High-speed rail sells saved minutes. Most passengers would settle for a train that turns up, runs often and meets a second train on arrival.
Apostolos Tzitzikostas and Raffaele Fitto stood up in Brussels in November 2025 to launch the European Commission’s high-speed rail action plan, which proposes some 345 billion euros by 2040, and perhaps 550 billion by 2050, on a network joining the Union’s capitals. The transport commissioner promised Berlin to Copenhagen in four hours rather than seven by 2030, and told the press conference that “high-speed rail is not just about cutting travel times”. Buried in the document is the standard the Commission actually set: an upgraded line must carry trains at 200kph, a new one at 250kph or more.
Oskar Herics and nine colleagues at the European Court of Auditors examined 5,000 kilometres of Europe’s network in 2018, in France, Spain, Italy, Germany, Portugal and Austria, and called what they found an “ineffective patchwork”. Trains on the lines they audited ran at an average of 45 per cent of the speed the track was built for. Only two of the 10 lines managed an average above 200kph. None managed an average above 250kph. The auditors costed a Stuttgart to Munich scheme at 368.69 million euros for each minute of journey time it saved, and concluded that an upgraded conventional line would have reached the speeds actually being run, for very much less money.
Alon Levy, who works on the Transit Costs Project at New York University, told Rail magazine in April that the French built quickly and cheaply because they began by refusing to tunnel, having what Levy called “an almost pathological cultural aversion to them”. Speed is what removes that option. A railway designed for 360kph needs curves of seven kilometres in radius or more, so it cannot be steered around a hill or a town, and goes through instead, in bore or on viaduct. The track sits on concrete slab rather than ballast. Level crossings are banned. Air resistance climbs with the square of velocity, so the electricity bill and the wear on the rail head climb with it. The auditors put the average cost of the lines they examined at nearly 25 million euros per kilometre, excluding the tunnelled sections; France finished the LGV Est at 16.5 million. Eric Goldwyn of the same New York team told a conference of rail executives in Washington that America builds “the most expensive high speed rail projects in the world”. Spain opened the 621 kilometres from Madrid to Barcelona in 2008, at roughly 13.1 million euros a kilometre in today’s money. California budgeted 33 billion US dollars for its first 277 kilometres, which is nearly 120 million a kilometre.
As fast as necessary
Swiss voters approved Bahn 2000 in a referendum in 1987, and the first stage arrived in December 2004: some 130 separate projects for 5.9 billion Swiss francs (6.4 billion euros), the largest of them 45 kilometres of new line between Mattstetten and Rothrist. Zurich to Bern fell to 56 minutes. That number was the point of the exercise, because trains reach Zurich, Bern and Basel just before the hour and the half-hour, pause, and leave just after, so a passenger going from more or less anywhere to more or less anywhere else changes once and waits five minutes. Rolf Bergmaier, a planner who worked on the programme and later redrew Zurich’s regional bus network, has described a method that begins with the timetable and derives the concrete from it. The campaign slogan promised trains that would run not as fast as possible, but as fast as necessary. The tilting stock tops out at 200kph.
Huw Merriman, then Britain’s rail minister, handed the Transpennine Route Upgrade another 3.9 billion UK pounds in December 2023, taking the government’s commitment to 6.9 billion UK pounds. Neil Holm, who runs the scheme, said the money moved two of its largest projects from design into construction. The work covers 122 kilometres between Manchester and York: full electrification, four tracks instead of two from Huddersfield to Ravensthorpe, digital signalling, eight trains an hour, and Holm has said the aim was to avoid digging the railway up again in ten years. The Department for Transport priced the job at 2.9 billion UK pounds in 2018 and now puts it at up to 11.5 billion, finishing some time in the 2030s.
Heidi Alexander told parliament in May 19 that HS2 would cost between 87.7 billion and 102.7 billion UK pounds, against the 15.8 billion to 17.4 billion quoted when the scheme was first priced in 2010. Originally planned as a vast, national high-speed railway connecting London, the Midlands, and the North of England, it is today a hugely scaled-back project that will run between London and Birmingham, cutting journey times by not very much. Sir Stephen Lovegrove, reviewing the vast expense for ministers that month, traced part of it to the decision to design the line for 360kph, as fast as any conventional railway anywhere. Alexander had asked Mark Wild, the chief executive of HS2 Ltd, to model 320kph in March; her department reckoned the effect on journey times would be negligible. Wild, defending the reset, said: “Speed has never been the primary objective.”
And nor should it be. Speed does not always need to be the primary concern. Knocking off a few minutes from travel times is a lesser outcome than providing quick, punctual services with good connections. Herics and his auditors wrote as much in 2018, and the Commission set the number itself last November. An upgraded line, in Tzitzikostas’s own plan, means 200kph.
Photo: Dreamstime.