» Boston’s Green Line extension, bloated after years of planning, gets slimmed down. A lesson for other cities.
Given how reliant the people of New York City are on their Subway, an outsider just looking at ridership data might conclude that the system must be paved with gold, or at least its stations must be decent to look at. After all, it wouldn’t be unreasonable to assume that the comfort of a transit system plays an essential role in encouraging people to abandon their cars and get on the train or bus. That’s why, some would argue, it’s so important to put amenities like USB charging and wifi into transit vehicles.
Yet anyone who has ever ridden the Subway knows first hand that its success has nothing to do with aesthetics or access to luxury amenities. Stations are hardly in good shape, trains are packed, and cell service is spotty at best. People ride the Subway in spite of these things; they ride it because it’s fast, it’s frequent, and it’s (relatively) reliable.
Too often, this simple fact is ignored by public agencies actually making decisions about how to invest. New York’s own $4 billion World Trade Center Transportation Hub—perhaps the world’s single-most expensive station—is evidence of that; rather than improve service frequency or speed, officials chose to direct public funds to a white monument that does nothing to actually ease the lives of daily commuters.
Initial plans for the MBTA’s Green Line extension, which would extend light rail service from Cambridge into Somerville and Medford—all three are close-in suburbs of Boston—featured none of the extravagances of downtown Manhattan’s new transit terminal. Yet it too was designed with unnecessary features that, while nice, did little to actually solve the travel needs of its future users. Its projected construction costs exploded such that officials announced last year the proposal could be cancelled. Now, after several months of review, the MBTA and the state government have voted to proceed with design changes meant to significantly bring down costs—but without compromising the quality of transit service to be offered to riders.
Agencies with pricey projects around the country should look for similar opportunities to minimize costs.
A rail line for one of the nation’s most transit-friendly communities
The seven-station extension of the Green Line proposed for Boston would be the region’s first rapid transit expansion since the completion of the Red Line extension in 1985. Running along two branches northwest from today’s terminus at Lechmere—one branch to Union Square in Somerville, the other to College Avenue, near Tufts University—the 4.7 miles of new track would run along existing commuter rail lines and connect to some of the country’s densest, most transit-friendly neighborhoods. See the Transit Explorer map for details.
The project would vastly improve connections of Somerville and Medford residents to jobs hubs in Cambridge and Boston and is expected to attract 45,000 daily riders by 2030. That would make it one of the most effective projects in the nation from the perspective of riders per mile operated.
The project has been in planning for decades. A 1990 lawsuit required that the line be completed by 2011 as a sort of trade-off in exchange for the completion of the Big Dig. But faced with limited funding, mounting MBTA debt, and a lack of adequate state political support, the project failed to gain traction and the state kept pushing it off. Finally, initial construction activity began in 2013 and the federal government agreed to provide a significant New Starts grant to the project in 2015.
Yet even as the project advanced, its estimated construction costs mounted ominously. Federal reports show total costs rising from $1.1 billion in 2013 to $1.7 billion in 2014 to $2.3 billion in early 2015. By late last year, the project’s budget had reached $3 billion, and the state announced that it was not only cancelling certain contracts related to its completion but also that, in the context of a transit agency stretched beyond anything it could handle, it was considering cancelling it altogether.*
Redesign by necessity
But the MBTA submitted the Green Line extension to a review by a project management team, and that group released its report on how to save the project yesterday. The document details how the project’s price tag could be substantially reduced, returning it to a (still-expensive but) doable $2.3 billion cost.
The changes are reasonable because, rather than cutting the quality of service provided to riders in terms of transit service, they focus on aesthetic elements that, even if they improve the general atmosphere of the system, likely do little to actually get people onto trains. The essentials, like the frequency of trains, their speed, and their capacity, are maintained.
What the team does recommend is vastly simplifying proposed station designs. As the below chart from the report indicates, the stations would be slimmed down. 15 elevators would be replaced by six (while maintaining wheelchair accessibility throughout); escalators and fare gates would be eliminated entirely; and full-length station canopies would be cut down to shelters. In total, these changes would slash almost $300 million from the project budget, with virtually no impact on ridership experience.
The changes will make the MBTA’s built footprint less visible; there will be no Calatrava extravagances here. As the below images show, Ball Square station in Medford was initially designed to feature a plaza, a headhouse (a multi-story building featuring elevators and escalators), a concourse, and a fully covered platform. What would be built in its place is an open-air and very simple train stop, with more room for future transit-oriented development.
Customers may suffer through the cold for a few more minutes, but trains will come frequently enough that shouldn’t be a major concern. Meanwhile, MBTA will save itself millions of dollars of future maintenance costs not upkeeping expensive and unreliable machinery and not keeping thousands of square feet of interior space clean. These savings aren’t even accounted for in the capital costs of the project but they’ll pay off in a lower operating budget for years to come.
Initial proposal | Revised proposal | |
![]() |
![]() |
The management team also proposes a reduction in the size of the proposed vehicle maintenance facility and affiliated transportation building, which together will save more than $100 million and not affect the MBTA’s ability to keep trains moving. An expensive parking deck will be replaced with a parking lot. Bridges that the initial plan suggested needed to be completely replaced will be simply renovated.
If the choices about what to eliminate seem obvious, consider the alternative: The Purple Line in suburban Washington, D.C. also underwent a considerable cost-cutting process earlier this year. Yet the changes there will reduce passenger quality of service by increasing headways between trains, reducing train capacity, and lengthening the walking distance between the line and a Metro station in Silver Spring. While these efficiencies aren’t dramatic enough to imperil the overall value of the line, they will hurt the passenger experience in the long term, while those on the Green Line will not.
The changes in Boston must be approved by the Federal Transit Administration, which has final say over whether the redesigned project meets the initial project goals. And local governments need to scrounge up an additional $73 million to meet the gap in project costs that remains—without this funding, the project could still be on the chopping block. Yet these are surmountable obstacles and the project now seems likely to move forward.
Nuance by design
Boston’s example is no panacea; the quality of the transit environment does matter. While nice materials, enclosed stations, escalators, and overhead canopies may do little to expand ridership, they improve peoples’ daily experience, and that’s important. The nicer we can make the public sphere, the better our cities will be to live in.
But it’s refreshing to see a transit agency propose a cost-cutting approach that does nothing to negatively impact the level of service being proposed. Rather than take out a constricted budgetary environment on riders by reducing service, the MBTA is proposing to stick to the essentials, and that’s the right move.
Were construction costs in Boston lower, the MBTA could afford to give riders both good service and a comfortable environment. But like transit agencies around the country, the MBTA has been unable to lower costs to international standards. In this environment, it serves as a model for other agencies looking to invest in transit on a limited budget.
* There is some question as to whether the state actually can ever cancel the project, given that it was mandated through the legal process.
Image at top from Flickr user Bill Damon (cc). Other images from Green Line project management team report.
21 replies on “Frequent service, not escalator access, is what attracts transit users”
Well, the article finally gets around to the real issue at the end: that MBTA has been unable to lower construction costs. The same issue that afflicted the big dig. Why is this?
Mention is made of the need to relocate the railroad tracks. There does not appear to be any indication of whether or not the cost of this is included in the cost of the extension. If the cost is included, then the total cost does not look too out of line with other schemes. It should be noted that only one track needs to be relocated; a new track should be constructed on the eastern side of the existing tracks, and then the western track can be turned over to the light rail system.
No indication is given of the likely peak hour headways; if these are five minutes or more it would be possible to single the track at stations, so that only one platform is needed. This saves platform and shelter construction, at the cost of sprung switches at each end of the platform. It also limits the need for passengers to cross one or both tracks to get to the other platform. The sole platform would then be constructed on what would be the second track, so saving substantial space – costly space if needed to be purchased?
I query the use of ‘redundant’. In English something that is redundant is unnecessary – to be disposed of. Why build ‘redundant’ elevators?
Would it not be possible to insert a light rail track either side of the existing railroad tracks, so that they do not need to be relocated? This would surely save a very substantial amount of money.
It may be possible to provide space for the new tracks at bridges at low cost; well and good. But if this is expensive, the possibility should be considered of raising the tracks to road level and using a grade crossing. This would also obviate the need for escalators and stairs, though absolute traffic priority for the LRVs would be required, with platforms on the far side of the crossings.
No, you can’t single-track at stations while running a train every five minutes. In theory you could schedule it, but the schedule would be fragile. Meets are best done at stations rather than on running track, and in this case, a meet would be required every station. For reference, the Canada Line has single-track outer termini, with service every 7 minutes each, but other than the terminal station, everything is double-track. On through-track, the Yokosuka Line has some single-track segments (with a meet at a station), with 15-minute frequency.
Moving the mainline tracks absolutely does not excuse the high cost. There’s a standard cost for an open cut, which would be widened to allow for two more tracks in the ROW. It’s on the order of a few tens of millions of dollars per kilometer, excluding property acquisition. Property acquisition is cheap, both because for most of the route the ROW is far enough from houses that it can be widened without takings, and for the rest, the property values per unit of land area aren’t that high. The cost of the Green Line per unit length is high for a fully underground subway.
The main cost item is the bridge modifications. There look to be 12 grade separations that require widening: 2 bridges on the Union Square branch, 10 bridges on the main line. A few other grade separations exist but are wide enough, because the ROW has four tracks. One additional grade separation involves a road undercrossing, and then the rail ROW is almost wide enough for four tracks anyway, and can be widened at minimal cost. An urban grade sep can run into the $100 million. The Chuo Line track elevation project cost about that much per grade separation, and the long-term Caltrain grade separation project uses that standardized cost. When the rail is elevated, you’re running into elevated rail costs, which work out to about the same per unit of length, taking into account how often roads typically cross mainline ROWs; when the rail is in a cut, it involves reconstructing road bridges, and either way it’s around the same.
So there you have it: the original budget was reasonable, given the scope. The current one, lol.
Eggs, not air conditioning, is what attracts customers to the supermarket.
Why spend money on such silly things as “comfort” when your captive audience is forced to purchase your product anyway?
The fact is, given the choice between waiting on a wet, wind-swept platform with only a bus shelter to protect you from the elements, or an uber, well, guess which one many of your customers will decide to use?
Treat your customers badly and they wont be your customers forever. New Yorkers deal with abysmal station conditions because the competition isn’t better. That’s not quite the case in Boston.
If the faux-cab breaks down en route that will not be better service. Still I get what you are saying. But since the faux-cab’s business model does not cover depreciation of the vehicle it could become a problem when the pyramid scheme begins to reach the bottom of the barrel of suckers willing to drive their car into the ground as if it were a payday-loan.
JJJ – Your analogy is disingenuous here. In the egg-buying market in Boston, there are plenty of competitors all selling eggs. In the rapid transit market in Boston, there is one supplier, the MBTA. Anybody seeking to buy eggs in Boston can choose from literally hundreds of suppliers ranging from Walmart to chain supermarkets to convenience stores to farmer’s markets. Anybody seeking to ride a subway train has to take MBTA.
You do cite a possible “competitor” to MBTA; namely, Uber. But Uber (and cabs in general) aren’t truly a competitor to MBTA on all aspects of taking a trip downtown. In particular, MBTA is faster and cheaper than Uber/cabs from most destinations to downtown during most of the weekday, mostly because (1) the Green Line in its subway gets to skip all traffic lights downtown and (2) the Green Line has a flat fare vs. Uber/cabs which charge for both distance and time. So, yes, for customers who value comfort over speed or per-trip cost, MBTA might lose them to Uber/cabs. And MBTA might also lose customers to Uber/cabs during weekends and at night when streets are emptier (and thus MBTA loses its competitive advantage in speed). But MBTA isn’t going to lose all its customers to Uber/cabs, because MBTA still retains some advantages over Uber/cabs for some riders.
Uber/Lyft are compliments to rapid transit. Where personalized transportation is required, they are better for passengers in non-commute hours.
*I should have made it clearer in the first paragraph: what I’m saying is that supermarkets (and all the other sellers of eggs) have to differentiate themselves further (by offering A/C or whatnot) to get customers to choose them BECAUSE of the fierce competition in their market. The MBTA doesn’t really need to do this, because it essentially is a monopoly operator in the marketplace for “people who want to purchase subway rides in Boston.”
People arent purchasing a subway ride. Theyre purchasing transportation. Look at Washington DC where ridership is collapsing because customers can no longer depend on WMATA for timely transportation. Alternatives exist. They are private cars, cabs, ubers, bicycle, bikeshare, walking, or telecommuting.
MBTA wont lose ALL their customers, but they dont have to. Death spirals happen because people start to leave slowly. Im not saying the lack of shelters will lead to a death spiral, Im just stating that losing 100 customers here and there can lead to a service cut which leads to other customers leaving etc.
Elevators are requirements, not luxuries. They are needed by people with disabilities and others for using the rail system.
Every station in the revised plan includes barrier-free access via ADA-compliant ramps. In many cases, these ramps become the primary egress for all passengers.
Whether this adds distance to the access route — or actually saves disabled/non-disabled riders access time — appears to vary on a stop-by-stop basis.
If I understand the two drawings, the “accessible” pathway will involve crossing one of the tracks to reach the platform. Going to be fun in the snow. While I am not here to burn funding, I have $5 that if this actually gets built, the stripper model with the usual overruns will cost nearly the same as the “loaded” model. Transit projects in the US are just overpriced.
Correct, it looks like passengers would walk across the track. That leads to the trolleys being faced with stop signs, 5mph speed limits, and other burdensome rules which slow down service, create jankier rides, and generally institutionalize 100 years of slow service to save a few bucks today.
Anyone who thinks Green Line drivers are limited to 5mph entering or leaving stations with pedestrian grade crossings has never used the Green Line in their life.
Snark is lovely, but it doesnt trump fact. MBTA Green Line trolleys ere limited to either 3mph, 5mph, 7mph or 10mph depending on the geometry of the crossing.
You are simply wrong. What the MBTA now intends to build in Somerville is, for all intents and purposes, another D Line.
Many if not most D Line stops involve riders crossing the tracks for access from one direction or another. The trains do not enter at a crawl. The trains do not leave at a crawl.
The Green Line is one of the busiest single subway lines in America, and its central tunnel has always had a very high traffic volume. To some degree “driving like you mean it” has been grandfathered into the operational culture. Light rail drivers in Portland and Phoenix and St Louis and Seattle may be overcautious as shit, but they aren’t here.
As Yonah explicitly states, this is a drawback of amenities, not of service quantity or speed. Both of which are calculable. The only genuine potential service impact is in the elimination of fare gates. All stations remain controlled-access, I would tend to join you in hoping for a push to reverse that decision.
I would also note that the MBTA has hardly been immune to the consultant-driven overdesign-industrial complex in its last few major projects. The results have been some of the slowest service chokepoints in the entire system: the Silver Line’s bus tunnel and the Green Line’s bafflingly lethargic segment below North Station/Boston Garden, submerged in the early 2000s. All future Somerville trains must pass through the latter, which, despite full grade separation, has stop signs and 5mph running and signals that default to red.
Overdesign more often creates problems than alleviates them.
[directed at JJJ, not Yonah]
I don’t like the stance that is conveyed in this article. The cost cutting of public transport projects are nothing but a capitulation to the car-lobby. I haven’t heard of any project down-scaling when the Big Dig ran over budget. So should it be done when a PT project costs a bit more?
The down-scaled facilities are by far not superfluous luxury but actual necessities. The difference between a properly sized canopy and a minuscule shelter will be felt by waiting passengers. And it won’t feel good for them nor will it do any good for the perception people have of this transport mode.
The issue is that American transit project costs are extremely high, an order of magnitude more expensive per project than technically-similar counterparts anywhere else in the developed world. These excessive costs make new transit projects more difficult to execute than they have to be, and prevents widespread implementation. When a moderate extension in Boston costs $3 billion to implement, do you think that transit planners in Cleveland would want to follow suit with their own light rail? … I didn’t think so.
This, in its turn, keeps mass transit infrastructure from being able to act as an option for a significant portion of the populace. That, in its turn, keeps drivers from being able to see it as an option in any meaningful way — which is ultimately why politics do not favor mass transit investment. A similar cycle is at work in the politics of bike infrastructure.
Contrast this with Britain. British passenger rail ridership is the highest it’s been since the 1920s, and that means it’s perfectly normal and acceptable to channel large chunks of infrastructure capital into their railroad network there. But that explosion in ridership was not brought about by shinier newer facilities. Far from it: it started because there was cheap, reliable, well-marketed service. The gewgaws came later.
Incidentally, while it by no means is as comfortable as a waiting room, a modern shelter with heat lamps is more comfortable than you may think, especially in a Northeastern winter.
MBTA should be spending its money on transit service, not station enhancement. While 2 elevators per station and tons of escalators might be nice, they should really be planned and funded by the local community. MBTA should focus on the lines frequency and operations.
[…] lanes alone, no one would have to wait long for a minibus, so not only would journey times be shorter; they’d also be more […]