Maglev, still not taking off after 10 years

I am often asked about Maglev and whether it is the technology of the future. So it was fascinating to travel on the maglev between Shanghai and Pudong Airport, to find out more about it.

The line which yesterday celebrated its 10th anniversary, is not a great success. The 19 mile line cost $1.3bn to build and has never recouped any of its capital costs, just about covering its operational expenditure in a good year  Partly this is because it goes to the wrong place in Shanghai, the Longyang Road metro station which is nine miles from the town centre. Since the metro actually then continues out to the airport for just 10 yuan (about £1) many people use that rather than hopping out and taking the Maglev which costs 50 yuan (40, oddly, if you have  plane ticket). Although the metro takes 30 minutes compared with the Maglev’s eight, since the latter only runs every 15 or 20 minutes, the time saving is marginal. As a result loadings are barely 20 per cent and on the morning Maglev I took, it was more like 10 per cent.

.  The ride was surprisingly bumpy and actually the maximum speed was only 300 kph, the same speed as most high speed lines, rather than the advertised peak of 430 kph because running it faster uses more energy which already accounts for two thirds of the operating costs. The whole operation of the line indeed seems a bit clunky and a weird mix of low and high tech. Before you can board the train, the various uniformed platform staff, of whom there are several, have to open a gate manually and take down a red rope strung across the gap.

While it is obviously exciting to travel at that speed, the infrastructure is immensely complex and expensive. Piling had to be done to very small tolerances in order to ensure stability and it is probably necessary to keep the track in the air on concrete tracks, which consequently is extremely expensive.

The whole scheme, therefore, does not make sense and would never have been built in that way in a country more attuned to the market.  In fact, it was really a testing ground for the technology and it has consequently failed. Initial plans to build a line between Beijing and Shanghai, which were mooted when the scheme was first given the go ahead, have been scrapped and a conventional high speed line constructed instead. While a few schemes around the world have being mooted, most have not passed a financial viability test and the incompatibility with rail and the cost of construction, combined with the difficulties of creating junctions and turn-offs, means maglev is unlikely to take off (except for the 10-15mm it is levitated above the tracks).


  • John Kirk

    I use it and the metro link from Pudong when I visit Shanghai , but it would be better if the link to the metro was a bit more obvious. Poor connectivity.

  • Speaking as a maglev technology consultant, I am also asked whether
    maglev is the technology of the future. My typical positive answer bears almost
    no resemblance to yours, so it was disappointing to read your article.

    Maybe there was a desire to confirm your own biases [“Should
    UK trains look to magnets?” from September 2006, at, but your comments after taking a trip on the Shanghai maglev could have been better-balanced.

    For example:

    · The system was opened to the public in January,2004 and was officially turned over to the Chinese in April of that year, so the 10th anniversary wasn’t yesterday. It was at least eight months ago.

    · When the system was laid out in 2001, the Longyang Road metro station was expected to become the edge of the business district; things just haven’t worked out that way.

    · Is it so odd that airplane passengers should get a discounted ticket on their trips to the airport?

    · The maximum speed was 300 kph rather than 430 kph because of the time of day of your morning trip – before 9:00am or after 10:45am, it indeed runs at 430 kph. That should not have been a surprise.

    · Running the vehicle faster does use more energy, which is why the daily schedule was adjusted years ago to cut back on high-speed running times in the morning and evening. Again, no surprise.

    · So before you can board the train, the platform staff have to open a gate manually and take down a red rope strung across the [platfom] gap. Ever go through security at an airport?

    · The infrastructure is seemingly complex because of the well-known alluvial soil conditions in Shanghai, which forced guideway beam spans to be halved from planned 50-meter spacings and the number of support piers to be doubled.

    · In the meantime, the construction technology has gone through significant cost-reduction efforts, so the system is not actually much more expensive than conventional steel-wheel trains when compared like-for-like.

    In your telling, it’s no wonder the whole scheme fails to make sense. Too bad some useful facts remain hidden from view.

  • Phil

    The ‘useful facts’ you cite don’t do much to blunt Christian’s criticism. The line has very low passenger load because it’s much more expensive than a competing option and has little demand on its route. It hasn’t recouped any of its capital costs. That’s the stuff of nightmares for Department for Transport officials and whichever politicians would sign off on the investment, and it’s not the basis of a strong case even for those of us more open to risk-taking in the name of progress and inspiring some national pride.

    I for one would welcome your insights into what can be learned from the Chinese experience and Japan’s work on the Chūō Shinkansen, which may well provide a much better example for the world.

  • John Harding

    I think it is time to concede that the Shanghai maglev has not done much to promote the technology. That should have been evident at the outset because it does not save much time over alternative methods of getting to the airport from downtown, and is certainly quite expensive. Yet it barely covers the cost of electricity. For such a short route it makes no sense to travel at 430 km/h. It was basically a demonstration of technology in a public arena. Without planned extensions to more distant locations it is an orphan.

    There were certainly more ideal projects for Transrapid like Berlin-Hamburg or even Orange County to Las Vegas, but insufficient political support. Fortunately for maglev the Chuo Tokaido bypass superconducting maglev line is a go, with over 42 km of the most difficult construction already completed. And there is every evidence that it will be a commercial success with amazing trip times and affordable fares that puts highway and air travel at a disadvantage. And although energy consumption per seat-km is three times that of HSR, it is still well below that of jets flying the same route.

  • My citation of ‘useful facts’ wasn’t meant to blunt
    Christian’s criticism, but to question the accuracy of his statements – statements that could’ve been checked on Wikipedia prior to departing for Shanghai – that set the negative tone at the outset.

    Yes, the line has very low passenger load because it wasn’t built to satisfy a market demand. It was mandated by a previous leader who had a politically driven timeline for a technology demonstration in mind. And no, it hasn’t recouped any of its capital costs, just like every high-speed rail line ever known. So what? I thought these problems were supposed to be maglev-specific.

    Let me think about insights that can be taken from the
    Chinese experience and Japan’s work on the Chūō Shinkansen. Thanks for asking.

  • Kevin Coates

    Like so many opinion articles that have come before on this subject, the
    author provides a few factual snippets to lend credibility to his
    incorrect assumptions and conclusions.

    To me, as someone closely involved with this technology and uniquely well informed by the Chinese maglev operators, such articles reveal an obvious
    steel-wheel-on-steel-rail bias while having no real clue as to what operational advantages are offered by maglev technology.

    Mr. Wolmar mentions the delay between maglevs to get to the airport and that
    it doesn’t save much time from taking the subway (you have to wait for Shanghai subways, too). The point is that Longyang Road station was never intended to be a stand alone station. It was simply to be the terminus of the initial operating segment from Pudong Airport, or IOS. The line has yet to be extended because a technology transfer agreement was never reached to the mutual satisfaction of the Chinese buyers and the German technology providers.

    If the maglev is eventually connected to Shanghai’s other airport to the west, Hongqiao Airport, the Number 2 subway trip from airport to airport would take about two full hours and it is usually extremely crowded. The maglev trip would take less than 20 minutes and would better accommodate travelers with luggage.

    Last October was my fourth lengthy trip to China. Starting with a ride on the maglev in October of 2003 as part of an official U.S. maglev delegation, I have ridden on the Shanghai maglev many times, at speeds of 431 km/h and at 300 km/h. (Please see the official SMTDC schedule here: Before that, I also organized trips for journalists to ride the Transrapid maglev at the German test facility in Lathen.

    As I pointed out in a recent James Fallows blog (Atlantic Monthly, October 18, 2014:
    ), the maglev line in Shanghai was an absolute engineering and
    technical success, but is not a commercial success because it was never
    designed to be one.

    The decision to place the line where it is was a political decision, which is not an indictment of maglev technology. If HSRwere deployed here instead, its energy consumption would be greater, the trip times longer, and the maintenance bill far higher. What’s more,ridership would be lower due to the slower trip times. And, a HSR line would still require an elevated viaduct due to the high speeds involved and, as a result, require extensive pile driving for support, due to the extremely unstable alluvial soil along the route – just like all the other elevated HSR lines going in and out of Shanghai – and, there are a
    bunch of elevated HSR lines in and around Shanghai.

    Having ridden the maglev last October 27th (at 300 km/h), I can tell you the ride WAS NOT “BUMPY.” “Bumpy,” by the way, is not a technical term. There is some minor sway, or oscillation, at higher speeds, but no vertical oscillations, or bumpiness, if you will. The ride was actually smoother than when I was in Shanghai in October of 2012. Since then, worn out rubber bushings that connect the vehicle to the magnet chassis have been replaced, which reduced lateral oscillation in the curves that I noticed to be unusually noticeable at the time. The point here is that the ride is smooth and one can get up and walk around at either 300 km/h or 431 km/h without holding onto anything. I know, because I do it every time without any problem.

    New developments in maglev manufacturing and construction since the Shanghai maglev was built now make the Transrapid maglev less expensive to deploy than traditional HSR. The maintenance on the Shanghai maglev guideway, by the way, has required only two weeks of labor in over 12 years. What rail system in the world can claim that? I have this information directly from both the SMTDC’s former CFO and present chief maglev engineer.

    The big cost issue facing HSR is life cycle cost, but no rail advocates want to talk about this dirty little secret. Maglev technology is the answer for escaping that problem while also providing financially sustainable and reliable higher speed service. This is why the CJR is presently building its new Chuo Shinkansen using its superconductor maglev technology – and this information came directly to my ears from the mouth of Yoshiyuki Kasai, Chairman and CEO of the CJR when we last spoke at a conference in Washington, DC.

    I suggest Mr. Wolmar contact people who are experts with this technology
    before he goes shooting his mouth off about something for which he knows
    nothing, or at least very little. Just because a person rides in a car
    doesn’t mean they understand its engineering.

    By the way, the Japanese Shinkansen HSR system operates on standard gauge track while the rest of the Japanese rail system uses narrow gauge. So what that if they are incompatible. Has no one ever heard of a cross platform transfer?

    There is, however, one thing Mr. Wolmar observed about which we both agree. Usability of the maglev station for travelers is not very convenient, especially if one is hauling luggage. But, then again, this has nothing to do with the transport technology, but the decisions made by the people running the stations.

    Fair to say, if this maglev line was built from airport to airport in Shanghai as a privately operated for profit maglev system, reviews of the system would be overwhelmingly positive because the vehicles have demonstrated for the last ten years that they are on time,to the second, 99.97% of the time. What train system (or airline) can claim that?

    For more information, I suggest Mr. Wolmar and others start with this updated version of my November 2004 cover story for Civil
    Engineering about how the Shanghai maglev was built:

  • marty

    “The whole scheme, therefore, does not make sense and would never have
    been built in that way in a country more attuned to the market. In
    fact, it was really a testing ground for the technology and it has
    consequently failed…”.

    Contradiction surely, if it was a testing ground for the technology then it was surely successful in that it proved the technology was not viable. “Markets”, whatever they may be, don’t enter the equation here.

  • John Harding

    “And no, it hasn’t recouped any of its capital costs, just like every high-speed rail line ever known”
    This is a common misconception, however the first truly high speed rail system, the 515 km Tokaido shinkansen has already repaid the government several times the 400 billion yen it cost to construct..
    The current operator, Central;Japan Railway, is still paying off the sum demanded to acquire the line, which is one reason for the delay in financing the maglev replacement of the line.

  • John Harding

    Mr. Wolmar hits the nail on the head with his comment, “…the maximum speed was only 300 kph, the same speed as most high speed lines, rather than the advertised peak of 430 kph because running it
    faster uses more energy which already accounts for two thirds of the operating costs.”

    This is indeed the elephant in the room. Compared to energy, all other operating costs, like maintenance and salaries are trivial. A typical passenger is paying about $5 just for energy or 26 cents per passenger mile. By comparison the Japanese N700 trainset with top speed of 270 km/h costs 0.3 cents per seat mile for energy, while the 500 km/h superconducting maglev uses less than 1 cent per seat mile for electricity.

  • RapidAssistant

    If history is any teacher, then they should have studied the fate of the Bennie Railplane back in the 1930s….a classic example of trying to reinvent the wheel with a clever, but fatally expensive to implement idea.

    Maglev is an expensive and overly complex attempt to solve a problem that doesn’t really exist, a bit like automatic handbrakes in cars…..

  • Dan

    Wasn’t the first Maglev system at Birmingham airport, and designed by BR Research, back in the early 80s?

  • David Faircloth

    Not quite.

    The Birmingham Airport system was the first COMMERCIAL maglev, and one of the cars is in the NRM.

    I believe the first actual maglev ran up and down the yard of the Railway Technical Centre; it looked a bit like an orange painted garden shed with giant BR double-arrows embazened on it! From riding on it (at low speed), I remember it as being “soft”, and I would describe the Birmingham Airport one as similar from the few times I rode it.

  • nozomi07

    Interesting article, confirming our experience in Germany: Many maglev lines had been projected here – both intercity and airport connectors. All of them were finally abandonned: Too expensive (twice the cost of classic rail), no access to city centers (except via tunnel, for even higher costs). And the advantages? Slightly faster, less noisy, thats all.

    As early as 1981, when the French TGV proved that high speed rail is very fast and economically feasable, we shoud have learned that maglev was just a nice dream of the past, same as the supersonic jet Concorde. But our government was pigheaded in not accepting this fact. The idea of levitating a lame duck has cost us a lot of taxpayers money.

  • Dan

    I saw a copy of the December 1932 Meccano Magazine at the weekend – great image of a ‘Futuristic maglev Train’ which is well worth looking at – I found a scan here – supported by a detailed artcile in the magazine.

    Looks like the artcile can be read here – aims for 600mph!

    A great blast from the past!

  • carlake

    The Transrapid is a very good solution. China and Korea both are developing Maglev equivalent to Transrapid but for 600 km/h Chinese CM1 and for 550 km/h Korean Suma.

    The Múnchen line was calculated till 1.85 Bln Euro and the DB recalulated that and come to 3.5 Bln Euro. What does that depend on was the question. Oh that is confidential. The IMB-forum Internstional Maglev Board forum found the calculation. 900 million Euro was added for 50 years operation garanti and further 500 million Euro for overtaking the operation not additional sums at all so the origin calculation was still valid. But at that opporunity the money was already redistined. DB was just sheeting and ought to be caught in yale.

    Similar was even the HSR-lobby in China were the responsible have been for corrupcy.

    Therefore China now built their own Transrapid. 2021 this new train will be built and will take passanger from Shanghai to Beijing i two hours. Another track will be between Jinan and Qingdao.

    Transrapid ought to get that respect it earns in Germany.

  • Ken Johnson