In early 2022,
@alon_levy
,
@elifensari
, & I are going to hire someone, who can help us figure out how to build high speed rail along the northeast corridor for less than $100 billion. We want to learn from other countries and apply those lessons to the US. Get in touch!
Today we published our Phase 1 of the Second Avenue Subway case study. We show oversized stations (specifically back-of-house spaces), extractive negotiations with city agencies, a diminution of in-house capacity and embrace of consultants over everything drive costs and delay.
@ndhapple
Baseball does have natural breaks in the game, which are not driven by commercials. BUT, they probably could shorten some of them. Getting average game time down to 2 hours would be amazing.
Last week we published our final report re: transit costs. All of this follows from our finding (not unique) that the US pays more for transit projects than other countries in our database.
People have been beating up on VTA for crazy costs. $12.2 billion for 33,000 daily riders is the most expensive cost per rider I’ve ever calculated (~$370,000). That’s more than 10x SAS phase 1’s cost per rider in inflation adjusted dollars. More here:
Enjoyed this
@palladiummag
piece on why we can’t build anymore. This closely aligns with things we have found in our ongoing transit costs work. The major issues re: diminished in-house capacity and nimbyism have slowed schedules and increased costs
Some preliminary analysis by
@elifensari
of IBX's job access improvements. You'll never guess which stops have the lowest population density, lowest FARs, highest car ownership rates, lowest transit usage, etc., etc.
Work with
@alon_levy
and me at
@NYUMarron
. We are trying to figure out why it’s so expensive to build subways in America and how we can do better. We are building a database with costs and project scopes and doing in-depth cases. Apply here
$450M per mile for a surface light rail line. What a mess. Even with a new bridge this is criminal pricing. Portland built a surface streetcar loop for $45m per mile. 10% of the cost. WHY IS IT SO EXPENSIVE?!
Reading Sound Transit board documents. The decision to study a new station alternative will cost ~$800,000 in consultant fees. It will likely delay the project by a year, which, just in inflation terms could cost ~$750,000,000, assuming 5%. Pencils down!
While chatting with
@mattyglesias
about US transit costs, he asked how the US compared to the world. I hesitate to give a concise answer because of variation. Here's a chart we (
@elifensari
&
@alon_levy
) put on our site re: costs/km, tunnel %, & country that tells the story.
New York's Public Service Commission employed over 1000 engineers during the Dual Contracts era of subway construction in the 1910s and 1920s. The PSC was also responsible for utilities! This is a very different model than anything that exists in the US today.
I love what the MTA is doing. I think we are moving in better direction. We need to resolve these congestion pricing lawsuits and tender contracts. Finally, we need to focus on speeding up delivery. Instead 7 years, how do we build phase 2 in 4 years?
Here’s an example from the second Avenue subway. This is an open station with no support columns as far as the eye can see, though not unique to NYC. The design inspiration, I have been told, comes from visiting new builds from the late 1990s in Singapore, Hong Kong, and Bangkok.
In New York, at least in about 2010, for every tunnel boring machine a contractor used for boring a subway tunnel, it had to pay Local 147 $350,000. Second Avenue Subway Phase 1 used 1 TBM. In Istanbul it is required that contractors have 4, 6, or more TBMs
"The high costs of the 96th Street station can be partly
explained by the inclusion of 65,000 square feet for the
MTA workforce in underground facilities and office space,
requiring expensive blasting."
For a little less than a year I’ve been working with a group in Seattle to help speed up project delivery & restructure Sound Transit’s capital plan. This has been a great opportunity to apply our academic work into the real world.
Good to see
@ndhapple
shining a light on the problem of excessive station sizes. Larger stations lead to $$$ projects & longer construction timelines. Smaller stations, specifically in terms of non-passenger back-of-house space, is one path to that goal.
look i'm a boomer, with transition lenses, but this doesn't instill confidence: "Rather, designers will spend the next several months working with community members to envision a transit system that feels uniquely Austin." h/t:
@urbenschneider
Are you interested in transit-infrastructure projects? Are you interested in understanding what drives costs across cities/countries? Do you read foreign languages? Do you know what a database is? Are you interested in speaking to
@alon_levy
and me? We may have a job for you.
In an image VTA has perfectly diagrammed the cost issues we have been studying. A lot of excavated earth to accomplish a mid-sized building worth of fans and very generous underground circulation. The tracks appear secondary at best in this rendering.
"The 'hard' construction cost of an underground center platform station was estimated at $53,500,000. By contrast, the cost of an open-cut center platform station was estimated at $3,500,000."
Let’s start the rezoning push along the IBX corridor. If we are going to spend $5-6 billion, let’s allow housing and commercial spaces to be built around it to support ridership and ease rent burden. Need CMs to start thinking about land use citywide rather than in their district
“When existing bus routes are unreliable and slow, focusing attention on microtransit is like trying to perfect dessert at a restaurant that routinely burns the entrees.”
Lots of the proposed Amtrak services look good on a map, but these travel times & frequencies continue ongoing problems. Houston to San Antonio seems like a useful service, but at 4:45 by train versus 3 hours by car & 1 hour by plane, it's hard to see why anyone would take it.
“The more places which appeared to be getting service, the better the prospects for the proposition passing. As one member of the LACTC staff put it ‘basically you just take a small map of L.A. and a big felt tip.’”
Wrote about governance & transit costs for
@Noahpinion
. Always happy to get into the details of specific projects & cost categories, but the issue of governance, while vague, is critical to alignments, wages, third party agreements, design, & timelines:
Meant to post the proposed Union Square station versus the actual Union Square station. In the initial GLX plan, the average station cost was ~$60 million, in the revised plan it was under ~$20 million. The overall square footage across the 7 stations was reduced by 91%!!!
Wrote about some of the lessons we can learn from REM for
@TransitCenter
. Re: governance: Quebec passed a law that established timelines and consequences for local permitting decisions. In the US we don’t effectively operationalize time.
Love this LA Metro chart showing level of influence over projects & expenditures during project life cycle. The basic point is that the ability to get in front of cost and delay drivers is in the earliest phases. Once contracts are tendered, the game is over. ht:
@numble
If you don’t want to work with
@alon_levy
and me, but you want to share information & help us understand why transit-infrastructure costs vary from country to country, dm/email us. We need data, experts, & people with information from inside agencies, consultancies, & the trades.
2023 was exciting for the transit costs project team. In addition to releasing our final report, we have engaged with agencies, consultants, media, academics, researchers, and done a lot of zooms.
Some tweet streaming
@EnoTrans
newest report on international project delivery. Average costs for tunneled projects. If you exclude New York, the US numbers basically get cut in half
One reason why
@alon_levy
,
@elifensari
, and I have looked at transit-infrastructure costs is that there is no discernible connection between GDP and costs. Rich countries like Norway and Switzerland build cheaply and poorer countries like Bangladesh and Vietnam don't.
Pro state capacity, yes! How do we resolve regional vs. local tension? That seems to be the crux of the permitting debate. There are cities and towns that were created to be separate from the central city. When we plan transport networks we need regions to work cohesively.
America is suffering from a lack of bureaucracy -- by which I don't mean rules and red tape, but a competent civil service capable of getting things built cheaply.
We need to bring back the bureaucrats!
We were impressed by Istanbul's approach to minimize station digs by excavating smaller station boxes and stacking back-of-house facilities above the passenger areas. Istanbul manages to build similarly sized passenger areas with much smaller digs.
Now that Amtrak is about to receive a windfall,
@EnglishRail
& I look back at Amtrak’s first 50 years. Let’s not repeat the mistakes of the past, and let’s not assume intercity passenger rail is completely obsolete, as naysayers did 50 years ago:
We,
@alon_levy
&
@elifensari
, compiled a comprehensive tunneling staffing/wages sheet for Second Avenue Subway phase 1. A few things to note, tunneling ran from May 2010 to September 2011 so even at ~600k/month, this was a tiny fraction of the ~$4.5 billion project.
Megaprojects breed extraction. NYC DEP wanted bigger & more expensive pipes; NYC DOT wanted bike lanes, bus lanes, floating curbs, new street lights, and uniform paving, even in areas w/o construction; NYC Parks wanted cash, replacement trees, and funding for additional staff.
This is why we need to figure out how to build faster & cheaper. Phase 2 of the second Ave. subway is estimated to cost *$4 billion a mile*. Long term planning, smaller stations, greater MTA capacity, genuine partnership with the city and unions, & political support are vital.
Im all about the 14th street busway. Everyone seems to think that the 27,000 daily riders on 14th street deserve more reliable service--they do. HOW about the 27,000 daily riders on Church Avenue? More busways, please
@NYC_DOT
@TWULocal100
@NYCTBus
This is just a teaser of the great work
@elifensari
is doing to transform
@alon_levy
,
@YaoYinan_
,
@AnanMaalouf
, &
@abdirashidmd
's rich transit-infrastructure cost data into a website. Can't wait until we have finalized all the details & launch the site with +500 projects.
Yes, we have higher wages. BUT, we are also less productive & overstaffed. In our GLX case, the general contractor proposed 1 supervisor for every 1.8 laborers. Internationally, this ratio is closer to 1:5. This is why some sites appear to have more people watching than doing.
Did some back of the envelope estimating for the
@BlvdSubway
. If we could build at global averages it would be a ~$4.5 billion project. A big number, but reasonable for a 15-mile project. At US costs, it's likely closer to a ~$15 billion project, which is probably unfundable.
Definitely an exciting project with enormous benefits, even if $$$. Let’s lower costs by right sizing stations, scrutinizing fire codes, and aggressively managing utility interfaces. This is a challenging because of the tie-ins, let’s get some help from abroad.
NY Gov Hochul is asking the MTA to consider keeping the 2 Ave Subway tunnel boring machines going past 125/Park (the end point of phase 2), continuing to 125/Broadway, thus saving $400m on a future 3rd phase by completing the tunnels early. Seems like a pretty good idea!
We,
@elifensari
,
@alon_levy
,
@ChittiMarco
, compared LA’s Purple Line 1 & Istanbul’s M1B. This is our first attempt to compare platform length, construction method, financing, etc. Let us know what you think. check it out:
2021 marks the 50th anniversary of
@Amtrak
.
@EnglishRail
and I wrote about how most didn’t think Amtrak would last 5 years, let alone 50, and how schedules, frequencies, delays, and lack of autonomy continue to plague it for
@mattyglesias
’ slow boring:
Was walking past an ancillary facility for phase 1 of the second Avenue subway on 69th street. The emergency exit door was wide open. I have always wanted to see inside. If you’re like me, here you go:
“The Seattle Transit Commission had urged the Washington State Highway Department in 1953 to include a 50-foot median in the new freeway as a future rail right-of-way. The state rejected the rail plan.”
Excited by the prospect of an Interborough Express. Much has been made about connecting Brooklyn and Queens, but havent seen much about land use.
@elifensari
and I took a quick look at the travel and land-use data for
@replica
:
Since the 1990s, the capital construction arm of New York City Transit, CPM, has been dissolved & overtaken by MTA CC, and MTA C&D. CPM had 1,600 staffers who performed 95% of construction management and 60% of designs. MTA CC during phase 1 had ~120 full-time employees.
“To illustrate the accelerated pace at which spending is taking place, consider that China used more cement in the three years 2011-13 than the United States in the entire twentieth century.”
I was surprised to learn that more bus passengers in Brooklyn transfer to the bus (37%) than the subway (35%). This is a strong indication that service improvements, greater frequencies and reliability, should be foundational to any redesign.
I spoke to someone about GLX today, which reminded me of these old presentations re: station design and community character. I love that someone gave a presentation and explained community character with this image.
@the_transit_guy
This is the opening vignette of something I wrote a while back for
@Noahpinion
. Universities, UW, UMN, etc. all trot this out. I was told GOT was willing insulate several rooms at duke, but the university countered with, well what if we move the machinery?
Folks are skewering the GAO transit infrastructure study.
@alon_levy
& I via
@NYUMarron
have a research project ready to share with any foundations or wealthy individuals who want to uncover why US transit infrastructure costs are so high. Get in touch if you want more info.
Wonderful deep dive on high-speed rail. Eager to see it come to the US. I’m most interested in the planning and decisions re: improving bus, subway, and regional/commuter rail connections to ensure people can actually use these services when they arrive.
Excited to share our site, . We definitely need your help adding data, finding documents, understanding nuance, making corrections, and improving site stability! Please get in touch if you know about any of these things: info
@transitcosts
.com
LIRR fare integration would really help Queens out! Look at all those blue dots (half-mile from LIRR stops) within the yellow boundary. If i were an elected in queens this is what i would harp on re: congestion pricing. cc:
@IDaneekMiller
@drichards13
@CMPeterKoo
@LeroyComrie
@patrickc
@AidanRMackenzie
Isn’t the joke that SNCF was so disillusioned with CA HSR that it left for Morocco? I think SNCF said Morocco was less politically dysfunctional than California.
Anyone who has visited phase 1 stations knows they're spacious. What is less obvious is all of the space blasted out of Manhattan Schist that you don't see or even know exists.
"For too long, Congress has treated Amtrak as a pork barrel, dividing the fat among the most influential representatives. The result has been service so expensive and inferior that the very idea of train travel got a bad name."
Expensive labor and low productivity is a recipe for expensive projects. What if we could dig a launch box for the tunnel boring machine in six months or a year, as they do in Turkey and Italy, instead of the three years it took in New York?
The first thing we need to do is contextualize costs & schedule so we can talk across projects and in detail about specific projects and make good decisions about tradeoffs: do we accept longer work windows to speed up construction and reduce costs?
Given how many billions of dollars we’re throwing away on inflated infrastructure costs, it feels like a no-brainer for Pete to spend a few million to hire
@alon_levy
@ericgoldwyn
@ChittiMarco
@nilocobau
@elifensari
to start up an in-house transit costs consulting arm
I will be speaking a talk tomorrow. I will answer the age old question: how many people does it take to tunnel in New York? I will also reveal a small example about escalators that explains why stations are really expensive in New York. Maybe you will be hearing a listen?
Sweden has high cost labor, not as high as New York, but not so far off. It manages to build infrastructure cost-effectively by using a mobile, experienced work force more productively.
There is a role for consultants, no doubt. BUT, the MTA also has to lead and have the bandwidth to manage its projects effectively. Consultants complained that they didn't always have clear instruction on what they were supposed to do, and were expected to figure it out.
In New York and Boston, we show how internal capacity has diminished over time. When GLX began, the internal team consisted of 4 to 6 managers. In New York, the in house team managing these kinds of projects has fallen from ~1600 to ~120.
Great to speak with
@mattyglesias
about transit infrastructure, high speed rail, streetcars, and ongoing research with
@alon_levy
,
@elifensari
, &
@ChittiMarco
. Also, *very* exciting to e-meet The Weeds’ producer Erikk Geannikis. The other (E,A)r(r)i(c,k,ck,q,ct, etc.) understand
On this episode of The Weeds,
@mattyglesias
and professor
@ericgoldwyn
discuss why transit projects in the US often fail — including the Second Avenue subway line in New York, the Green Line Extension in Boston, and the DC Streetcar.
Did a long semi-rambling interview with
@GOVERNING
about our transit costs research. All of our key topics are here: politics & decision making, agency capacity, coordination w/3rd parties, design, labor, & procurement. More soon.
When transit projects are $$$, we get less transit per dollar spent. It also makes the marginal project look cost ineffective. Make no mistake Phase 1 delivered great value, but it was also the single best project in the city. How do IBX, Utica Ave, W ext. to Red Hook compare?
Now, we aren't all doom & gloom! We see hope in our low cost cases. In Italy, station standardization works. Stations, outside of Naples, aren't monuments. In Turkey, they have optimized station digs and use of underground space. Building shallow stations via cut & cover is key.
This is one of the most dramatic TOD turnarounds I've found in the US. MAX debuted at Orenco Station in 1998, and by 2007 the lots closest to the station were basically vacant. Much of the writing on this project predate the massive changes depicted here. How do I learn more?
We detail how stations in our Boston and New York cases made up a large share of project costs, and that they were designed beyond international norms we found in Italy, Sweden, and Turkey: more back of house space, full-length mezzanines, bespoke rather than standardized, etc.
A $2 billion grant is great, but even with that $$$, the CTA still can’t assemble all of the funding it needs for the red line: “The project is about $300 million short of being fully funded…”
Lawmakers are furious, but based on this, it seems that the real problem is a lack of local funding, which is solvable so long as lawmakers choose to solve it. Lawmakers: translate your fury into a funding commitment
This article really captures domestic transit projects: Non-subject-matter experts micromanaging station locations, parochialism vs. regionalism, missed connection, and delayed timelines: