How One Bad Project Can Poison the Entire Mode
There are a few examples of rail projects that fail in a way that poisons the entire idea among decisionmakers. The failures can be total, to the point that the project isn’t built and nobody tries it again. Or the outcome can be a mixed blessing: an open project with some ridership, but not enough compared with the cost or hassle, with decisionmakers still choosing not to do this again. The primary cases I have in mind are Eurostar and Caltrain electrification, both mixed blessings, which poisoned international high-speed rail in Europe and rail electrification in the United States respectively. The frustrating thing about both projects is that their failures are not inherent to the mode, but rather come from bad project management and delivery, which nonetheless is taken as typical by subsequent planners, who benchmark proposals to those failed projects.
Eurostar: Flight Level Zero airline
The infrastructure built for Eurostar is not at all bad: the Channel Tunnel, and the extensions of the LGV Nord thereto and to Brussels. The UK-side high-speed line, High Speed 1, had very high construction costs (about $160 million/km in today’s prices), but it’s short enough that those costs don’t matter too much. The concept of connecting London and Paris by high-speed rail is solid, and those trains get a strong mode share, as do trains from both cities to Brussels.
Unfortunately, the operations are a mess. There’s security and border control theater, which is then used as an excuse to corral passengers into airline-style holding areas with only one or two boarding queues for a train of nearly 1,000 passengers. The extra time involved, 30 minutes at best and an hour at worst, creates a serious malus to ridership – the elasticity of ridership with respect to travel time in the literature I’ve seen ranges from -1 to -2, and at least in the studies I’ve read about local transit, time spent out of vehicle usually counts worse than time spent on a moving train (usually a factor of 2). It also holds up tracks, which is then used as an excuse not to run more service.
The excusemaking about service is then used to throttle the service offer, and raise prices. As I explain in this post, the average fare on domestic TGVs is 0.093€/passenger-km, whereas that on international TGV services (including Eurostar) is 0.17€/p-km, with the Eurostar services costing more than Lyria and TGV services into Germany. This includes both Eurostar to London and the services between Paris and Brussels, which used to be called Thalys, which have none of the security and border theater of London and yet charge very high fares, with low resulting ridership.
The origin of this is that Eurostar was conceived as a partnership between British and French elites, in management as well as the respective states. They thought of the Chunnel as a flashy project, fit for high-end service, designed for business travelers. SNCF management itself believes in airline-style services, with fares that profiteer off of riders; it can’t do it domestically due to public pressure to keep the TGV affordable to the broad public, but whenever it is freed from this pressure, it builds or recommends that others build what it thinks trains should be like, and the results are not good.
What rail advocates have learned from this saga is that cross-border rail should decenter high-speed rail. Their first association of cross-border high-speed rail is Eurostar, which is unreasonably expensive and low-ridership even without British border and security theater. Thus, the community has retreated from thinking in terms of infrastructure, and is trying to solve Eurostar’s problem (not enough service) even on lines where they need competitive trip times before anything else. Why fight for cross-border high-speed rail if the only extant examples are such underperformers?
This dovetails with the mentality that private companies do it better than the state, which is dominant at the EU level, as the eurocrats prefer not to have any visible EU state. This leads to ridiculous press releases by startups that lie to the public or to themselves that they’re about to launch new services, and consultant slop that treats rail services as if they are airlines with airline cost structures. Europe itself gave up on cross-border rail infrastructure – the EU is in panic mode on all issues, the states that would be building this infrastructure (like Belgium on Brussels-Antwerp) don’t care, and even bilateral government agreements don’t touch the issue, for example France and Germany are indifferent.
Caltrain: electrification at extreme costs
In the 2010s, Caltrain electrified its core route from San Francisco to Tamien just south of San Jose Diridon Station, a total length of 80 km, opening in 2024. This is the only significant electrification of a diesel service in the United States since Amtrak electrified the Northeast Corridor from New Haven to Boston in the late 1990s. The idea is excellent: a dense corridor like this with many stations would benefit greatly from all of the usual advantages of electrification, including less pollution, faster acceleration, and higher reliability.
Unfortunately, the costs of the project have been disproportionate to any other completed electrification program that I am aware of. The entire Caltrain Modernization Project cost $2.4 billion, comprising electrification, resignaling (cf. around $2 million/km in Denmark for ETCS Level 2), rolling stock, and some grade crossing work. Netting out the elements that are not direct electrification infrastructure, this is till well into the teens of millions per kilometer. Some British experiments have come close, but the RIA Electrification Challenge overall says that the cost on double track is in the $3.8-5.7 million/km range in today’s prices, and typical Continental European costs are somewhat lower.
The upshot is that Americans, never particularly curious about the world outside their border, have come to benchmark all electrification projects to Caltrain’s costs. Occasionally they glance at Canada, seeing Toronto’s expensive electrification project and confirming their belief that it is far too expensive. They barely look at British electrification projects, and never look at ones outside the English-speaking world. Thus, they take these costs as a given, rather than as a failure mode, due to poor design standards, poor project management, a one-off signaling system that had very high costs by American standards, and inflexible response to small changes.
And unfortunately, there was no pot of gold at the end of the Caltrain rainbow. Ridership is noticeably up since electric service opened, but is far below pre-corona levels, as the riders were largely tech workers and the tech industry went to work-from-home early and has still not quite returned to the office, especially not in the Bay Area. This one failure, partly due to unforeseen circumstances, partly due to poor management, has led to the poisoning of overhead wire electrification throughout the United States.
Pedestrian Observations 
“The frustrating thing about both projects is that their failures are not inherent to the mode, but rather come from bad project management and delivery, which nonetheless is taken as typical by subsequent planners, who benchmark proposals to those failed projects.“
Arguendo, this is the system working exactly as it should? Since we are talking about a system’s first project of this kind, an experiment if you will, by assumption the system doesn’t have a detailed (“mature”) understanding of this project type, which is what it would take to attribute a failure to extraneous causes. Putting it differently, randomly grabbing one datum from the set is an unbiased estimator.
Furthermore, I would argue that the estimates (so to speak) are actually in the right ballpark! If I read this right, then Caltrain electrification cost ~10x European electrifications …just as every single American project costs ~10x corresponding European projects, no? Thus for the purposes of the system internally deciding whether to build [proposal], it does exactly the right thing by benchmarking the cost against Caltrain — whereas of course, the same cost should be flagged as bonkers for the purpose of evaluating how the system works, or what would be reasonable from the point of view of a different system. (“It is an abomination beyond human comprehension; if someone looks at it too attentively, they go mad and start raving, like Richard Mlynarik.”)
To elaborate on this further, the phenomenon of declining project benefit-cost ratios with subsequent projects should apply in the usual way. The fact that, as you pointed out, Caltrain was plausibly the best test case for electrification shows that the system was roughly correctly ordering potential electrification projects, descending by benefit-cost ratio as-achievable-by-the-system.
Likewise, most EU cross-border HSR would involve SNCF (and/or DB).
Also, obligatory xkcd.
“The excusemaking about service is then used to […] charge very high fares, with low resulting ridership“
Allegedly the actual airlines had a similar problem. They treated Concorde as definitionally above-first-class, priced tickets to match, and consequently had remarkably poor seat utilization. It was only after the decision was made to discontinue Concorde that a different faction(?) could run an experiment, selling the seats at prices comparable to business class, thereby filling the planes and earning more revenue per flight.
“the EU is in panic mode on all issues“
Can you expand what you mean by this?
I am guessing you are referring to Richard Branson’s offer to take over Concorde when they announced they were discontinuing it. Was he disingenuous? Poking the bear when he was at peak animus with BA? Hardly matters as I believe it was a French decision (they never had made as much as success as BA did with the London-NY & DC flights) and the original agreement called for both, or none, to operate.
Or Club Concorde which was to run charter flights to the Arctic or wherever, which BA had been doing anyway.
Part of my job involves creating estimates for engineering projects. (Not rail, but bear with me.)
The most accurate source of data is historical actuals, in other words actual costs incurred. However, the data source has to be recent enough, similar enough and suitable enough to be used as a historical data basis. If it isn’t, or the project that is being estimated significantly diverges from the historical project, the data set must be reduced or a factor applied (with substantiation) to make the historical project data applicable.
I’m vastly oversimplifying here, but let’s pretend like I’m going to use Caltrain to estimate New Haven-Hartford-Springfield. Caltrain scope not applicable in ( ), mods in [ ].
Assumptions Both lines are fully double-tracked (this is not true for NHHS, but close enough) / NHHS will use existing M8 EMUs and CT maintenance facilities
Caltrain scope
Material: rolling stock, catenary structure, overhead wire, substations x 80km/51mi, mains interconnect, rail storage yard, maintenance facility, maintenance tracks, X bridge and Y tunnel modifications, bespoke PTC signaling
Labor: construction hrs x CA prevailing hourly union labor, engineering design, permitting, project management, environmental impact x CA prevailing salary labor, % consultant offload for professional services
= $2.4B
NHHS scope
Material: (rolling stock,) catenary structure, overhead wire, substations x [100km/62mi], mains interconnect [not sure about this – one terminus is electrified], (rail storage yard, maintenance facility, maintenance tracks), [Springfield layover yard], [X] bridge (and Y tunnel) modifications, [unified New Haven Line/Amtrak NEC signaling]
Labor: [construction hrs] x [CT prevailing hourly union labor], [engineering design, permitting, project management, environmental impact] x [CT prevailing salary labor], [% consultant offload for professional services]
= ~$350M or somewhere in that neighborhood
Is this what drives the conversation? No. What will likely happen because no one has the attention span, or is looking for an excuse to throw the project in the garbage, is $2.4B x (100km/80km) = $3B. At which point policymakers’ eyes glaze over and it’s done.
This is before we even get into “I just (stupidly) spent $310M on unpowered coaches*, $60M on diesel-electric dual-mode Chargers, and $25M per battery-diesel Charger x MTA usage share x #locos, and now you want me to spend $3B more?” They won’t even get to “Hey, cancel that nonsense first. Pay the termination fee. It’ll still be cheaper.”
And now? The tariffs? God help us all. We’ll be making catenary structures out of Maine white pine.
Alon’s point is a good one. As a result of exceptionalism and incuriosity, American rail delivery organizations will likely not consider projects in Europe or Asia as similar enough or suitable enough to base cost estimates on, and ignore the cheaper and in many cases more accurate benchmarking that comes with it.
* from a garbage vendor who’s 5 years late on Avelia Liberty and is withholding product support from Acela I to make up for eating losses on that contract, forcing Amtrak to cannibalize 4 out of the 20 trainsets for spares
Actual trains, even though they offend your delicate sensibilities, are lot better than not having any. If there are no overhead wires the electric trains can’t go very far.
There isn’t any conversation. Railfans love to imagine all sorts of things including that Connecticut will do all the stupid things California did. Connecticut DOT is less frothy and seems to be doing the study to determine whether a study should be done. Expect progress by 2060 or so
https://portal.ct.gov/dot/ctdot-press-releases/2023/feasibility-study-of-electrifying-hartford-line-and-danbury-and-waterbury-branch-lines-underway?language=en_US
Bad troll! Shoo! Back under your bridge!
Unfortunately, just as there is a housing crisis for humans, reduced infrastructure construction has lead to a spate of troll shelterlessness.
(sorry)
More serious (albeit whinging): I live in Hungary. Our recent rail projects look like:
There are plenty of hollow trees up here in the mountains.
That doesn’t change that diesel trains are better than no trains. Or that dual mode locomotives changing modes is better than changing locomotives.
I’m going to make this crystal clear to anyone reading your comment (not you, because you’re a troll) who thinks you might have a point (you don’t):
The state had 3 rational choices here: 1) do nothing, and continue to pay high o&m costs for the existing fleet of locomotives and coaches, 2) buy new rolling stock, or 3) electrify the line. Option 3 would have then required a subsequent choice of using existing M8s or buying additional EMUs
The correct choice here would have been option 3 with using existing M8s. The performance metrics of this choice fare comparably or better with all other choices with respect to capital spend, o&m, reliability, customer satisfaction, and potential service expansions.
Instead, the state decided to be poor stewards of taxpayer money and picked option 2, then compounded its mistake both by overpaying and picking the least efficient choice for a commuter or regional service (DM locomotives and unpowered coaches). The details are in the transit costs project’s rolling stock database and in a previous post on this blog.
Perhaps people who think Real Americans(tm) drive everywhere and trains are a communist plot to sap and impurify theie precious bodily fluids, like to point at it. Americans, outside of a few metro areas, don’t care about trains because there aren’t any.
Americans who do have electric trains, extend service now and then.
https://en.wikipedia.org/wiki/Montclair_Connection
I don’t know or care if the people of Upper Montclair are still objecting to running trains on the weekend. Something about how trains running in quintessential railroad suburb would have their suburban splendor disturbed if there were weekend trains. Gawd only knows in what decade they will restore service to Andover. Trains to Scranton by 2317?
SEPTA has been up to something too.
https://en.wikipedia.org/wiki/Wawa_Station
This reflects more about Alon’s views of “French elites” than the reality. The silly state of Eurostar today is different to how it began. When I first used it, the custom clearance at the British end was no problem and was very fast. It is only the Brexit mentality that has screwed it up; its origins is the pre-Brexit paranoia about unregulated migrants and refugees though one wonders who imagined thousands or tens of thousands somehow evading French immigration control to board Eurostar. Of course it was all performative when Tories were running the show. Nothing to do with French, elites or plebs.
Also nothing to do with French elites was the high cost. This is directly attributable to Thatcher who insisted not a cent/pence of government money go into the tunnel venture. Groupe Eurotunnel was loaded with €8bn of private debt when the tunnel opened. Through bankruptcies (caused by unprofitability due to the huge interest payments on that debt) and restructuring there have been some reduction of the debt, for example by converting it to equity but the debt remains about €5bn. Even though SNCF now has majority control it would take a very generous/brave French government to decide to wipe the debt. SNCF itself is not going to do it, except that further restructurings etc that the Thalys merger might allow. Then there is the Covid effect which it still hasn’t recovered from though largely because of the fiasco of slow immigration control in St Pancras causing reduction in number of trains running, despite demand.
Be that as it may, I doubt it has much real influence on future thinking on international services because, despite all this, both Eurostar and the freight and car services are considered successes. I don’t know what Alon would or could do to overcome the fundamental problems, political (British not French elites) and financial.
Well, there was talk about running a Ouigo like service between Stratford and Interconnexion-Est at CDG (hah, those elites again, trying to fool customers they are actually flying!) which doesn’t sound bad and could allow both higher pax numbers and lower fares, without impacting much at all on the elite services from St Pancras-Nord. I’d consider using it myself as long as the overall cost, with transit at both ends, remained low and they didn’t force you to book weeks or months ahead.
Michael, I agree with your post but you confuse Eurotunnel, now a GetLink subsidiary, and Eurostar. Eurotunnel’s debt is indeed close to 5 billion Euros. No government, and certainly not the very indebted French one, will wipe out this debt and get sued by the ferry companies for doing so. Getlink appears capable of paying its debt before its concession ends in 2086.
Eurostar, in which SNCF now has majority ownership, is only a large GetLink client. Its own debts of 650 million Euros are manageable given its 2 billion Euros annual revenue.
The headwinds faced by Eurostar and would be competitors will only marginally fade away. HS1 and Eurotunnel were costly endeavors. HS1 concession, in private hands until 2040, charges 7600 Pounds (8708 Euros) for a 109 km St Pancras-Folkestone Eurostar run. Alon’s assertion that HS1 “is short enough that those costs don’t matter too much” and that Eurostar is profiteering don’t make sense. For years, GetLink has been trying to find new clients, so far without success because Eurostar operations are only marginally profitable. HS1 toll is higher than the discounted one charged by SNCF Réseau on the 380 km between the Tunnel and Gare du Nord, and the Eurotunnel passage is even more costly at about 20 Euros per passenger.
Alon’s “Europe itself gave up on cross-border rail infrastructure” is even more absurd. The EU is financing 50% of the Brenner base tunnel. It has promised to pay 40% of the Mont Cenis base tunnel and the 700 million Euros it sent in 2024 are enough to finance the works up to 2026.
This is a big part of the problem. As you know their concession was extended to that date to alleviate the impact of high interest charges on the debt, and rescue Getlink from bankruptcy. All of which flow into charges on Eurostar and the freight operators; ie. the high charges remained while Getlink’s costs reduced. The only way to reduce them is to get a lot more pax thru Eurostar or potential competitors (or alternatives like Ouigo) or make more profit thru something like the joint ownership with SNCF of rail freight operation Veolia Cargo. Though I suppose since its debt repayment is now “under control” any excess profits will flow to its shareholders (largest shareholder is private but do you know if there is any government ownership?).
Michael, GetLink is the dominant player on the Channel passenger market. Between Paris, Brussels and London Eurostar carries more passengers than the airlines. GetLink Shuttle has a 62% market share for private automobiles. It is not doing as well with freight. More trucks (64 to 36%) use ferries than GetLink. DB Cargo stopped sending trains one year ago, and GetLink’s own rail freight hauler Europorte is not yet using the tunnel.
These are neither the numbers of an operation geared towards the elites nor numbers that can easily be improved through lower charges. Hedge fund managers who disagree could launch a buy-back offer. GetLink’s market capitalization of $10 billion is higher than a few years ago but a small number for financiers around the world.
Rating agencies classify GetLink as speculative BB+ business. I think it is a solid one but not a money-printing machine like SNCF’s LN1 to 5. HS1 and Eurotunnel bankruptcies showed that Margaret Thatcher was wrong and that infrastructure projects with high upfront costs and long term benefits were beyond private investors horizons and patience. The EU understands that and is willing to finance such endeavors if they align with its long term goals of European integration and decarbonation. In my opinion, HS2 and CAHSR troubles show that some degree of private investor discipline is also needed.
Eurostar ran 15 trains today to Paris and 10 to Brussels and pays a total of £48m/year (February 2023 prices) to High Speed 1 (https://www.orr.gov.uk/sites/default/files/2025-02/2025-02-11-periodic-review-of-hs1-pr24-final-determination.pdf) in both fixed and variable costs, assuming that train load is typical (and there are 25 trains in the reverse direction as well) that is £2630/train (or £2750/train today which is €3150).
But those fees as far as I can tell include their fixed and variable costs, so new extra services would attract a somewhat lower fee.
I do agree the fee is to be fair high at €30/km.
The ORR is only counting the regulated access charges, which it regulates. Congratulating itself on lowering them by 3.8% the ORR doesn’t seem to notice that these charges are still very high. In the final determination you quoted, they amount to 126.7 million Pounds per year, i.e. 1.16 million Pound/double track km. Annual HSR maintenance costs are currently about 100 000 Euros/km in Spain, the country with the highest costs in a 2005 study comparing it with Belgium, France and Italy. Eurostar and would be competitors face other unregulated charges to recover HS1 long-term costs as stated in the ORR report page 9 paragraph 2.4. These should explain the discrepancy with the 7600 Pound number cited April 4, 2025 in The Guardian by Gwyn Topham and in Skiflightfree.org.
Alon, my reply must be stuck in spam filter.
Network rail more generally is spending £145k/year/km on maintenance and £250k/year/km on renewals – so I would be surprised if those figures include money into a sinking fund to cover track renewals which to be honest they probably should.
That said I think you can also see the cost savings from having fewer tunnels etc – another downside on how they have done HS2 I think.
Agreed. I think £7600 is about right including covering the construction costs.
But I do believe HS1 has offered discounts for new services which also shouldn’t have to make a contribution to the construction costs if otherwise they wouldn’t run.
It seems the ORR access charges cover maintenance and renewal. In Table 1.3, the 126.7 million Pounds total regulated income results from adding 90.1 m for Route Operations & Maintenance, 28.1 m for Route renewals and 8.5 m for Stations renewal.
I don’t think the 7600 Pounds cover the entire construction costs. The UK government did not set the 2010 sale price on a cost basis. It looked at what various consortia were willing to pay for a 30 year concession and chose the offer from two Ontario public pension funds over bids from consortia including Eurotunnel or BT pension fund, Allianz and another Canadian public sector pension fund. In 2017, the Ontario pension funds who had invested about 2.1 billion sold the remainder of the 30 year concession to Korean and British investors for 3 billion Pounds. HS1 will be returned to the UK when the concession is over. Whether that will result in another lease auction or simple public ownership is yet to be decided.
Still stuck in the spam filter. Here (below) is the essence without any links.
Applied to the Tunnel it would have meant the UK and French governments covering about €3bn each, 75% of the total, leaving only 25% for Groupe Eurotunnel to recover over a reasonable concession period (not the century or whatever they have been given). Rates for Eurostar and the freight operators would have been much more reasonable and ridership would be higher etc etc.
This kind of thing was anathema to the Thatcherites. However I note that HS1 cost £5.8 billion [€6.7bn] but I doubt that the includes the rescue fiasco of the attempted PFI (=PPP) (various real estate speculators who couldn’t pull it off which necessitated a public bailout). Then in a few short years that company sold the 30y concession for only €2.1bn dropping more than €3.5bn right there! Worse, after a few years the purchasers re-sold the concession for £3.6bn another £1.5bn money-for-nothing at taxpayers (and users) expense. As has been pointed out the user charges for HS1 are 5 times that of Calais-Paris.
Ugh, I only saw this now. (I added in the links from the comment this is more or less a duplicate of to consolidate the two comments into one.) Sorry.
Swiss Rail benefits from well respected operators and competent builders. This ensures strong public support as shown in the referendum. This is a virtuous circle. CAHSR and HS2 can only look with envy. As for Eurotunnel, I wonder if the financing you suggest would have survived P&O and Stena Lines legal challenge. Instead, they got some financing “a fonds perdus” from retail investors who bought into the IPO but it wasn’t enough.
@Michael
I suspect it probably does include the rescue costs, that is how we roll.
The Jubilee line extension costs include related bus stations etc.
Awful handling.
I really don’t feel the customs clearance is a major problem. Also immigration is definitely a top issue in lots of Western countries and in rich Asian ones too.
Going to Europe I would allow 60 minutes in London St Pancras and 45 minutes in Paris. It’s not a massive difference.
So all the reports about the trains carrying only 2/3rds of capacity purely for the reason that that is the only way they can stick to schedule due to the long complicated immigration clearance? I would have expected the new biometric system to have speeded it up but apparently the prediction is that it would slow it down further, for reasons only the Brits might understand. Numbers are still lower than pre-Covid, never mind reaching the target of 30m by 2030/35 whatever. This is what Alon is talking about. How it has failed expectations and obvious potential. But due entirely to the Brits and nothing to do with French elites! It’s depressing to think it may not be remedied until 2086!
Those HS1 charges you mention are the least of it. The Getlink charges, including the €20 per pax, are much bigger. Eurostar barely breaks even and no matter how many additional pax it carries it mostly is just a funnel of cash towards Getlink especially with that extraordinary per pax charge. The Brits (and Alon) can complain that Eurostar doesn’t have any care about British national interest but then 1. why did they sell out their own stake and 2. it should have been a state-owned bit of infrastructure because of that ‘national interest’. Economists claim it has brought untold billions of benefit to the British economy from both freight, pax and LeShuttle and that is exactly what national ownership of important infrastructure is all about. Thatcherism couldn’t see past its own nose.
Incidentally when Groupe Eurotunnel had its first IPO (I think, unless it was Eurostar itself?) I considered subscribing (it was widely promoted by governments) because at a certain investment it conferred lifetime free travel on Eurostar! However that perk was jettisoned at its first bankruptcy.
I do hope they develop the Ouigo service. It’s just a 2h journey so who cares about so-called luxury etc. and, unlike most LCCs, Strathfield and CDG have excellent connections to their respective cities. They could even capture some international travellers because they could avoid Heathrow … and the cost of a Ouigo ticket would probably be about the same as the Heathrow Express (and a fraction of a London taxi) …
and
OK this I agree is bad.
My objection is people pretending that turning up an hour before in London is essential but that you can turn up 10 minutes before in Paris.
That said I think you could easily get 1/3rd of passengers to board at Stratford International which would at least partially address these issues.
If we are talking advance fares the Heathrow Express is £10. That’s certainly what I paid last time I used it and is very similar to the Elizabeth line fare from Heathrow.
You’re a Brit so of course you accept that there should be a dozen different fares depending on when you purchase and the colour of your disability sticker. Drives me crazy. I want to turn up and go.
Heathrow Express has 12 different fares, including a business class. The standard Express Saver (bought at the station on the day): US$33.20 (rtn $54.00). An online-only 90d-advance-purchase: US$10.80 (rtn $18.40). [this is from my old files, terribly out of date]
It’s a wonder they don’t introduce dynamic pricing then it would be an infinite number of different fares combined with total opacity.
Incidentally the concept of running Ouigo trains from Stratford to CDG also avoids what you seem to be suggesting: having St Pancras trains stop at Stratford. DrAlain didn’t answer whether that concept is advancing.
Michael, I like the idea but this is the first time I hear about a Stratford-CDG Ouigo. CDG2 train station is mutating but the changes I read about never mentioned the creation of passport checking facilities. I believe an existing platform will be used by the CDG Express which should start running southwards towards Gare de l’Est in 2027. Northwards, the ligne nouvelle Roissy-Picardie is a less controversial 6.5 km project. When completed, TGVs starting from Amiens, a mid-size Prefecture bypassed by the LGV Nord, will join the Ile-de-France bypass north of CDG and reach Lyon and Strasbourg in 3h10, saving about 40 minutes over current options. Every 30′ the connection will also be used by a Amiens-Creil-CDG2 TER shuttle which should use a 5th platform created in place of a running track and is expected to be popular with the airport workforce.
Must have been someone’s fantasy.
OTOH since the bottleneck is British the Stratford service could still use Gare du Nord? I suppose that doesn’t fit the concept of Ouigo …
When Eurostar went to Disneyland was there passport control there or did they force everyone off to do it at Lille?
With all these problems I suppose the plans for La Defense as a Eurostar terminal are in deep freeze.
Michael, Ouigo’s initial concept copied Ryanair’s use of cheaper peripheral airports. Ouigo, and to a certain extent Ryanair, have expanded to more central termini. Ouigo uses Gare du Nord for its classic service to Bruxelles-Midi so Stratford is a possibility. Peripheral stations like Stratford are often badmouthed by rail fans but a large fraction of the population find them convenient.
Eurostar had seasonal services to various stations in the Alps and Provence. On the way back to London, passport control required stepping out of the train in Lille.
There are no plans for Eurostar terminal in La Defense. The ground is getting crowded. Metro M1 platforms are possibly dangerously undersized, and finding space for Grand Paris Line 15 was difficult. This is not the place for new 400 meter long Eurostar platforms.
I believe however that the future CDG Express should have gone to Magenta, Haussmann and La Defense over RER E tracks. There is space for turnaround at the next stop La Folie, a former freight yard. The CDG Express tracks are directed towards Gare du Nord. Because the station is already crowded and connected to CDG by the RER B, CDG Express created a connection to nearby Gare de l’Est who lost half of its ridership when the RER E started running. There will be space for a nice terminus but I don’t think many travelers want to go to this neighborhood. Opening is expected in 2027. In a few years we should know whether the ridership target is met.
Many LCCs do use small regional airports that then require a bus or worse (taxi, Uber) to the nominal city destination, sometimes >50km away. The thing about Stratford (and CDG) is that it has excellent transit connections to the city, especially now that the Elizabeth Line is running. The cost and politics of running immigration control at Stratford will be the barrier. Maybe it all awaits Brejoin along with Schengen. Yeah.
For La Defense, it was surely in the context of it being one of the 12 potential suburban TGV stations, presumably at La Folie. It then makes sense for Eurostar trains … except of course for the immigration b.s.
CDG Express is entirely weird. I can’t see who would use it over the RER-B or eventual M17. Also I can’t see its logic since it doesn’t relieve the RER-B3 straining the B corridor, since it shares 24km of track with RER-B thus in some ways making it worse? It’s a pity it wasn’t privately financed as originally intended, so that it could be taken over when it went bankrupt. As it is, it is simply using up unnecessary public funds that would be better spent elsewhere. Was it a Keolis fantasy provoked by Heathrow Express?
CDG Express is indeed the weirdest thing. I would not blame Keolis. They were just part of the winning bid for its exploitation. A fantasy inspired to politicians by the Heathrow, Arlanda or Hong-Kong Airport Expresses looks like a better explanation. It seems they thought that international visitors having to travel on the RER B among immigrants from Seine St Denis was unbecoming.
Tickets will be expensive but it will also be financed by a tax on air trips starting or ending in CDG. The EU said that the tax was legally acceptable. The receipts should pay back the principal and interest on the loans financing the construction. The fact that private banks were not interested and the State had to step in as the loaner is not a good sign. Another weird thing is that CDG Express website emphasizes how easily it will be possible to reach the Magenta-Gare du Nord complex through new underground passageways. That is good news because it takes currently about 500 m though public streets and staircase but begs the question: Why didn’t you send the trains to Magenta in the first place ?
As for Eurostar towards La Defense, La Folie or Nanterre, it always was a fantasy. This is the short platform corner of France. During episodes of RER A interruption, Gare St Lazare could receive a 214 m long pair of MI84 but cannot take in the 224 m long pair of MI2N. The Ligne Nouvelle Paris-Normandie would pass through Nanterre but platforms won’t be lengthened.
The thing about Stratford is that it is actually quicker to get to for passengers coming into Paddington, London Bridge, Waterloo, Fenchurch Street and Liverpool Street which is a decent chunk of the London terminals.
CDG is more like Ebbsfleet – frankly a cheaper, less frequent Ebbsfleet-CDG service would be very interesting. That would allow some good price differentiation.
Curious regarding the CDG Express project and necessity. Now I know airport express projects aren’t popular among the cognescenti on this blog comment site, but my experience which is admittedly only restricted to Japan and namely Tokyo’s two international airports, is that express type services are quite useful to travellers, especially international travellers (both outbound Japanese and inbound foreign visitors), as their luggage tends to get in the way on crowded commuter trains, which not only inconveniences them and others but also may affect timekeeping on impacted railway lines. Rail mode share at Haneda AP is around 50%, and Narita is 56%. You also have the excellent airport access bus network (Limousine Bus), which not only serves the major hotels in the city center for the inbound tourists, but also the major railway terminal stations and some major inner suburban railway stations. Unfortunately the Le Bus Direct services were ended to CDG. What do the French normally use to get to CDG? The RER, private automobiles, taxis, uber?? Do they travel light?
In a 2011 study, 43% of travelers reached CDG by public transport, a number in the mid range among the 14 European airports under study : 32% used rail, 11% buses, 27% taxis and 30% private cars. CDG Express say that they expect the automobile share to decrease by 16% once they are running. Their rolling stock will be well suited to people traveling with luggage. Personally, I like lines serving the airports and I can afford the fare but am skeptical they can achieve such a shift with Gare de l’Est as a terminus.
Alon – I heard you on Volts where you mentioned the Caltrain electrification, good to see this post today! I use Caltrain a couple times/week. As you mention, it’s much faster and quieter than the diesel trains were before. That + more trains/hour has led to increasing ridership (though still below pre-pandemic levels).
I’m using your numbers to figure out how much we overpaid:
We spent $2.4B to electrify 52mi = $46m/mi = $28m/km
“Netting out the elements that are not direct electrification infrastructure, this is till well into the teens of millions per kilometer. ” –> Lets say $15m/km ( * 1.6 km/mi * 52 mi = total of $1.2B of the project for the electrification part)
“the RIA Electrification Challenge overall says that the cost on double track is in the $3.8-5.7 million/km range in today’s prices, and typical Continental European costs are somewhat lower.” –> Could have cost $315m to 475m
So depending on which comparison you use, we spent 3-5x as much as should have been, overpaying by $725-800m.
Caltrain reported a 76% ridership gain for June ’25 over June ’24.
Yes, it was ‘over budget’. Sadly this is how US patronage distribution works. projects get built only if enough bakshish is spread around. The CTA Red Line Extension is a prime example. A new piece of the ‘elevated’ is being built at $5.7billion (and likely more) to serve a minority housing project rather than restoring stations on two ‘higher class’ commuter lines which used to serve the same area. Separate but not even equal.
I have to ask, is Caltrain electrification actually poisoning the idea of electrification elsewhere in the US? All the coverage I have seen has been very positive, both as to speed and ride quality as well as the huge jump in ridership as you noted. Caltrain is now at about 60% of pre-pandemic ridership, by comparison Metro North and LIRR in New York are around 75%. Second Avenue Subway and East Side Access were overpriced, but NY moving forward with SAS Phase 2 and Gateway. Central Subway in SF was overpriced and BART to Downtown San Jose will be, but the Downtown Extension (DTX) for Caltrain is plodding along in planning and there is talk of a regional transit measure to include a new Bay crossing for BART.
Does too-expensive projects affect the will/desire to built things, or does it just limit how much will get built?
Alon’s point was that due to the incuriosity and exceptionalism of American policymakers and passenger railway managers, be it Amtrak or joint powers authorities or public-benefit corporations like the MTA, or state agencies like CTDOT, they will not consider electrification projects outside the United States as a basis of cost estimates; and of recent electrification projects, Caltrain is the most extant. When Brightline West is completed, that’ll be another data point (but again, a bad comparison for regional or commuter rail – it’s a greenfield private enterprise development for a high-speed intercity service with federal grants but no federal involvement in engineering other than permitting). What won’t be considered is French, Spanish, Italian, British, German, Israeli, or Japanese electrification projects.
Too-expensive projects absolutely affect the will/desire to build things. The treasury is not infinite and an exorbitant cost can and will dissuade policymakers from pursuing certain transport modes. I was told by State Sen. Rick Lopes (D-CT), vice chair of the CT Legislature’s Transportation Committee and committed rail enthusiast, that an exorbitant cost estimate for a New Britain-Hartford-Bradley Airport light rail line ($1B), spun out of whole cloth by a DOT commissioner opposed to rail, poisoned the project entirely and knocked it back to a poorly patronized (but somehow eligible for FTA cost share!) BRT system.
Caltrain carries 33k passengers per weekday and less on weekends. These are, unfortunately, atrocious numbers for a commuter line running through a heavily urbanized area. If I were a policymaker I would not brag about spending $2.4B and not achieving a return to pre-pandemic ridership. “I spent $2.4B on trains!” “Yeah? How many more people are riding it?” “Well…less than before. But that’s because of those remote tech workers!”
But they do brag. The metric is public funds spent, not constituents benefited, not objectives achieved. That’s why Gov. Ned Lamont (D-CT) put out a press release touting the purchase of 60 unnecessarily heavy unpowered passenger coaches for $5.25M apiece, which is locomotive money in normal countries. That’s why Amtrak brags about a billion-dollar tunnel in Baltimore that provides less benefit than its original design which also cost a billion dollars. That’s why the MTA brags about $25M per battery-diesel Siemens Charger, per-unit the most expensive road power in the world, instead of electrifying lines.
What happens when IIJA funding is exhausted? How will these service providers keep equipment running and infrastructure in a state of good repair?
I suppose they could offer bus service for the decade or two or three it takes to fund the study to study whether or not a study is warranted all the way through the final environmental impact statement of 2066.
I’m going to make this clear to anyone reading your comment (not you) who is trying to figure out what point you’re trying to make (there is none):
The antidote to inaction, poor cost estimating, poor management practices, and any other obstacles to successful project execution which has clear benefits to constituents and achieves the objective of fast, frequent, highly patronized passenger rail service, is sustained engagement with legislators and executive branch leaders (e.g. authority executives or agency directors) , or leaders at public-benefit corporations (like the MTA). In other words, lobbying. That’s what I’m doing, within the admittedly constrained limits of my knowledge, ability, and connections (almost none). I’m an amateur at this, but at least I’m showing up. I have the data, courtesy of Alon and their collaborators at Marron, and ETANY, and many other organizations who are better at this and have been fighting for longer. It’s up to me – us, really – to create the constituency, and to persuade.
I don’t expect to see overhead catenary from Springfield to New Haven in my lifetime. That is my starting expectation. Any positive movement from that zero position is a victory.
Actual trains carrying actual passengers even though they don’t satisfy your epicurean railfannery, does that.
Monorails would be a great example of this. Its a niche technology that because it was initially treated oversold as a magic technology that could escape traditional rail limitations, is treated as never working. Also like Nuclear power suffered from very successful Simpsons delivered propaganda.
The best example of this is actually the LRT craze in the US which is dying down city by city due to horrible alignments. Portland and Denver once praised (never understood that) have stopped building all together since freeway and freight ROW produce so little ridership they barely transport 1000 pax per mile. Even cities that build lines that more closely resemble french tramways have stopped building or definitvly going to stop after choosing turds like the green line extension (MSP) and purple line (Houston). Phoenix out of all places is continueing to build since it more or less has always picked sensible alignments and thus actually gets a boost in ridership (relative to what place Phoenix is). Denver could have easily made the ridership of its current quagmire with a single line along Colfax.
The EU is not a United States of Europe. There’s simply no political appetite from member states. The unfortunate reality is that member states remain by varying degrees inward looking, protectionist and conservative. Legislation on technical common standards and non-discriminatory access is as far as the European Commission realistically can go. The European Commission is a civil service – it should be on tap not on top. There’s no point in it pushing for anything the Council of Ministers have no interest in pursuing. This is far too big and intricate area for ‘creeping federalisation’ – it’s not going to happen.
https://www.economist.com/europe/2025/07/17/despite-enormous-challenges-the-eu-sticks-with-its-puny-budget
This is the reality of EU budget setting. Member States are not voting for a bigger EU, and EU spend on cross-border transport will remain limited to the CEF pot. It’s no surprise CEF money is mostly going to Eastern European countries where there’s often not a functional modern railway (i.e. existing lines are in such bad state of repair they have to operate at 30-60km/h), and every Euro goes much further than in western Europe. Building a modern 160km/h railway between Zagreb and and Bucharest is far more impactful than the same money spent in western Europe.
One of the biggest challenges to running cross-border services is actually getting enough staff competent in all languages to operate the services. This is what’s plaguing the Liege – Maastricht – Aachen service and it’s nothing to do with infrastructure.
EU passports work in the UK e-gates. Even so, there is a EU (really, French) requirement for a stamp on a piece of paper at the moment; you pass through the automated UK control in the GdN and a lone French immigration officer perched on a stool desperately tries to keep up with stamping everyone. A few weeks ago, I went through and my UK passport worked in both sets of the juxtaposed e-gates but, you guessed it, still stamping.
I am not sure whose “Brexit mindset” is at work here.