Construction Costs: Metro Accessibility
It is much harder to find estimates of construction costs for elevator access to metro stations than to find estimates for the costs of new tunnels. Accessibility isn’t as flashy, so it does not get reported in the international trade press. If you’re a humble blogger who has no idea what the Japanese phrases for “elevator,” “wheelchair accessibility,” and “construction costs” are, you have no way of figuring out how much it cost Tokyo to get its legacy subway network to be about 80-90% accessible. Thankfully, there are enough European cities where I do know those phrases that I do have a list. It comprises just eight cities, two in the US and six in Europe, and in at least one case (Milan) I don’t fully trust the numbers.
New York:the current capital plan spends $740 million on 19 stations, and a single elevator is $10 million if I understand Curbed correctly. The headline figure is $39 million per station.
Boston: MBTA head of systemwide accessibility, Laura Brelsford, told me that Symphony is a $25 million, 4 elevator project, and that Copley and Arlington were in the $30-40 million range each. I can’t find a link, unfortunately.
London: here is a source saying a £76 million fund would make 12 stations step-free, but I don’t know which stations. This source states a higher figure, £200 million for 13 stations, which in PPP terms is about $22 million per station. The lower figure is only $9 million per station.
Paris: nothing is accessible except M14 and the RER and there are no plans to make anything accessible, but a disability rights organization estimates the cost of making all 303 Metro stations accessible at €4-6 billion. Subtracting M14-only stations this is also about $22 million per station at the midpoint.
Madrid: a 2016-20 plan breaks down costs per station. They’re about €5 million per station per line it serves (so a 3-line transfer is €15 million), and a single elevator is €1.5 million. The relevant numbers are on PDF-pp. 12-17. This document used to exist in English too but the URL broke and the Web Archive doesn’t have it. Lumping all remaining stations together, few of which are transfers, gives €531.9 million for 93 stations, or $7.5 million per station.
Berlin: follow the link from my tweetstorm. The U-Bahn is about to complete step-free access by 2020. The cost is €800,000 per elevator, €2 million per station, or $2.6 million per station.
Barcelona: I have three separate sets of numbers, and am placing the city on this list with the lower two. Here is a source describing the installation of elevators at 14 stations for €26.8 million. Here is a source describing a single station on a different line to be made accessible for €2.1 million. In contrast, a document describing the plan for universal accessibility states the cost up to 2009 at €390.97 million euros (PDF-p. 87) for 73 stations (PDF-p. 83). The first two sources average to $2.5 million per station, but the third one, covering more stations, including complex transfer points, averages $7 million per station in what I imagine are in 2009 prices.
Milan: this article seems to be saying that elevator installation at 21 stations is budgeted at €24.5 million, all on Lines 1 and 2, which are shallow cut-and-cover lines built under narrow medieval streets. I’m not sure whether all of these are elevators, though – the Milan Metro site says that it has elevators at some stations and stairlifts at others. This totals $1.5 million per station.
The low construction cost of elevator accessibility in Berlin is notable since the city’s tunneling costs are not so low. I am still trying to talk to local transit officials about this, but my surface understanding is that fewer elevators are needed per station. The city uses island platforms rather than side platforms and has shallow subsurface stations with access from raised street medians, so a single elevator covers the entire station.
That said, Paris and New York both have extensive cut-and-cover construction under very wide streets, much like Berlin. A multiple of 2 over Berlin costs is to be expected as they have side platforms, and thus all station infrastructure needs to be doubled. However, Paris’s costs are (apparently) higher than Berlin’s by a factor of 9 and New York’s by a factor of 15. Paris is not a high-cost city when it comes to tunneling, and it’s possible that the activists just estimated it based on London costs, or that the government made up a high number in order to sandbag the calls for step-free accessibility.
Pedestrian Observations 
In Japanese, the term for making an existing facility accessible is バリアフリー化 “Barrier Free Change”. The general term for accessibility is “Barrier Free”: バリアフリー
Words for construction cost generally include this kanji: 費
建設費
費用
費
In other cities, do elevators have their own fare control areas? I know that in New York, even when there are island platforms, there is an elevator from platform to fare control, and another from outside fare control to the street.
Berlin doesn’t have fare barriers, but I think (am not sure) Madrid, Barcelona, and Milan do.
In general the four German “true subways” work on the “honor system” but in two of them (Munich and Hamburg) there are still “Bahnsteigkarten” for those who wish to access the platform but not ride a train. Technically being on the platform without a ticket is treated as fare-dodging. It is obviously an anti-homeless measure.
On the Nuremberg subway there are “symbolic faregates” where one can stamp single or multiple ride tickets that aren’t yet stamped and which usually bear words like “riding without a ticket will cost a 60€ fine and include criminal prosecution” or “no access without valid ticket”…
Ugh, I constantly go on the platforms without paying because I use the U-Bahn for hotspots. I’m pretty sure I’ve tweeted while schwarzstehen at Gesundbrunnen (not schwarzfahren, I didn’t actually ride a train).
I’m not sure that’s actually the case for Berlin. Berlin police are comparatively liberal compared to Munich police…
What I mean to say is: in systems with fare control and cheaper costs, are there elevators on either side of fare control, or do they have the elevator go directly to street level with its own dedicated fare barrier at the surface?
IMO part of the reason why New York stations tend to be so palatial is the placement of fare control at an intermediate mezzanine level (necessitating large excavation underneath the street), rather than at the surface.
Depends on the individual station. If they dug a hole to put in the tracks and platforms, filling up the hole could cost more than making the concrete level and slapping some tile up.
At the same time getting rid of an entire floor reduces the total depth you need to dig. The IRT stations without crossovers are much shallower.
Stack it like they did on Central Park west you only need half as much street level stuff. It depends on the station.
I don’t know for sure, but the Madrid document mentions multiple elevators per station even without transfers.
DC’s Metro has a few stations with elevators similar to the ones you describe. The dedicated fare gate is on the platform, and not the street. (The elevators don’t actually go directly to the street, but they are outside of the normal fare control area) Of course, Metro isn’t low-cost, and it’s stations are palatial… though for different reasons.
Fortunately, DC Metro was built with elevators nearly from the get-go, despite the best efforts of its first general manager. Only a few stations were under construction before Congress intervened to enforce disability access.
Yes to Madrid and Barcelona. There are even stations on Madrid metro where you have to change trains across the platform and pass through barriers again just to continue along the same line, an extra fare. Such as Tres Olivos. At Villaverde Alto (and likely many stations) you can see the side platforms through a simple fence but to reach them requires going underground, via barriers, and then back up. Quite obnoxious compared to simple German POP stations.
Sometimes it feels like the cheap construction cost leads to excessively large stations. Chamartín metro has four levels including two mezzanine levels. A stunningly large underground complex. I remember looking down from one platform and seeing two more fully built platforms with everything but tunnels. A placeholder for future expansion. Very forward thinking I suppose.
Re: Berlin. The lack of fare gates also means a reduction in the number of elevators necessary to make stations step-free.
Yes, definitely, but at New York subway stations built directly under the street with no mezzanine, like 110th on the 1 and 72nd on the 1/2/3, only a single elevator is required per platform.
True, and that’s how they built out the new entrances at 96th and 72nd St. But no island platforms for nearly all of the subsurface IRT stops.
Another reason for the disparity between tunneling and elevator costs in Berlin is the ground: sand is not a good material to dig tunnels in, so most lines stayed relatively close to the surface(this also cuts down on elevator costs), but the U5 has to go rather deep to go below other lines, increasing its cost (it also doesn’t help that it has to cross beneath the Spree; in certain areas the soil had to be frozen to allow tunneling work to take place [http://www.projekt-u5.de/de/bautagebuch/arbeiten-im-eis/]).
Plus the U5 extension went through the medieval core of historic Berlin. Archeological find were almost a certainty and as if on cue they found a medieval city hall…
…are they going to dig it up and display it the way they do when they find artifacts in Mexico City (and I think also Athens)?
https://www.berlin.de/landesdenkmalamt/denkmalpflege/bodendenkmalpflege/grabungen/u5-grabung/ This article mentions the findings bit not what they will do with them. I find it amazing how precise dendrochronology has become…
https://www.berliner-woche.de/mitte/c-verkehr/richtfest-fuer-bahnhof-rotes-rathaus-ausstellung-zu-archaeologischen-funden_a108531 here they indicate it’s a question of costs…
Regarding costs in Japan, it’s hard to find figures for individual stations as most material online gives only aggregate costs, which may include items other than elevators such as platform barriers, additional safety railings, raised tiles for the blind, etc. spread over hundreds of stations and across systems or regions. However I did find a figure in a publication of the Japan Subway Association (No. 199, Nov. 2013) of a general amount *per station* of 1 billion yen plus in the more difficult retrofit cases. At current exchange rates that would be around 9 million USD.
Page 14 lower left corner paragraph:
Click to access HWnpRJbHAhOL.pdf
Too add for comparison and in acknowledgement that JR lines also serve as metro type transport, the cost to add (I assume mainly above-ground) elevators to 87 JR-owned stations nationwide came out to 12.1 billion yen (~108.7 Million USD) or about 1.2 million USD per station.
http://report.jbaudit.go.jp/org/h12/2000-h12-0631-0.htm
And this study estimates average costs per elevator in 2003 (including associated removal of debris and other materials) of 72 million yen (roughly 600,000 USD, give or take some exchange rate fluctuations; with a couple elevators per station, works out close to the numbers Andrew in Ezo dug up.) Equivalent costs per escalator were 54 million yen (450,000 USD). This is averaged across Japan, although given that barrier-free work has been focused on stations with more than 5000 daily riders, this probably is pretty close to Tokyo costs except at particularly complex stations.
As Andrew in Ezo points out, a lot of work gets bundled together in barrier free retrofits, including platform doors and new toilets. Incidentally costs are split three ways: national goverment, local government, and the operator al pay a third.
Click to access 95.pdf
Do you know what the cost of platform doors is? In Paris it’s around 4 million euros per double-track Metro station.
According to the chairman of the Odakyu Railway (also the chairman of the Private Railway Association), it costs 400~500 million yen per station in the case of private (non-JR) railways.
https://newswitch.jp/p/9581
Chicago Transit Authority has estimated the costs of accessible-izing the remaining stations. Some of the projects are combined with larger station or line reconstructions (e.g. Red Line North, Blue Line Eisenhower).
https://www.transitchicago.com/accessibility/asap/
So, $1.9 billion for 42 stations, not including elevator replacement? That’s New York bad…
Looks like $1.2 BN in 2017 dollars for Phase 2-4. Comes to $75M each, six are elevated/highway median and ten are subway.
The Red North future phases and Eisenhower Blue aren’t included in the estimates.
It’s not totally out of line. They’re relatively specifc on the estimates (though note they’re in inflated “year of expenditure” dollars). Several of these stations have annoying restrictions like being protected historic buildings. As you can guess, the subway stations in the Loop are the really expensive ones.
$24.0 million Austin, $16.1 million Montrose, $34.6 million California, $119.4 million State/Lake. This gives you an idea.
Clinton St. Blue Line is probably going to be really expensive due to location *under* the expressway.
Something similar is happening in Philadelphia where City Hall (Broad Street) is probably going to cost as much to make accessible as 10 other stations, and it’s pretty obvious why.
In Madrid the costs seem linear in the number of stations served, so City Hall should cost twice as much as the other stations. The numbers I have for Berlin are consistent with that as well, but this is out of a very small number (maybe just 1?) of still-inaccessible-as-of-2016 stations.
Another data point from Montreal. CAD$213 million for 14 stations, most of which are deep level with side platforms and concourses. Average 3 elevators per station.
http://www.stm.info/sites/default/files/pdf/fr/rapport_avancement2017.pdf#page=62
Berlin: this month’s BVG magazine gives a figure of 1-3M Euros per elevator. Project timespan is three to seven years with 40 to 50 institutions involved. During which time we have building cost inflation plus new building regulations which require replanning etc. In the latest absurd example of this process the preferred location for the elevator was deemed unsuitable by the monument protection authorities. Because… it would remove some ‘historic’ planters from the 60s or 70s. That was three years ago. The BVG hopes to start construction this year.
On page 6 (10-11) in this pdf:
Click to access PLUS_0319_Doppel_FINAL.pdf
Does the size or number of elevators increase the cost significantly
No idea about size, but costs seem linear in the number of elevators required.
Is there anything that fare-gating doesn’t ruin? (Aside, obviously, from the profits of contractors.)
Another data point for you: the MTA is currently spending $11.5 million to add two new elevators to the Flushing-Murray Hill LIRR station, more than doubling the original $5.7 million budget. It has also been badly delayed, due to unspecified additional design requirements (likely related to lead and asbestos abatement issues).
MTA doesn’t seem to budget properly for lead paint removal. They’ve got peeling lead paint all over their system and should probably have a systemwide program to replace it.
Something doesn’t add up here. Hydraulic elevators in two-story buildings with about the same rise as a subway elevator can be had for under $200k. I could maybe see excavation and utility relocation costing another $800k, but the same cost seems to apply equally to elevated and underground stations.
The information available isn’t helpful and is often misleading, but your London figures are not correct. The £200m cite is a boost to another figure, and isn’t specific to the 13 (simple stations, and some of the quietest ones too, so inexpensive as there’s no need to spend much due to simplicity, and a low value for money if high cost. No way would sleepy North Ealing be done ahead of the far busier stations in the area if it wasn’t a very cheap scheme!) stations listed as just gaining approval. To quote that Jan 2018 press release:
“Since the Mayor announced his £200m funding boost in December 2016, the much-needed delivery of step-free access has been brought forward at outer London stations. Newbury Park, Bromley-by-Bow and Buckhurst Hill are scheduled to be step-free in 2018, with work completing at Harrow-on-the-Hill in 2019. Work at Amersham, Cockfosters, Mill Hill East, Osterley and South Woodford will start this year. Work is also progressing at Knightsbridge, and vital interchanges including Bank, Finsbury Park and Victoria.”
Which is another 13 stations on top of the “Further 13” that the press release is about (Boston Manor, Burnt Oak, Debden, Hanger Lane, Ickenham, North Ealing, Northolt, Park Royal, Rickmansworth, Ruislip, Snaresbrook, Sudbury Hill and Wimbledon Park)
Bromley-by-Bow’s opening press release from a year ago explicitly mentions the £200m boost. South Woodford (opened Monday) and Victoria’s press releases don’t mention it, but these more recent ones don’t seem to anymore (the funding issues TfL is facing meaning that not all the £200m will be spent? certainly approvals have dried up).
So what we have is an existing sum that had £200m added to it, funding step free access for a certain number of stations (though complicated as other pots of money might part fund step free upgrades too – eg Crossrail funding those stations), that might not all be spent. It might be possible to get an accurate figure, but that’s a lot of work. £200m/13 stations is definitely not right though!
Wait, so how is this not “an extra £200m gets you 13 extra stations”? Is it because it is spread across 26 stations while at the same time there are other pots of money available, like the Crossrail funds you mention at Crossrail stations?
Construction Costs (vol2): Grade separation
Just how much should a grade separation cost?
I ask because eleven years after passing Prop 1A, the first work on HSR in southern california is beginning, with a single grade separation project. Durfee Avenue
Because everything HSR related is incredibly and totally insane this single grade separation is going to cost a mind numbing 105 million dollars
Why?!
Not to mention the total insanity of it taking eleven years to get this project started, similar project for Cal Train started almost immediately.
http://www.progressiverailroading.com/c_s/news/ACE-partners-to-start-grade-separation-in-Southern-California–57175
NYC is installing elevators in a single controversial station for $116 million (and counting…):
https://thecity.nyc/2019/04/long-fight-over-68th-st-subway-elevator-doubles-price-tag.html