Trade-offs
- Posted by ryan on March 25th, 2008 filed in Cities
Let me first congratulate BeyondDC on putting together a really exciting transit vision for the nation’s capital. His long-range transit plan, like those at Greater Greater Washington, helps to get us thinking about what is possible and achievable. Also, they’re really fun to look at. At some point in the near future, I’ll have more to say about my particular transit vision for the area.
I want to comment now on one passage from BDC’s vision page. He writes:
Any attempt to cost this system out to more than a ballpark figure would be highly speculatory on our part, but to get a general sense of the costs involved we can compare the known figures for the Tysons Corner / Dulles (Silver line) MetroRail and the Columbia Pike streetcar proposals.
The Virginia Department of Rail and Public Transportation estimates that Phase 1 of the Silver line (the portion from East Falls Church to Wiehle Ave) will cost $2.1 billion for 11.6 miles and 5 stations. Phase 2 is estimated to cost $3 billion for 11.5 miles and 6 stations. Averaged out that corresponds to a Phase 2 cost per mile of $261 million, or $223 million per mile for the entire Silver line.
According to Arlington County, the Columbia Pike streetcar is estimated to cost $140 million for 5.2 miles and several stations. That is, the entire line can be built for approximately half the money it will take to build a single mile of Silver line. Per mile, the streetcar will cost $27 million – approximately 10% the cost of a mile on Silver line Phase 2.
Broken down per anticipated rider:
Silver line:
$5.1 billion divided by 91,200 anticipated daily riders = $55,921 per rider.Columbia Pike:
$140 million divided by 21,000 anticipated daily riders = $6,666 per rider.The streetcar is literally eight times as fiscally efficient.
If Virginia were to build the truncated Tysons line as we’ve proposed, ending at Wolf Trap rather than Loudoun County, that would free up enough money to build about 120 miles of streetcar. If we can get ridership on all 120 miles that’s similar to what we’re getting on Columbia Pike (21,000 riders / 5.2 miles = ~ 4,000 riders per mile = 480,000 riders for 120 miles), that’s many times more efficient at producing transit riders than the Silver line. Even if we don’t get so much as half Columbia Pike efficiency on other streetcar routes, we’d still be doing far better than the western segment of the Silver line.
It’s numbers like these that have BeyondDC convinced that more streetcars is a better investment than more MetroRail.
I’m not a big fan of this presentation of the heavy rail versus light rail decision for several reasons. First, we shouldn’t talk about trade-offs between light and heavy rail at all without also discussing value per dollar spent on highways. It’s worth noting that more streetcars and more MetroRail is a better investment than more pavement.
Second, and I know this analysis is limited by available data, but I don’t think it’s right to assume that we can readily substitute one transit mode for another on any given corridor. Different modes are right for different circumstances. MetroRail is best suited to very high capacity short routes as well as medium distance routes, between dense neighborhoods, where speed is important. I think that building a streetcar line between Tysons and downtown would be cheaper but not cost effective, because the drop in service speeds would defeat much of the purpose of the extension–the creation of a rapid transit connection between Tysons and other dense neighborhoods, including downtown.
Finally, I think that transit advocates need to get in the habit of explaining to people how there is more to these choices than just cost per person per mile. The density and capacity that can be supported by a Metro station significantly increases the value of surrounding property, above and beyond increases delivered by streetcar lines. That should be taken into account. In BDC’s example, for instance, one might consider the expected value of new development planned along the new Metro stations compared to the expected value of new development along Columbia Pike. In both cases, the planned increases are impressive, but the gains along the Tysons line are an order of magnitude greater than those along the streetcar line.
A new heavy rail subway line through downtown might not look particularly cost-effective using BDC’s numbers, but it would significantly increase the robustness of the Metro system, providing redundancy, enabling more effective maintenance schedules and fewer delays, and generally increasing the value of existing Metro lines. These things need to be taken into account.
And we should also consider factors like the relative effect of Metro versus streetcars on congestion, pollution, overall neighborhood walkability and lots of other stuff.
But the bottom line is this: transit should come in a range of shapes, sizes, and speeds, and we should be committed to providing the right mode in the right situation. The big gains come from providing a credible alternative to highways, and we need to be establishing the arguments for that shift, not negotiating with ourselves about where we can afford to reduce planned transit capacity from heavy to light rail. Our long-term interests aren’t served by selling leaders on a transit-on-the-cheap world. The goal is to get people out of their cars. In many cases, MetroRail will do that more effectively, and we need to keep that in mind.
March 25th, 2008 at 2:12 pm
Great critique. I’m all for BDC’s plan, but I think it does sell Metro short, and ought to include at least the ‘New Blue Line’ expansion proposal through downtown, as that project would do wonders for increasing operating redundancy, service area, and offering a huge increase in capacity for both the new line as well as all of the lines that feed into the previous bottleneck.
March 25th, 2008 at 2:33 pm
Fitting the mode to the corridor and building what makes the most sense is the whole point. I presented things the way I did because
A) I think there’s a resistance in this area to embrace any mode that isn’t Metro.
B) I think it’s not just more fiscally efficient, but flat out better to get this sort of system coverage with streetcars rather than a couple of new Metro lines. Alex says I should have included the new blue line. OK. But part of the point is that if I did that, I’d have to cut out several of the streetcar lines (that’s what makes it a “vision” instead of a “fantasy” - in a fantasy I’d include everything). Nevermind cost. At this point, with our trunk Metro lines already in place, I’d *much* rather have a web of streetcars going everywhere than one new crosstown subway.
It wasn’t my intention to suggest that we should build a transit system on the cheap. If that’s how the text comes across I may want to rewrite portions. But I think it’s missing the point a little to focus on the money. The trade-off isn’t money. It’s coverage versus trunk service.
March 25th, 2008 at 2:46 pm
I just think you’re framing the trade-offs in the wrong way for our purposes. You’re saying that every Metro line costs us this many streetcar lines, and that simply need not be the case.
Transportation departments assume that the road network will grow ever larger. We ought to be questioning that assumption, saying look, under business as usual we can afford to build a new Metro line or a streetcar network. If we rethink our priorities and stop building every massive road project that comes along the pike, we can do both. And we should do both. Better trunk service increases the value of feeder lines, and better feeder lines increase the value of the trunk service.
If we assume that new transit construction is essentially going to involve making tough decisions about what to do with a very limited pile of transit money while highways continue to get the bulk of the dough, then this is all kind of silly. We’re conceding that transit will never play a primary role in our transportation system.
March 25th, 2008 at 2:50 pm
More…
one might consider the expected value of new development… the gains along the Tysons line are an order of magnitude greater than those along the streetcar line.
Comparing a single Metro line to a single streetcar line misses the point that we get more than one streetcar line for the money of one Metro line. If we\’re to make such a comparison, we\’ll need to compare the development on the one Metro line to, say, 100 miles worth of streetcar. I don\’t think it\’s clear at all that Metro would come out ahead. In fact I suspect very seriously that it wouldn\’t.
A new heavy rail subway line through downtown… providing redundancy, enabling more effective maintenance schedules and fewer delays, and generally increasing the value of existing Metro lines. These things need to be taken into account.
In addition to the Rosslyn Metro tunnel I\’ve provided two streetcar routes and one express BRT route crossing the Potomac between Rosslyn and the West End. How is that not redundant?
You\’re talking about building super lines that can handle any problem and still run smoothly. I\’m talking about providing options so when a problem comes up it can be bypassed.
It\’s like building arterial highways surrounded by cul-de-sacs versus building a grid of like-sized roads.
March 25th, 2008 at 2:57 pm
>Transportation departments assume that the road network will grow ever larger. We ought to be questioning that assumption, saying look, under business as usual we can afford to build a new Metro line or a streetcar network. If we rethink our priorities and stop building every massive road project that comes along the pike, we can do both.
Of course I agree. And we *should* be fighting that battle.
But I don’t think it’s necessary or legitimate to say that either we revolutionize the way this country spends infrastructure money, or we give up entirely. Of course a revolution would be ideal, but just in case we lose that battle, I still want the best system I can get.
March 25th, 2008 at 3:08 pm
It should be clear that I’m not at all talking about doing Metro only. And I think it’s quite likely that just the four Tysons Metro stations would generate as much new development potential as a complete streetcar network.
I’m not talking about super lines, etc. I’m saying that additional Metro core capacity is needed to correct the system’s design flaws and ensure that it can be made to work for the very long term. A new crossing and some streetcar lines will do nothing to diminish the growing ridership from inner suburbs into and through the core. And it would be nice to be able to take a few core stations out of service for a weekend without crippling the entire system.
But my main point is this–we should not be negotiating with ourselves about new transit capacity. We should be saying that new transit capacity is worth the investment and new road capacity is not. Then we can prioritize transit investments. But making the case for streetcars rather than heavy rail lines, at this point, is just doing the job of the highway lobby.
March 25th, 2008 at 3:09 pm
The argument with the New Blue line is that it adds redundancy within the Metro system. You may provide for alternative means, but streetcars are not the same as Metro - I’m guessing that all of those streetcars won’t be able to run on totally dedicated ROWs, thus they’ll be slower. You’d be asking for patrons to transfer more often, for example, to get them around the Rosslyn bottleneck.
The level of service is not the same, and should not be treated as such, regardless of cost. Streetcars have huge improvements over buses in many fashions and certainly appeal to choice riders much more, but in operational terms, they are much more bus-like than metro-like when forced into mixed traffic. (If your plan includes substantial dedicated ROW for the cars within the core of DC and I missed it, please correct me)
Why a new Metro line like the New Blue makes sense is because it would provide redundancy within the Metrorail system, as well as effectively double the max capacity of the tail ends of the orange and blue lines (to the max of about 26 trains per hour - one every 135 seconds). The interlining of the Metro was one of the compromises made when they built the system, so the logical next step is to fix those problems, as they then offer major spillover effects for the rest of the Metro system.
More pragmatically, I think a new downtown tube will take a long time to plan, fund, and implement, thus it’s important to include it in a vision and begin chipping away at the process now.
March 25th, 2008 at 3:12 pm
I’ll second Alex’s points. Heavy and light rail should be viewed as complements and not supplements, particularly when we’re making the case for these projects to the public.
And, by the time any new heavy rail line in the District approached the construction phase, the city would have at least three, and possibly more, light rail lines under construction or in operation.
March 25th, 2008 at 4:21 pm
> If your plan includes substantial dedicated ROW for the cars within the core of DC and I missed it, please correct me
I said we could do either/or depending on the location. Given the width of streets in DC there’s no good reason why the majority couldn’t operate in some sort of exclusive ROW.
>A new crossing and some streetcar lines will do nothing to diminish the growing ridership from inner suburbs into and through the core.
Huh? Why not? Folks who live or work closer to a streetcar stop will take it rather than transfer. Any of the tens of thousands of workers in the lower West End near State and IMF will find that the streetcar delivers them much closer to their door than Foggy Bottom. And it’s not only those lines, but it’s also the connection up to the extended Purple line. Under my plan the entire segment of riders going between Virginia and Upper Northwest or Montgomery County can bypass downtown entirely.
*And* there’s the MARC/VRE improvements I’m suggesting to boot. Doesn’t help a whole lot in the Dulles/66 corridors, but should elsewhere. A lot.
> Heavy and light rail should be viewed as complements and not supplements
I agree. Metro should provide the trunk service and light rail should work around the edges. But we *already have* the heavy rail component. If this were LA, where there’s a single solitary Metro line, I’d be singing a different tune, but this isn’t LA. I am making this case in the context of 100 miles of existing Metro and zero miles of existing light rail. Build this system, then maybe it makes sense to include more Metro in the next wave of expansion. Talking about balance now is like talking about spending equal cash on highways and transit. The existing system has a huge gap that needs to be filled for balance to legitimately enter the equation.
>I think it’s quite likely that just the four Tysons Metro stations would generate as much new development potential as a complete streetcar network.
I don’t think so, but regardless, Tysons Corner is a special case. You may notice I did provide full Metro service to those four stations because I do think Tysons needs to be plugged in to the primary trunk network.
>making the case for streetcars rather than heavy rail lines, at this point, is just doing the job of the highway lobby.
I will modify the text to clarify that we should invest more in transit and less in roads, but we’re just going to have to agree to disagree on that point otherwise. I do not believe it is an all or nothing question, nor do I believe Metro needs the protection, nor do I believe that shutting out reality is the best way to bring about desired change.
March 25th, 2008 at 4:53 pm
I think you really underestimate the stress at the core of the system, and the additional demand that will be placed on that system by the recent boom in TOD. There is massive employment growth in the core, and a lot of that labor force is going to come from places where a streetcar commute will simply take too long. I think a streetcar network is vital, and so is additional core capacity.
Frankly, I think you misread public opinion on streetcars. We’re going to see a lot more streetcar mileage appear in the next few years than we will heavy rail, and it’s not as if feeder systems are completely ignored; half a million people ride Metrobus every day. I’d go so far as to say that the consensus position on Metro is that we shouldn’t expect any additional core capacity. That’s a mistake, and I absolutely believe that we should be reminding people that better core service isn’t something we should consider optional.
And excuse me, but I think “reality,” as you’d have it, suggests that the time is right for a shift in transportation spending priorities, and I intend to argue that that’s just what we need. I’m not willing to accept a vision in which key transportation needs are ignored, and I think it’s incredibly counterproductive to make statements that can and will be construed by others as concluding that “heavy rail is too expensive.”
March 25th, 2008 at 5:02 pm
And I think you drastically misread the ability of streetcars and interurbans to move large numbers of people. They’re all we had when our city was far more transit dependent than it is now, and they worked very well.
At this point we’re going in circles on the rest.
March 25th, 2008 at 5:54 pm
I added the following two paragraphs to the top of the text:
For the record, BeyondDC doesn’t mean to suggest that MetroRail isn’t worth the money. Metro’s effect on the Washington region has been nothing short of revolutionary; it has been worth every penny spent, many times over. Furthermore, streetcars alone would be incapable of providing all the rail transit necessary for a city as large and diverse as ours. This vision can only be taken within our regional context of 100 miles of existing MetroRail service. Indeed, where we think MetroRail is the best mode for a new route, such as between the Orange line and Tysons Corner, we’ve included it despite the cost. The point here is that given what we already have, streetcars and interurbans are the best capital investment we can make right now. After these lines are built it may once again make sense to build significantly more MetroRail.
A final note before getting into technical explanation: Making such hard decisions as MetroRail versus streetcars is only necessary in the first place because our country’s transportation spending is wildly out of whack. Decades of widenings and new roads have brought our planet to the brink of disaster, ruined our cities, and taken our economy hostage, all with the result that congestion is far worse now than when we started. Like pre-industrial medicine focusing on bloodletting, highway engineering is junk science. The time, lives and money spent on it don’t change the fact that it simply doesn’t work. For all the arguments and numbers BeyondDC is throwing out contending that Metro isn’t the most efficient use of our transportation dollars these days, anything given over to road or parking capacity is severely worse-spent. More MetroRail would be good; it’s just that more streetcars would be better. More roads, in any event, would be utterly counterproductive.
March 26th, 2008 at 9:31 am
There’s NOTHING that more hovercraft ferries can’t improve.
Doesn’t seem like residents around Huntington, Largo, or most of “Ward 9″ are particularly well served by surface rail here. I always envisioned the terminus stations radiating out a spider’s web of light rail lines. You’ve got existing surface transit corridors like Route 1 in Alexandria that are desperate for that kind of public transit. I appreciate getting people from Lorton and Belvoir to the Pentagon, but there’s a lot more than doesn’t really help the THOUSANDS who file through Huntington every day.
Do that many people really need to get from King Street to George Mason?
March 26th, 2008 at 9:55 am
I’m not sure I understand your comments.
Line F2 goes right along Route 1 in Alexandria. I just didn’t mark all the stops for streetcars since there are so many.
And PG County, Fairfax and Montgomery all get about the same number of new lines. Not as many as DC or Arlington/Alexandria, true, but then those jurisdictions are closer in and more urban. I’m also not sure where else to *put* any lines in PG County. I admit I’m less familiar w/ it than with any of the other regional jurisdictions, but most of the natural transit corridors are already covered by Metro. I’m not opposed to proposing a line heading out from Largo on principle, but where would it go? Nothing jumps out to me.
March 26th, 2008 at 10:07 am
Oh, regarding the King Street / GMU line: I doubt very many people would ride all the way from one end to the other, but then who does that on any of the Metro lines? The point is it’s a good *corridor*.
It starts in Old Town, then hits all the following in succession:
1) King Street, which is effectively Alexandria’s downtown.
2) The Landmark Mall area, which is home to the densest census tracts in the region outside the District of Columbia.
3) Annandale, which sucks now but has potential to be a perfectly nice local downtown.
4) Northern Virginia Community College (college students = riders)
5) Fair City Mall, which has enormous TOD potential.
6) Downtown Fairfax
7) GMU.
In other words, Little River Turnpike has potential to be the sort of string-of-pearls corridor that really thrives with rail.
March 26th, 2008 at 10:28 am
Alright, now I see. I was reading the F2 line as a subway. Need more coffee.
The area bounded by 197, Pennsylvania Avenue and 301 has exploded with residential and commercial development (most affluent African-American suburb in America, remember?) And Bowie isn’t just one town but 10 miles of sprawling suburb. A single Bowie Metro station terminus would be as swamped as Vienna.
There’s A LOT of opportunity to serve that area with light rail, particularly feeding people not just westward to DC but eastward to Annapolis as well. It’s not just a commute-to-work-downtown thing; it’s a let’s-hit-the-Chesapeake-and-have-fun thing. And it would help Davidsonville from looking like Capitol Heights.
March 26th, 2008 at 5:31 pm
If we expand suburban transit we will create bottlenecks in the heavy rail system. The capacity across the Potomac is already emerging as a bottleneck. A second bottleneck will arise on the Red Line between Union Station and Gallery Place if there is a major expansion of commuter rail service.
The lowest-cost remedy to the first bottleneck is to separate the Green and Yellow Lines so that the entire capacity of the Yellow Line bridge becomes available - essentially adding half a bridge. One configuration would be to extend the Green Line from Waterfront under 4th Street SW to connect to Federal Center SW (with a foot connection to L’Enfant Plaza) and on to Judiciary Square. Going one stop farther to Union Station would also relieve the future Red Line bottleneck.
March 27th, 2008 at 8:03 am
I’m with br. If they’re going to expand Metro service, they’ll get more bang for the buck adding capacity at the core to deal with the passengers light rail will bring, than they will creating a Brown Line to East Bumf**k.
And it’ll take more than just a walkway between Gallery Place and Metro Center.
March 27th, 2008 at 2:58 pm
I agree. When it comes time to expand Metro, the places to do it are those that will add capacity to the *trunk* system, not as new tails in the suburbs.
Tysons Corner, by virtue of its six-figure employment population, status as the region’s #1 shopping destination, and outrageously high TOD potential, is the only exception.