't be too hard to close off the most of the lane in the future to create a hybrid where it's got a mostly dedicated right of way with limited number of crossings at intersections and such.

Anyway, the next step is actually getting the damn thing built. Preferably without stupidly expensive stops...

by Zeus on May 14, 2014 11:55 am • linkreport

I love how the article makes no mention of the $100 million cost increase at all. This weak argument is almost akin to those put forward by Tea Partiers foaming at the mouth against Obamacare. It's based on a single (estimated), statistic and instead of countering critcisms, completely ignores them.

Ridership might be the qualifying factor to build transit in the first place, but it's only a small part of the picture. Efficiency, need, total cost, and cost/mile are just as important. At $585m, the streetcar project is ridiculously expensive.

Comparing a streetcar line to rapid transit (Metro), commuter rail (MARC), or even light rail (Purple/Red Lines) doesn't make any sense. For the high price you pay for those systems you get a ton of capacity, dedicated Right-of-Way, and much, much higher speeds. Streetcars are practically high-capacity buses on rails that run in traffic, have limited infrastructure, and operate relatively slowly. They definitely have their use in certain applications at certain costs, but this definitely isn't one of them. A Columbia Pike/Rte 1 BRT system makes far more sense.

I guess this type of thing isnt unprecedented though. Arlington has a history of careless and wasteful spending, and is well known to be the most spendthrift jurisdiction in the area. In addition to this half a billion $ white elephant, the county also wasted taxpayers' money on $1 million bus stops (half a million is still ridculous, but at least they will actually function), $100 " aquatic centers," and $3m+ a year on an art center with usage.

by K Street on May 14, 2014 12:00 pm • linkreport

Even if the pike streetcar exceeds projections, VRE will still beat in in Passenger miles handily.

Ridership is only one component of use, how far those passengers are transported is also important.

by Richard on May 14, 2014 12:04 pm • linkreport

The 585M figure includes the Crystal City Streetcar.

The bump from 250M to 310M and now to 358M is in part because the federal government wants larger contingency fees.

Your evidence for the county wasting money is thin as well. The county found that they would save money by doing the bus stop project themselves rather than relying on WMATA. The aquatic center is on hold because the construction bids (from private contractors) were high and the county didn't want to commit to that.

Meanwhile the Artisphere performs well and suffers from claims that it would be profitable rather than any specific waste.

by drumz on May 14, 2014 12:08 pm • linkreport


Notice, Cavan, that those are all "light rail" and not streetcar. Seriously, what a tremendously wasted opportunity to not give this it's own right of way.

If the Commonwealth/VDOT was never going to allow it to happen, then there was no opportunity to be wasted.

Anyway, for a true streetcar/non-light-rail example, we can use the (somewhat disputed) New Orleans Loyola-UPT example, which may mirror what will happen with PikeRail:

Moreover, while the Loyola line has exceeded expectations when it comes to ridership, much of the increase is due to the fact that the Regional Transit Authority altered bus routes to force riders onto the streetcar, transit advocates say.

The RTA estimated in 2010 that the Loyola streetcar would draw 1,000 to 1,400 riders a day, and it’s already averaging about 1,700 a day.

But Rachel Heiligman of Ride New Orleans, a nonprofit public transit advocacy group, said that because the RTA cut off the downtown segments of the Freret and Martin Luther King buses at the Union Passenger Terminal, those bus riders now must transfer to the Loyola streetcar if they want to get to Canal Street.

“What we’re doing is really just shifting the ridership from one mode — the bus — to the streetcar,” Heiligman said.

by Dizzy on May 14, 2014 12:17 pm • linkreport

It fascinates me that Arlington's 'well known spendthrift ways' boil down to a bus stop that they're already working to address, an aquatic center that's already canceled, and an art center. That's it. So, basically, something on the order of $10M out of a $1B budget. Let's get some perspective--if that's the worst people can come up with, they're doing a pretty impressive job.

by Mike on May 14, 2014 12:17 pm • linkreport

Okay so light rail is a non-starter. What exactly are we getting for $350M? It sounds like we'll have a bunch of street cars (and infrastructure that requires maintainance) that will be stuck in the same awful traffic we have now. (At least on the Eastern side we will mostly have dedicated lanes.) As you can see by the numbers, a ton of people already ride the bus on Columbia Pike now. Transit use will only increase if it becomes more efficient. Without dedicated lanes, how will this become more efficient?

by movement on May 14, 2014 12:20 pm • linkreport


It's a bigger vehicle. Meaning more people on fewer total vehicles.

It's more investment over what a buses only option on the pike would provide. Which means more people overall using public transit on Columbia Pike. It means less bunching, it means better use of the stops.

That's the other key. There's a lot of griping about the costs but many are ignoring the benefits. It's worth spending more if more is returned.

Is it worth spending 80-100M on the enhanced bus option (going off what the last ROI study said) that will basically help things break even. Or go with the more expensive streetcar which is predicted to provide way more investment than enhanced bus.

by drumz on May 14, 2014 12:33 pm • linkreport

Subway systems such as New York (8,733,300), WMATA (855,300), Atlanta (221,200), and even Baltimore (48,500)

I think New York's daily ridership is closer to 5.5M (less on the weekends), WMATA's is about 760k (less on the weekends).

by Scoot on May 14, 2014 12:33 pm • linkreport

The cost increase can't be ignored.

$356m for a 4.9 mile streetcar is ~$75m per mile.

Next month, Minneapolis/St. Paul will open their newest light rail line - the Central Corridor between the two downtowns. It largely runs in dedicated lanes down the center of University Avenue. The total cost is $956m for an 11 mile line, or ~$87m per mile.

Point being: the difference in operational performance for that small increment in capital costs that gets you from 'streetcar' to 'light rail' can make a huge difference, and at relatively low cost.

VDOT's objection to losing any car lanes are well known, but that doesn't change the fact that mixed-traffic operation will put a rather large damper on the ROI for this project.

by Alex B. on May 14, 2014 12:40 pm • linkreport

What you're describing isn't an improvement. Buses are what 5% of the vehicles on CP? 2%? Even if you slash that number further, you're not solving the traffic problem.

Why aren't more people taking the bus? Is it lack of capacity or lack of efficiency? From what I've heard it is the latter. If the Streetcar doesn't solve this, then it isn't worth the investment. I don't care what the prognosticators say. They could easily be wrong.

by movement on May 14, 2014 12:43 pm • linkreport

Just for the sake of argument, let's say Arlington grows a pair, flips VDOT the bird, and breaks ground on dedicated lanes anyways.

What happens next?

by LowHeadways on May 14, 2014 12:56 pm • linkreport

I don't know if it is extraordinarily bad timing with this article and the revised budget Arlington published this morning, but the c pike street car is entirely out of hand.

Started at 161 million in 2007, then 250 million in 2011 ( increase of 55%) and now 385 million in 2014 ( another 54% increase!). The cost of this thing has soared 140% since the original proposal. Then Uncle Sam telling you "no" on the grant application because your cost estimates are so far off.

How anyone on the Arlington Board can discuss this in session without being embarrassed is beyond comprehension.

by Cpike on May 14, 2014 12:59 pm • linkreport

The number of people taking transit will increase as development on the Pike increases. Period. Neither side disputes this. Pretty soon you run out of room on the current bus system - they have the capacity of conventional buses, and adding buses beyond a certain point (which IIUC we are close to) is not possible as bus bunching becomes too big a problem. Ergo, capacity improvement is needed - via stops with off fare payment (the dreaded half million dollar stops) and larger vehicles. The alternative the anti street car folks come up with is articulated buses, with the SAME off fare payment, etc. Street cars get you somewhat more capacity than the articulated buses. They also will probably draw more transit mode share (for reasons from ride quality to hipster elitism, depending on your POV and agenda.) They will probably also goose development a bit more, either because of their permanence, or because they look nicer to elitist hipsters. The antis basically discount the added transit share and the development impact (other than the antis who oppose development) play down the capacity improvement, and suggest you get 99% of what you need with articulated buses. The pro's emphasize the development, mode share, and capacity gains, and also suggest articulated buses cost advantage shrinks when you look at life cycle costs rather than upfront capital.

Thats it. I am tired of rehashing this debate. We may from time to time refine the numbers, but what the advantages of each mode are has been reiterated 300 million times.

Whats new, now, is a revised ridership projection, and revised costs. But the latter isn't a substantive change, its different assumptions about contingency and timing (which effects inflation.) The bigger news, I think, is that County seems commited to getting federal funds, something some antis claimed would not happen.

by AWalkerInTheCity on May 14, 2014 1:21 pm • linkreport

"Then Uncle Sam telling you "no" on the grant application because your cost estimates are so far off."

One more time - the FTA said a higher contingency was needed, and that pushed the project out of the category for which ArlCo had applied (Small Starts) into New Starts. FTA strongly suggested that they apply for New Starts.

by AWalkerInTheCity on May 14, 2014 1:23 pm • linkreport

VDOT's original estimate for the cost of the Springfield Interchange (Mixing Bowl) was $350M. The final cost was $676M. The Wilson Bridge's original estimate was $1.6B and final cost was $2.4B.

Major infrastructure projects always go up in cost from the original estimate to the final baselined budget. There are unavoidable reasons (like inflation) and avoidable reasons (under-projecting, increases in scope) for that. It's likely that part of the reason original cost estimates are so far from the final estimate is that officials low-ball the original figure to garner support.

by Falls Church on May 14, 2014 1:27 pm • linkreport


You try to make the same argument with every cost increase. Reasonable people can agree that the metrics behind this project have changed significantly as the price has increased 140% over the past 7 years.

I get that you, among others here are a fan at any cost. That you would think this was a good idea whether it cost 160 million or 10 billion, but you are in the minority. Arlington County has burned whatever credibility it had by virtue if the fact that they obviously have no idea what is involved in a project line this.

Rest assured that these latest prices will have consequences in the forthcoming election in the County.

by Cpike on May 14, 2014 1:33 pm • linkreport


What you're describing isn't an improvement. Buses are what 5% of the vehicles on CP? 2%? Even if you slash that number further, you're not solving the traffic problem.

If more people choose to take a streetcar more than they choose to ride a bus (this happens, though debate rages as to why) then you're removing those private vehicles as well.

Why aren't more people taking the bus? Is it lack of capacity or lack of efficiency? From what I've heard it is the latter. If the Streetcar doesn't solve this, then it isn't worth the investment. I don't care what the prognosticators say. They could easily be wrong.

This is all examined and answered in both the latest ROI study and the Alternatives Analysis. This is what informs my view that the streetcar is the best choice and answers most of the questions that people bring up.

by drumz on May 14, 2014 1:41 pm • linkreport

Falls Church,

Yes, costs go up. The mixing bowl went up 90%, the bridge went up 50% and that is after construction is done. This streetcar has already gone up 140% and it's years from breaking ground.

Also, the mixing bowl handles 430,000 vehicles per day and the bridge 230,000 vehicles per day which are millions of people per day. The proforma performance between those two projects and this streetcar are time zones apart.

by Cpike on May 14, 2014 1:48 pm • linkreport

It's not about metal wheels vs. rubber tires, it's about creating a service that attracts ridership. A streetcar might be good at attracting riderhsip, but the question I haven't heard a good response to yet is: Is there a way to attract as many riders with improvements that can be built quicker and for less money? I don't really know, but it seems that this article is playing up differences between bus and rail that are not really intrinsic to either. Saying that a bus system could never move 60,000 people per say is simply wrong, and there are tons of existing examples to disprove that.

I'd rather the debate focus on the actual merits of several option to determine how to produce the best results in the shortest timeframe with the least amount of money. Until the conversation moves in that direction, I have to say that I'm pretty turned off by the rail folks.

Jarett Walker breaks down the actual differences between bus and rail here:

by TransitSnob on May 14, 2014 1:58 pm • linkreport

To piggyback on FC's point, $585 million is not a lot of money! For transportation, at least.

The new 11st bridge was 600 million. One small bridge. New highway interchanges are typically in the 600 million range. Just for an interchange!

Build the rail as is now, and have the discussion about transit only lanes later.

by h st ll on May 14, 2014 1:59 pm • linkreport


You will note well that no where in my comment did I say if the project should be pursued or not. I merely pointed out some facts about the justification (presenting the POV of both sides) and clarified some issues with the cost. I do not appreciate you charecerizing my POV or motivation like that. As for supporting it were it 10 billion, that is indeed odd, as I have often stated that the claimed alternative of a heavy rail line is not feasible due to a cost in that ballpark.

I was not involved in this discussion when the cost estimate was below 200 million. So to me the debate is about the higher estimate from 250 million, and I have explained the drivers of that increase. The rationale for that has not in fact changed, as you can see if you read the slides that present the change. It is what it is.

Again, while the higher contingency and inflation costs may sway some fence sitters to oppose, the clear path to seeking federal funds, which will mean a much smaller ArlCo contribution than some feared, may sway others to support. I don't know. IF it does not pass now, it will be up to the antis to implement the articulated bus option. Between discontent among street car supporters, and opposition from the large segment of street car opponents who simply do not want any large investment in transit on the Pike, I do not think that will happen. And of course FFX county will keep pushing for it. It will be revived again in a few years, hopefully in a county with a stronger office market, and with the swimming pool and dog park fights forgotten.

by AWalkerInTheCity on May 14, 2014 2:03 pm • linkreport


Note that JWalker says this

"Capacity. Where demand is high, rail can serve that demand at a higher ratio of passengers to on-board staff, which means that once you absorb the (often large) construction cost, you will be able to offer greater capacity for a given operating cost. A transit vehicle that’s too crowded to board doesn’t meet any of our seven desires for useful service, so this point is often decisive in favor of rail"

He has explicitly said that his oft noted street car skepticism does NOT necessarily apply to corridors reaching volume capacity. Which many proposed street cars are not, but Columbia Pike IS.

by AWalkerInTheCity on May 14, 2014 2:05 pm • linkreport

WADR, the LA report is garbage. As someone knee-deep in proposal development, the "motherhood and apple pie" nonsense would never get out of pink team review, much less get approved for publication. The report makes a number of outrageous claims without suggesting how or why they could possibly be true. There is only one rational explanation for not publishing that information - they don't know.

by movement on May 14, 2014 2:09 pm • linkreport

I don't know what LA is supposed to mean.

If you're talking about the Alternatives Analysis you're going to have to be specific about what's supposedly garbage. That report was required by the federal government. And by and large, they agreed with it, rejecting the project on a technical consideration (that's been explained before).

Besides, it's usually where you can find the answer to most questions about the streetcar.

by drumz on May 14, 2014 2:18 pm • linkreport

Build the rail as is now, and have the discussion about transit only lanes later.

This idea has merits but it shouldn't be relied upon when stacked streetcars become a problem.

by Fitz on May 14, 2014 2:33 pm • linkreport

Richard -- Ideally, you use appropriate metrics particular to each mode. Route-ridership miles matters far less to streetcar, which by definition serves shorter trips.

More importantly, VRE service and transit service on Columbia Pike are both important to Arlington, but measured differently, and have differential impact.

AWITC - word.

TransitSnob -- the bus thing is tough. Basically, in areas where you have a lot of transit dependence or hyper density, you can get ridership numbers for buses that are quite high, because of limited alternatives. (E.g., one of the biggest "snow jobs" about BRT is that proponents in the US tout South American results, when in SA people are much more bus-dependent and willing to withstand passenger densities that are much greater than in the US.)

I think that 20-25,000 passengers is about the highest we can expect on the highest use bus lines in the DC area. Most of those lines are in DC, with the exception of Columbia Pike.

If you figure that most of the market is captured for the transit dependent and you think for other reasons that you can capture trips from the non-transit dependent (choice riders) and you want to position your community more highly within the metropolitan landscape as a choice place to live, work, locate business etc. vis-a-vis others, than investing in fixed rail transit makes a lot of sense.

I guess the only way we could say for certain that the kinds of bus improvements that I suggest we should do ( would actually result in much higher ridership would be to do a real test.

The way BRT is positioned in the US isn't really for these kinds of existing corridors of medium range that already have decent ridership, say 15,000 to 20,000 like Georgia Ave., 16th St., Pennsylvania Ave., H Street, Florida Avenue.

by Richard Layman on May 14, 2014 2:34 pm • linkreport

I meant to say LPA=Locally Preferred Alternative.
The LPA contains insufficient information to perform a realistic cost-benefit analysis. Specifically,
* the goals and objectives are completely immeasurable
* the benefits are described anecdotally, not scientifically
* while the relative benefits from one mode to another are assessed, no scale information is provided (how much better is A than B? or B than C?)

This project never should have been green-lighted.

by movement on May 14, 2014 2:35 pm • linkreport

Agaan, I question the metrics on Columbia Pike. There is no question that there are a lot of bus users there. But they aren't neccessarily going to where the streetcar will run -- up and down the Pike.

And in the future, I don't see Pentagon City/Crystal City or Skyline becoming larger employment centers. BRAC has permanently hollowed out those jobs from Arlington, and they are not as attractive for private tenants as alternative locations.

by charlie on May 14, 2014 2:42 pm • linkreport

>Just for the sake of argument, let's say Arlington grows a pair, flips VDOT the bird, and breaks ground on dedicated lanes anyways. What happens next?

First, at a bare minimum, the agreement between Arlington and VDOT that turned over ownership of Columbia Pike to Arlington from VDOT in exchange for not converting lanes is revoked, and the road returns to VDOT ownership. Once that happens, construction ceases immediately and the project is dead.

But that's not the only result.

Second, very likely, since Arlington can no longer be trusted to honor agreements, VDOT refuses to sign any new ones. This effectively grinds Arlington's entire transportation program to a halt, since almost everything they do requires VDOT sign-off. Say goodbye to street improvements, bus stop easements, Metro station entrance expansions, everything else they're doing.

Third, probable but not definite, the state withholds future funding from Arlington. If they're nice and only withhold a couple of years of VDOT funding then maybe Arlington only loses a few tens of millions. If they're spiteful then it might extend beyond VDOT or for a longer period, and cost Arlington hundreds of millions.

Finally, very unlikely but not out of the realm of the possible, the state could revoke Arlington's charter to operate as an independent locality. Or if that's a step too far, the state could revoke any number of self-rule privileges Arlington has, such as zoning, or budget-making, or authority to tax property.

Virginia is a Dillon Rule state, which basically means the state is god and localities have no intrinsic right to do anything, unless the state gives them permission.

In short, ignoring the state would be disastrous.

by BeyondDC on May 14, 2014 2:47 pm • linkreport


True that the Streetcar's estimate has increased 140% while projects like the mixing bowl "only" increased 90%. However, VDOT (and the contractors they use) have built lots of interchanges, so you'd expect it to be easier to estimate cost. Streetcars are a newer thing (in modern times) and hence a little harder to estimate.

At any rate, I'd be less concerned with how much the project's cost has increased and more concerned with whether the current (and eventual final) price is less than the threshold where it becomes a bad investment.

Also, the financing mechanism makes a big difference. Part of the price of the streetcar is being paid for by the expected increased property values along the Pike (what makes this possible is that streetcars increase property values a lot more than buses). That makes the project higher returning for everyone else not paying that special tax on higher property values. For the people who are paying the special tax, they're making tons of money, so they're better off too.

by Falls Church on May 14, 2014 2:54 pm • linkreport


The LPA is just a summary of the analysis done in the Alternatives Analysis. If you want more analysis then go to the main Alternatives Analysis which has 20+ chapters.

You're third point was also why the county commissioned the latest ROI study that specifically included the enhanced bus option.

by drumz on May 14, 2014 3:00 pm • linkreport


The argument has been made, based AFAICT on an off hand remark by Tejada, that the BoS would NOT seek federal funding - ergo, the amount for the street car would all come from the special tax fund, meaning there would be nothing left for things like road paving, ergo those things would be paid from the general fund, ergo TAXES, ergo LESS MONEY FOR SCHOOLS.

Thats why I see the big news as being the fact that they seem focused on getting New Starts funding for PikeRail.

by AWalkerInTheCity on May 14, 2014 3:02 pm • linkreport

Richard -- Ideally, you use appropriate metrics particular to each mode. Route-ridership miles matters far less to streetcar, which by definition serves shorter trips.
More importantly, VRE service and transit service on Columbia Pike are both important to Arlington, but measured differently, and have differential impact.

I agree it is important, but as this article does compare VRE to the projections of the Pike rail, I wanted to point out that it was comparing apples to oranges.

by Richard on May 14, 2014 3:14 pm • linkreport

430k cars a day does not equal millions of people. With HOV-3 the numbers are probably higher but the average number of people in a vehicle is 1.1 - 1.2.

by NikolasM on May 14, 2014 3:31 pm • linkreport

Clearly no one relevant has read this thing.
p. 6-20
Streetcar Build Alternative
The travel time and travel cost savings associated with the Streetcar Build Alternative would be greater than the savings associated with the TSM Alternatives. The Streetcar Build Alternative attracts 1,300,200 new riders annually compared to the No Build Alternative and generates $5,180,148 in annual travel time savings and $915,525 in annual travel cost savings for 2030. These cost savings support livability in the corridor and would be consistent with each county’s efforts to reduce households’ need to own cars.

That's it, no references, no sources, no calculations, and no explanations. This is supposed to be supporting documentation. It is garbage.

by movement on May 14, 2014 3:57 pm • linkreport

The "entire Richmond bus network" is nothing to boast of. Almost no service outside of downtown.

by Willow on May 14, 2014 4:00 pm • linkreport

I would be remiss not to point out in this discussion that the Portland Streetcar carried 15,986 passengers at the end of the 3rd Quarter, 2014. As the first project to introduce modern streetcars in the U.S, the Portland Streetcar opened in July, 2001, and initially carried 4,982 weekday passengers. Ridership has now tripled to 15,986 weekday passengers, a statistic made even more impressive by the fact that a downtown fare free zone (which encompassed a significant portion the streetcar route) was abolished in September, 2012. A further extension across a transit only bridge in 2015 should result in another increase in ridership. Development along the streetcar route has been phenomenal, the latest surge owing to an extension into East Portland in 2012.

by Glen on May 14, 2014 4:03 pm • linkreport


whats even more impressive, is that some have said the ridership gains of street cars is due to novelty, and will fade. So Portland contradicts that.

by AWalkerInTheCity on May 14, 2014 4:05 pm • linkreport


Uh, the sources are at the bottom of each page. And that's a final summary after several pages outlining the methodology and data that takes up 28 pages just for that chapter.

by drumz on May 14, 2014 4:11 pm • linkreport

slight edit: what you quoted is a final summary of the analysis of the data that's presented.

by drumz on May 14, 2014 4:14 pm • linkreport

Falls Church,

"Also, the financing mechanism makes a big difference. Part of the price of the streetcar is being paid for by the expected increased property values along the Pike (what makes this possible is that streetcars increase property values a lot more than buses)."

This is the part that makes me shake my head in amazement. Proponents can't have both sides of the equation because if Arlington is woefully underestimating costs, it is also likely greatly overestimating the revenue side to make the project more palatable to the public.

You say cost estimates are off for this project (more so than in relation to road projects) because so few of them are built, getting good numbers is difficult.

Well, ok...then doesn't it also make logical sense that the County's pie in the sky, wholly speculative numbers on revenue, and increased development are also off similarly, especially since increased development and property taxes are highly subjective and subject to largely uncontrollable economic forces to begin with?

Arlington can't be so awful at the nuts and bolts part of the streetcar (infrastructure, design, construction), and experts at the revenue side at the same time. It defies reason.

If Arlington can't get the pre-construction price to within 140% (and lets be honest here, we are years before groundbreaking, this thing will continue to suffer "surprise" increases), then how can the voting public have any faith that the revenue numbers are in the realm of legit.

You say the cost of the thing doesn't matter unless it passes the threshold of being a bad investment. Well, what if the increased development and valuation numbers are also off by 140%? Does it then become a bad investment?

by Tom on May 14, 2014 5:00 pm • linkreport

The costs went up from 250 million (the earlier numbers were not based on even rough engineering, IIUC) because they are piling on contingencies, not because they made errors in the nuts and bolts of design.

by AWalkerInTheCity on May 14, 2014 5:23 pm • linkreport

Richard Layman thanks for linking that article.
"Night Owl bus service"!!

by asffa on May 14, 2014 5:28 pm • linkreport

How many billions have been spent to help North Arlington and Crystal City riders access the Metro? And look how it's helped North Arlington and Crystal City wealth explode over the decades. The Street Car option is a HUGE bargain by comparison.

Please, don't kid yourself that BRT is any kind of option for Columbia Pike. That's a joke. BRT works when there is a dedicated lane -- you know, the whole "RAPID" part. There is no possibility of adding a dedicated lane on Columbia Pike. All you're offering are bigger buses.

But, never mind those of us who can't afford the million dollar mansions to live on the right side of the Route 50 tracks. Just keep on spending your hundreds of millions on North Arlington Metro access and fancy dog parks. Maybe just tack on a bit extra for free wine and cheese. We South Arlingtonians will just keep our mouths shut and be grateful you even allow us to call ourselves Arlingtonians.

by Sam on May 14, 2014 5:55 pm • linkreport


Are you saying the entire 54% increase is because of the required Fed contingency? Where is the breakdown of the cost increase you seem to be referencing?

by Huh on May 14, 2014 6:28 pm • linkreport

I found it, 34 million of the additional 108 million is additional contingency.

The remaining 74 million, or 68% is from
an additional vehicle, longer vehicles, additional maintenance facility equipment, higher unit costs, higher engineering and start-up costs.

Basically, everything that a bottom barely competent project manager accounts for, especially after having already increased the price from 160 to 250 million, or 56%.

So yes Awalker, this increase IS because of deficiencies in the "nuts and bolts" part, and only marginally because of contingency.

by Huh on May 14, 2014 8:42 pm • linkreport

Build the rail as is now, and have the discussion about transit only lanes later.

This idea has merits but it shouldn't be relied upon when stacked streetcars become a problem.

I don't think it should be relied upon, but for those demanding a dedicated right-of-way it should enter their consideration.

What does one think the odds are that VDOT will relent and allow the streetcar to run in a dedicated ROW if Arlington keeps asking? I'd say single digits at best.

What does one think the odds are that VDOT will relent and allow the streetcar to run in a dedicated ROW if Arlington builds a streetcar, it gets bogged down in traffic and an analysis can be done that shows that relenting would increase throughput for the corridor? Not 100% for sure, but not 0% either and almost surely higher than in the other scenario.

IMHO, the best way to get a dedicated ROW is to build the thing. That's not a sure thing, but it is probably the best path.

by David C on May 14, 2014 9:09 pm • linkreport

As to the new cost, in addition to the higher contingency (which isn't the same as spending) and inflation (again in real dollars, that isn't an increase) there is the fact that the system being considered now isn't the same as the one last considered. The size of the streetcar has been changed, for example.

What's unfortunate is that they didn't further explain what caused the "higher engineering and start-up costs," or why they proposed a change in the size of the streetcars.

Still, this price doesn't even really represent an increase. The $310 million price was a “most likely” cost. But the FTA determined the price range as between $255 million to $410 million. And the current price is in that range. So let's tone the freak out down a bit.

by David C on May 14, 2014 9:18 pm • linkreport

Analysis paralysis. We've only been debating the streetcar for 15 years. What the heck. Let's keep debating it for another 20 and by that time it will cost $500 million.

Folks, fish or cut bait. But, I will say this, if we don't do this for South Arlington, this is one lifelong South Arlington voter who will never vote for another bond to help the millionaires up north. Screw that.

by Sam on May 14, 2014 9:41 pm • linkreport


The revenue and property value increase side is a lot easier to estimate. There are developers waiting in the wings to build further density but that's not possible unless Arlington allows it, and they won't allow it unless there's additional transportation capacity. The streetcar creates that additional capacity.

That said, we could enter an economic recession in the DC area that saps demand for space. That's always possible and in that case the developers will back out and revenue projections won't be met. That's a risk every investment faces whether made by the private or public sector.

Btw, I don't buy that the increase in contingency isn't a real increase in cost. Contingency reserves almost always get exhausted in major infrastructure projects, and then some. Usually they end up at least a little over the baselined budget. The contingency was likely underestimated initially to increase support for the project and establish momentum.

by Falls Church on May 14, 2014 9:49 pm • linkreport

"Arlington's streetcars will carry more riders than VRE, or the entire Richmond bus network combined"

Sound to me like LowHeadways and I were think the same thing.

Those kind of numbers could justify utilizing the Pentagon junction provision and building the subway line to Skyline and beyond to the Mark Center instead of building the streetcar line.

I happen to believe that a streetcar line along Columbia will put any proposals to put a subway line in the same corridor further into the future if at all.

by Sand Box John on May 14, 2014 9:54 pm • linkreport


As an example of what I'm talking about, take the Potoma Yard infill station. The cost of that station has increased nearly 80% since original estimates. They were totally wrong about costs. But, the millions of square feet of new development is still on track, much of which has already broken ground.

The NoMa infill station similarly went way past original cost estimates but the development/property value increase has surpassed estimates by far more.

Rail projects only get built because developers have an interest in building there in the first place.

by Falls Church on May 14, 2014 10:06 pm • linkreport

I'm all for a subway line running either up Columbia Pike or Rt 7, but if a streetcar price tag of a few hundred million makes people blanch, try quintupling that cost. Ain't gonna happen.

In these sorts of debates with big capital investments, there are always a million arguments against making progress. It's easy to say no. The hard part is to move forward with a plan that will work.

Digressing a bit, but take the debate over the 2009 stimulus bill. Every honest economist knows that should have been at least twice as big, and probably three-times as big. To start with, we should have done A LOT more to keep people in their homes. Instead, the TEA Baggers took over the Republican Party, and what did we get -- Quantitative Easing, which is the Fed's last resort and is a very inefficient way to keep the economy from collapsing.

Fiscal stimulus would have been a lot better, more efficient, more focused, more favorable for ordinary, hard-working, middle-class Americans. But there were too many idiots out there and the politics became hard and we squandered good policy to placate those idiots.

At a much smaller scale, that's what's happening here. People object to the streetcar. I don't know -- maybe it's too European. Maybe it interferes too much with people who love to drive cars. Maybe some people just think it will cost too much and are missing the forest for the trees.

Whatever the excuse, we either do this thing or we go with some jacked up, ineffective so-called enhanced bus solution that won't solve any real problem or accomplish anything meaningful for the county, but will definitely save money and allow the North Arlingtonians to afford their next $5 million dog park.

by Sam on May 14, 2014 10:11 pm • linkreport

Noma went way past? Final price was 21% over estimated price (19 million overage), and there was more than 2 million SF of new office in the entitlement pipeline in 1998, a year before any formal cost sharing agreement was in place with landowners to kick in 25 million, and 3 years before they ever broke ground. NOMA was well on its way without the Gallaudet stop.

Similarly, the existing retail at Potomac yard has been there since 1998. Pulte bought and rezoned adjacent Potomac yard landbays in early 2004 (I managed the entitlements for the Pulte Centex JV at the time), yet no Metro stop. The City of Alexandria didn't even have the appropriate zoning in place to accommodate the future metro stop until 2010, so again, crediting Metro with the development at Potomac Yard when there wasn't even a commitment to build a station there until 2010 is simply not true.

by Dakar on May 14, 2014 11:03 pm • linkreport

I'm all for a subway line running either up Columbia Pike or Rt 7, but if a streetcar price tag of a few hundred million makes people blanch, try quintupling that cost. Ain't gonna happen.

I am two, but factor 10x is more likely

by Richard B on May 15, 2014 8:02 am • linkreport

One consideration that has not entered the debate at all is the effect of driverless cars on future transit needs. Given the current state of the technology, it seems almost inevitable that driverless cars will be ubiquitous by 2035. This may lead to a shift in mass transit from larger vehicles (e.g., streetcars and buses) to smaller vehicles (e.g., vans). Most of the cost of a taxi today is labor, so taxis will likely be quite cheap in the future, especially if shared. I doubt many will wait for the streetcar with inexpensive shared ride vehicles that can take them much more directly where they want to go.

by Michael on May 15, 2014 10:58 am • linkreport


Maybe, maybe not. If it happens great, I don't see how we should suspend planning until that happens (especially since there is no guarantee that it will).

by drumz on May 15, 2014 11:02 am • linkreport

I don't think there is any information out there that indicates that "driverless" cars are going to be so prevalent and revolutionary that we need to start changing everything about how we plan our cities.

by MLD on May 15, 2014 11:17 am • linkreport


Ok, I guess 21% over-budget isn't "way past". Fine.

For Potomac Yard, in anticipation of metro construction, the area was upzoned to allow additional 1.4M sqft. During station construction, they're allowing an additional 3.4M sqft and once construction is complete, they're allowing an additional 2.7M. While the desire to build all that space would be there regardless of whether the station was built, Alexandria would have never permitted it without a plan (even an early, non-committed plan) to build additional transportation capacity.

Similar thing with NoMa. DC isn't going to allow for millions of square feet of new development without some kind of transportation plan (even if the plan isn't yet approved or funded at the time initial allowances are made). As early as 1996, DC was proposing a huge makeover to the New York Avenue corridor that included the eventual metro station, a traffic tunnel, and a reconfiguration of New York Ave. Some of those things happened, some didn't but DC wasn't going to upzone NoMa without addressing transportation in some way (whether rail or roads).

by Falls Church on May 15, 2014 11:28 am • linkreport

Pulte bought and rezoned adjacent Potomac yard landbays in early 2004 (I managed the entitlements for the Pulte Centex JV at the time), yet no Metro stop.

Additional transportation capacity, including a metro station has been in the master plan since 1987. That's what allows the upzoning. Pulte doesn't determine zoning. The government does.

Starting in 1987, about the same time that the City began updating the 1974 Master Plan for the Potomac Yard/Potomac Greens portion of the site, RF&P Railroad began to explore development alternatives for the Yard because it was no longer being used for the classification of trains. The first proposal for the Yard, Alexandria 2020, was a mixed-use, neighborhood development which continued the street grid of the adjacent neighborhoods and replicated typical setbacks, heights, and architectural styles. The plan included a tree-lined interior boulevard, parks, and interesting pedestrian gathering places. The plan included a Metro station near the center of the Yard with the potential for commuter rail service and bus connections. The densities proposed in the 2020 plan were much higher than the currently adopted plan.

The 2020 plan was never formally submitted to the City for approval, but the City did approve new zoning for the site in the context of the Master Plan update in 1992. The new zoning, CDD, provided for a lower overall density of the development than was proposed with Alexandria 2020.

by Falls Church on May 15, 2014 11:38 am • linkreport

Dakar -- re your comments about NoMA, discussions about the likelihood of the stop being constructed predated 1998 and were relied upon by developers in making decisions about building there. Even though it's close to Union Station, probably, especially then (You have to remember what it was like then, e.g., I got mugged-assaulted a few times on K St. between 1996 and 2001 before I finally adjusted my night route home) it's pretty likely that the development we see there today wouldn't have occurred, at least not for a long long time, maybe starting around now. Same goes with "Union Market", nothing would be happening there without the Metro station.

The reason that the cost was higher is that a switch needed to be moved-replaced in order to construct the station in that spot.

Having lived in that area from 1987 to 2005, it's pretty clear that the development in the "H Street corridor" is pretty much reliant on the creation of that station, which changed the willingness of people "with choices" to live north of H St. Just wrt housing values of existing properties in the section of "H St." between H St., FLorida Ave., 2nd St., and 14th St., the value increase of a minimum of $200,000 per house (it's higher now) resulted in a $340 million increase in property value. That's only about 1/6 of the catchment area of the station. Union Market is another area, as is NoMA more generally, and the industrially zoned property along 2nd St. on the east side, which is being rebuilt mostly as multiunit housing.

Now H St. ec. dev. energy comes from the forthcoming streetcar, but again, ec. revitalization of the extant section of the neighborhood was stoked by the new Metro station.

by Richard Layman on May 15, 2014 12:01 pm • linkreport

Or you consider skyline, which was originally upzoned in anticipation of the never built columbia pike metro. Or we can just say that the infratstructure is just several decades late I guess.

by drumz on May 15, 2014 12:04 pm • linkreport

"I'm all for a subway line running either up Columbia Pike or Rt 7, but if a streetcar price tag of a few hundred million makes people blanch, try quintupling that cost."

quintupling? generously allowing 400 million for pike rail, quintupling would be 2 billion. Seeing as Silver Line phase 1 is 3 billion and is mostly in highway median, I dont think building a new underground line of that length for 2 billion is close to feasible. More like 10 billion, I imagine to get all the way to Mark Center.

Really not possible without much higher densities than anyone currently envisions. If anything Pike Rail, by bring development forward, will make a future heavy rail line more likely. but either way it wont come before 2050, and then only if the region continues gangbuster growth.

by AWalkerInTheCity on May 15, 2014 12:15 pm • linkreport

They should call it "stander-ship," because on a streetcar, most people will be standing. Enhanced bus service has a much greater percentage of sitting vs. standing room, and as any regular metro bus rider knows, sitting is not only much more comfortable, it's also a lot safer. Anyone who has been standing on a bus that had to brake suddenly can attest to that. #goseethedoctor.

by Busman on May 15, 2014 12:36 pm • linkreport

A: Saying the streetcar is inferior because more people (might) have to stand is one of the more bizarre criticisms I see. Especially since it has nothing to do with why we're going with a streetcar in the first place which is to help accomodate growth on what's already a very busy corridor.

B: A streetcar accelerates and decelerates more smoothly than a bus anyway. Rendering the fear of people falling over each other moot.

by drumz on May 15, 2014 12:41 pm • linkreport

Standing isn't 'riding'? Who would have known?!

The "no seats" thing is getting old. It may be safer, but being in a transit accident is rare. I don't have the exact numbers but I would guess your walk to and from the stop is more dangerous than your trip on the vehicle.

The research shows that streetcars attract more riders than even enhanced bus service. So clearly there are other factors that outweigh the desire for a seat.

by MLD on May 15, 2014 12:42 pm • linkreport

Personally, I find sitting in a bus less comfortable than standing in a modern streetcar.

by Neil Flanagan on May 15, 2014 12:54 pm • linkreport

drumz, your skyline example reminds me of another example I meant to include, Prince George's Plaza. The developer of the office complex there understood in the early 1960s that what became the Green Line was going to be built. It's still an outlier, reliant on federal leases, but it is an example of building in advance, based on future planning.

The difference between Skyline and PG Plaza is that the line serving that area of PG County was always planned as part of the base system while the Skyline area was part of future possible but indefinite expansion-extension.

by Richard Layman on May 15, 2014 1:54 pm • linkreport

This is another absurdly wasteful scheme which will serve as a roadblock to every single person in NOVA that doesn't live right on this corridor. Arlington residents and the region would be better served finding unused space (off road) for additional (bike etc) transportation modes instead of taking road space that serves greater need/population to serve a smaller population that will greatly underutilize it.

by tocker on Jul 4, 2014 6:43 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