Help Us Write More Construction Cost Cases
At the Transit Costs Project, we’re planning to return to writing more infrastructure construction costs cases, and we need help to improve our coverage to new places – potentially, your help. See job ad here.
How many positions are you hiring for?
This number depends on the applications we get, but there will be multiple hires.
What kind of cases are you interested in?
Cases studying construction costs for urban rail in a specific city or group of cities; so far we’ve done rapid transit projects (commuter rail counts), but we will look at proposals that look at adjacent issues, for example mass expansion of light rail if it is at sufficient scale to qualify as a megaproject. We’re not going to be interested in highway-only or highway-focused studies, but a public transport-focused report could still talk about a contemporary highway megaproject in the same place to give more context – for example, at the same time that it built more than 100 km of new metro in a decade, Madrid also built motorway tunnels, and this may be relevant to understanding its decisionmaking, engineering practices, and contracting practices.
As we say in the job ad, you should be familiar with our existing cases, which can be read here. This doesn’t mean you need to memorize them (I have not memorized the one I wrote myself, let alone the others). But they should give some guidance about what issues we’re interested in seeing in each case: a sketch of the history of urban rail in the city, an explanation of the decisionmaking that led to the project, a description of the ongoing discourse about it informed by local opinions on how things should be done, engineering drawings, and similar issues leading to a synthesis of what works and what doesn’t.
Do the cases need to agree with the program focusing on decisionmaking, project delivery, and so on?
They need to be on-topic but absolutely do not need to agree with us. I’m fairly hardline against certain procurement elements that we call the globalized system in the cases, such as large contracts and design-build delivery, but also if you think I’m wrong and set out to find a case showing that design-build has actually worked and improved things, that’s entirely on-topic. Marco Chitti came to my attention after he vociferously disagreed with me on Twitter about private competition in intercity rail in Italy.
Which places are you interested in seeing reports for?
We’d like to have a wide variety of examples. So far we’ve done two American cases (Boston and New York), plus Italy, Stockholm, and Istanbul. Potentially, any example of urban rail megaproject elsewhere qualifies; that said, all other things being equal, we would prefer environments that are different from all of those, such as medium-cost places (like France or Germany), Asian examples of any kind, Latin American examples (cheap like Santo Domingo or Santiago, or more expensive like Brazilian cities), the United Kingdom with its relatively recent cost explosion, and so on.
This also, frustratingly to applicants, means that we’re going to hire based on who else we’re going to hire. What this means is that if we decide to hire three people, and then we narrow the shortlist to five good applicants with good proposals for a Paris case and then two more with good proposals for Tokyo and Santo Domingo cases, we’re not going to do two Paris cases.
What connections do I need for the interviews?
Any. You don’t need to have a list of 20 names of people to talk to in advance, but we would like to see evidence that you can get to this many interviews. This can come from connections with local advocates, politicians, civil servants, academics, contractors, union organizers, journalists, agency heads, engineers, or policy experts. Interviewees often naturally suggest other people to interview or documents to read, whose authors are then natural interview subjects; this in our experience has included both disgruntled planners eager to publicize the failures of the projects they’ve been involved in and planners who think their superiors are doing good work and would like to publicize their agencies’ successes.
I think I have enough knowledge and connections in multiple countries and can do two cases. What do I do?
Talk to us; it’s likely we’ll only be able to make you an offer for a single case, but you never know.
(Update) Where is the job?
We work fully remotely. I met Elif in person two years into the Transit Costs Project and Marco even later.
Pedestrian Observations 
I know why you don’t want to, but I’d like to see this study done for highway mega projects. (one might exist that I’m not aware of) Is the US bad at megaprojects in general, or is it just (mostly?) transit projects that blow up costs is the question I’d like to get answered.
https://en.wikipedia.org/wiki/State_Route_99_tunnel is $1bn/km and https://en.wikipedia.org/wiki/1915_Çanakkale_Bridge is just under $1bn/km.
Given Turkey is supposed to be cheap, I would say the Alaskan Way Tunnel is a good deal.
Well, Alaskan Way Tunnel was a bad deal in that a second freeway through downtown Seattle didn’t need to be built to begin with.
Not even mega. Single public buildings e.g. libraries in New York cost 4-8x per square foot of floorspace what similar-sized office buildings cost in other parts of the US. Source. For that matter, much of the Anglosphere (to name: US, UK, Aus, and within Canada Toronto and Vancouver but not Montreal) famously has problems with a quantitative inability to build housing in growing large cities. The latter, not being public projects, are even less relevant to Transit Costs, though political aspects are shared — e.g. the harmful attention paid to building a block of flats if that looks like a “generational project”.
Other categories of public megaprojects with their separate analyses:
– Nuclear power plants. Here a single item of their US regulation is identified as particularly perverted: the regulator is to assume that any cost-per-kWh benefit over coal-fired plants should be actively used up on improving safety, no matter the existing safety record.
– Bulk powerlines, especially ones crossing state borders. However, even some single states (e.g. California) are so bad at this that allegedly the idea to transport energy by …unit trains of batteries… could pencil out if existing freight railways have the spare line capacity. (“Hey, it worked for coal.”)
– Pipelines.
– Wind farms, especially offshore.
– Dredging, subject to the Foreign Dredge Act.
– Ship construction, possibly partially excepting exactly the most complicated categories (e.g. nuclear submarines), which is counterintuitive if true. In any case, surface ships have either come out of procurement as abominations (LCS) or as several times more expensive than international comparisons and extremely delayed (frigates, Coast Guard icebreakers). Civilian-customer-oriented shipbuilding is within a rounding error of not existing (see any discussion of the Jones Act).
– SLS, the Orion capsule, and some of the related lunar program. Particularly egregious is the Gateway: a comedy of errors has lead to the situation where some component systems lack sufficient performance to do the obvious (and overall efficient) thing. Rather than reworking the systems to be more capable, the decision was to: 1) add a freaking space station into the critical path of the project (which is here because it already has management issues), 2) add a detour, increasing total performance required (hence the derisive nickname “Tollbooth”) but bringing the individual hop within the capability of the Orion-atop-SLS system.
You know what I’d say about that …
But seriously it raises the question that the causes of the difference must be fairly obvious to pinpoint when essentially the same thing is constructed within the same overall jurisdiction at the same time and sometimes by the same companies. Back in Europe it is the same/similar thing for the nuclear EPRs being built at Hinkley Point UK and in Flamanville, France. Essentially it is the same company building all three reactors, France’s state-owned EDF (which includes Areva the designer of the EPR). Costs in both countries have skyrocketed compared to initial estimates (the Hinkley costs are for twin reactors):
versus in France (a single EPR):
At minimum this is a 100% premium and up to 216%, per unit, for building in the UK. I’ve never read an analysis to attempt to explain this difference.
This is interesting: https://bcis.co.uk/news/european-construction-costs-compared-bcis/
As is https://www.ft.com/content/9cc19ce5-fbdb-4285-80ac-498f01f97dfd according to what showed up in my search.
For that matter, much of the Anglosphere (to name: US, UK, Aus, and within Canada Toronto and Vancouver but not Montreal) famously has problems with a quantitative inability to build housing in growing large cities.
Growing US cities like Raleigh, Charlotte, Austin and Dallas have no problems building enough housing.
House prices on average in the US are up 50% over 2019 vs 18% here. I think that is a big part of why Kamala lost. It’s huge.
And yeah people vote on stories not cold economic data, but anyone who has bought a house in the last two years will feel poorer.
Right; Calgary and Edmonton are likewise building rapidly enough to keep prices down. Some additional qualifier is called for. “Old large”? “Any even slightly interesting geographical constraint”? — probably not, Toronto barely qualifies with being merely cut in half, and in particular London only has one forbidden line segment downstream.
In the cities with a problem, housing prices have been unreasonable for a long time. By 2018, the phenomenon was established practice and fairly widely talked about (I don’t know off the top of my head when the YIMBY and market urbanism movements congealed). Heck, the 2008 recession was at the time described as a housing bubble popping.
People buy their first house once in their life. The Democrat’s plan for first time buyers was a $25,000 subsidy. The Republican plan was a secret. Because Dear Leader will heal all. You don’t need details.
Votes are still being counted. 16 million fewer people voted in 2024 than in 2020. Not-showing-up means you don’t care about the cost of housing, food, etc. Or reproductive health or corrupt judges or that people who beat up cops will be pardoned. Or anything else.
House prices now average $500k in the US, in late 2020 they averaged $330k. Adjusted for inflation that would be $400k now, and for wages $412k.
A $25k discount for first time buyers isn’t awful, but a) it stokes demand even higher and b) it isn’t really enough to touch the sides – it would need to be ~$90k to do that.
What part of “don’t care” didn’t you understand? Most people don’t care because they didn’t vote or voted for Dear Leader will fix everything.
The Federal Reserve disagrees. https://fred.stlouisfed.org/series/MSPUS
$420,400 for the third quarter of 2024.
And it doesn’t matter. Because the majority of people voted “I don’t care because I didn’t vote”. And the slim majority of people who did vote, voted for “It’s doubleplusungood to question Dear Leader. He alone can fix it.”
Sunbelt cities aren’t really cities. They can build because they have no or little existing urban patterns to worry about.
NIce round numbers:
Metro Raleigh has the population of Bronx. The Combined Statistical Area has the population of Queens. It was real easy to build lots of housing in the farmlands of Brooklyn as the elevated wended out there. And real easy to build lots of housing in the farmlands of Queens a few decades later. I’m an old fart who remembers when the A&P had Staten Island potatoes. Staten Island isn’t potato fields anymore. The last farm in the city was converted to a museum many years ago. The combined statistical area of Charlotte is a bit bigger than Brooklyn. Austin is bit bigger than Queens. it’s real easy to build housing on farmland that isn’t too far from downtown. Or the beltway.
The Dallas-Fort Worth CSA is almost as populated as New York City. Since they decided to splatter single family houses almost exclusively, all over means they choke on their own traffic. It’s what they voted for. They can sit in traffic.
Density is what defines urbanity, not population.
If you don’t have population you have undeveloped land nearby that makes it easy to build housing. Once someone builds something on it, it is much more difficult to build more on it.
Vancouver builds more housing than Montreal. The Montreal exceptionalism is that rents there are lower, because the city is poorer, not that it builds more housing.
This is very true. Quebec was historically among the poorest provinces, but for Montreal specifically french language laws of Quebec and the political uncertainty associated with separation have been major forces pushing businesses out of province.
That is such a neocon and econocratic attitude! “Poorer” by how much? And if one accounted for life’s biggest expense, housing, is it still true? It is very close to impossible to quantitate because it comes down to qualitative issues and personal choice on things such as housing and transit and lifestyle. Your claim that french language laws of Quebec and the political uncertainty is speculation. It hasn’t altered that Montreal remains Canada’s second city (4.3m) and one of the biggest in North America, quite an achievement for a French-speaking enclave surrounded by the world’s largest Anglo population, in a ratio of almost one to 50.
Also, as an example of how little value lies in such monetary comparisons, the UK still has a larger GDP than France, although France is higher on per capita and PPP basis. Paris is expensive but actually still a lot lower than London. But get out of Paris into wonderful secondary cities (Lyon, Montpelier, Toulouse, Bordeaux etc) and the difference with anywheresville-UK is huge. And that is on an econocratic perspective, never mind the quality of life. Lots of people, especially Anglospherians would conclude the UK is clearly superior, but like so many things they’d be wrong.
It’s not even clear if your primary assumption is correct:
Oh, and Australia is one of the “richest” nations in the world (ca. 22% higher GDP per cap compared to both UK and France), yet despite being one of the largest land masses it now has some of the most unaffordable housing in the world. It has been run as a Ponzi scheme for about 4 decades now, but it was never sustainable and now a lot of next-Gen (today’s genZ and later) have zero prospect of owning their own house and rents are absurd.
So I am not the least bit convinced Montreal is ‘poorer’ in all the things that really matter.
Among others, RBC and Bank of Montreal moved their HQs to Toronto in the 1970s for this reason, and CP rail moved to Calgary in 1996 for this reason.
There is a huge literature on how businesses don’t like political uncertainty.
More directly, Montreal is *less desirable*. Few jobs compared to Toronto, bad weather compared to Vancouver, unfamiliar language. People don’t want to move to Montreal, so there is no upward pressure on rents, compared to Toronto or Vancouver.
Poor people and rich people are more or less the same size and take up the same amount of space in the housing whatever it is. Builders whether they are building for poor people or rich people don’t build stuff for the hell of it. There have to be buyers and since Vancouver is growing faster than Montreal it’s going to have more housing built.
Quite right. And V has a supercharged property market, not particularly for any Uber-entrepreneurial reasons etc but because it was the most amenable place for Hong Kongers (and mainlanders) to create their refuge homes and resting place for their capital. Not only on the same Pacific Ocean but also much milder climate. They are hardly going to choose Montreal, or even Toronto, over it are they? But it still doesn’t necessarily mean anything deeper. However I suspect those Hong Kongers would feel more at home on the Montreal Metro versus Vancouver (and no, I don’t mean rubber wheels!).
For me, I can’t really say for sure since I have only too briefly trudged thru the 2m snowdrifts of Montreal … but isn’t tout le monde, especially a few days after that election, just a bit weary of relentless idiocy and single-minded focus on money and whinging and “winning” of the Anglosphere? Speaking of lowest common denominator shit, dinner parties here and no doubt in London, New York and Vancouver are all about real-estate … (or college fees).
So, Alon and Weifeng Jiang and others here, I am seriously over that kind of narrow econocratic thinking and practice. I mean, FFS the top two countries exemplifying that kind of “thinking”, the UK and USA have been skirting failed-state status and have the shittiest rail/transit in Europe and the world, respectively. And amongst the most expensive (so much for ‘efficiencies’).
@Michael, the hard right is doing better in France than Britain. Farage isn’t polling at 30-40% here or anywhere close – https://en.wikipedia.org/wiki/Opinion_polling_for_the_2027_French_presidential_election
Give it a rest. You’ve just come out of 15 years of the worst government in modern history, not just dysfunctional but with extremist policies like nothing in any of its peers. So far the Far-Right in France has been politically contained. They only have a high vote due to the run-off system which is used by voters as a protest. They have very little power while Farage hauled you out of the EU! Even the RN have removed their Frexit policy in the light of what everyone has seen with Brexit.
Vancouver would be very expensive without any immigration at all. Canada has the same sort of weather related movement dynamics as the United States. Vancouver may not be sunny Los Angeles, but for most people, the rainy winters of Vancouver are more pleasant than the freezing cold of the Prairies. Besides that, it’s geographically beautiful, has excellent transportation, and has strong entertainment amenities.
I can’t because I’m not logged on to the octopus like tentacles of Google.
I just tried opening it in private browsing, where I’m not logged in, and it did work.
I wish that I were in a position to and even more that I were competent to do this. I don’t know if things that were not built are within the remit of the Transit Costs Project, but I would very much like to know how SEPTA’s King of Prussia Rail grew from a project estimated to cost a few hundred million dollars to one that was expected to cost three billion before it was *achem*, “paused.” (Heaven help us if there’s also reason to ask what happened to the costs of Trolley Modernization)
https://www.constructiondive.com/news/brent-spence-bridge-lawsuit/731204/