Realism about high-speed rail

Super-high-speed version: Australia has better things to do with $100 billion than building a high-speed rail line. It’s all summed up in this exchange from the ABC TV series Utopia:

The fundamental point made in this clip is exactly right: the engineers, economists, and transport and logistics experts who study this issue – in Rhonda’s words, the lunatic fringe – all tend to wind up finding that Australian HSR doesn’t make sense.

Now here’s the non-express version …

In looking at high-speed rail (HSR) at various times since the Hawke years, you can make a few fairly defensible conclusions:

  • High-speed rail is economical for certain routes, but right now Australia is probably not one of them. The “sweet spot” for HSR trips is between two and three hours. (The fastest Tokyo-Osaka bullet train takes just over two-and-a-half hours; a Paris to Lyon HSR takes just under two. These are the ideal HSR routes.)
  • Benefit-cost calculations for Australian high-speed rail almost always show a Sydney-Melbourne or Brisbane-Sydney-Melbourne high-speed line would struggle to deliver net economic benefits.
  • That may change in the future, if you can get Sydney-Melbourne travel times below three hours.
  • You should assume these routes would actually cost more than currently projected; research by people like Danish economic geographer Bent Flyjberg shows this is the norm for such projects.
  • The proposed Australian HSR would probably bring about a net increase in emissions. This is counter-intuitive until you consider the knock-on effects. For instance, Sydney Airport’s limitations currently mean there’s pent-up demand for Sydney flights; a new train connection will not mean less flights out of Sydney Airport, but rather will mean different flights. Building the line would also use a lot of emissions-intensive concrete and steel.
  • There is not a huge obvious inherent value in a decentralisation project that effectively makes Albury-Wodonga or Canberra or Shepparton into an outer suburb of Melbourne or Sydney. On the contrary, in many ways, crowded is good and decentralisation is bad.
  • To quote Spanish high-speed rail expert Daniel Albalate, the opportunity costs of building uneconomic high-speed rail can be huge, because the projects cost so much. That’s what happened in Spain,where the high-speed rail network cost an estimated €40 billion. $100 billion on HSR is $100 billion you can’t spend on other things.

If you’re not bored yet, here’s even more analysis:

  • My original 2013 criticisms of high-speed rail, eagerly greeted by Club Troppo commenters with plaudits such as “very unconvincing”, “I reckon you’re on weak ground” and “how about fast inter-urban type trains?”
  • INTHEBLACK magazine’s global survey of high-speed rail, written by the admirable Susan Muldowney and including comments from Daniel Albalate and David Hensher and much other insight besides.
  • INTHEBLACK’s survey of megaproject costs and benefits, also by Susan Muldowney and including Bent Flyjberg’s comments.

The Australian high-speed rail proposal comes up every few years and the reasons why it’s currently a bad idea haven’t changed that much.

But the latest proposal is accompanied by a bunch of talk about a new thing – “value capture”. The Prime Minister is quoted as saying this week that “that’s how railways were financed in the 19th century, actually”.

The PM is right: value capture is a real thing. However, it’s a thing that in the past mostly worked around suburban railway stations at greenfields sites, usually with free or cheap land donated by the government. The line itself is not the source of the value; the stations are. My guess is that a high-speed rail project will get much of its value capture not from greenfields sites or small cities like Shepparton or Bowral or Canberra but from real estate development right next to the big-city stations. (Feel free to make an opposing case in the comments.) But I’m not sure how big this value will be, or how it can be measured.

As it happens, there have been a lot of high-speed rail lines built in the past couple of decades, which means there is now a bit of a literature on the value added to various locations by the arrival of high-speed rail. At first glance, the short version of this seems to be that it’s really hard to measure the value, it doesn’t seem to be huge, and in some cases it may be negative because of the disruption associated with putting a big new station in the middle of a busy city.

One of the best studies to date seems to be the 2012 study by the University of Sydney’s David Hensher and two co-authors. Its findings suggest that “HSR does not always lead to (or is positively correlated with) the growth in property or land values in the short term (one to three years after the HSR project); and longer run evidence is not available.” For instance, as best anyone can can tell – and it’s hard to tell – HSR seems to have pushed property prices up in London but down in Paris, while having no effect in Strasbourg. If the changes in value are uncertain and unpredictable, “capturing value” starts to look like a pretty hopeless task. (Again, feel free to comment.)

As usual, Alan Davies at The Urbanist is worth reading on these issues too. And by “is worth reading” I mean, of course, “comes to the same conclusions as me”.

God help me, I’m getting as cynical as the Utopia crew.

[Side note: I don’t know who Working Dog had advising them on Utopia or its predecessor, The Hollow Men, but those advisers had certainly spent some quality time at that ugly point where well-informed policy collides with the worst type of media spin. This Utopia episode even finds time to make the heretical point that the Snowy Mountains Scheme may not have been a very good idea.]

About David Walker

David Walker runs publishing consultancy Shorewalker DMS (shorewalker.net). David has previously edited the award-winning INTHEBLACK business magazine, been chief operating officer of online publisher WorkDay Media, held policy and communications roles at the Committee for Economic Development of Australia and the Business Council of Australia, and run the website for online finance start-up eChoice. He has written professionally on economics, business and public policy since 1987 and spent three years in the Canberra Press Gallery for News Limited and The Age.
This entry was posted in Uncategorized. Bookmark the permalink.

11 Responses to Realism about high-speed rail

  1. Patrick says:

    I couldn’t agree with you more.

  2. conrad says:

    I agree with you on most of that (although I have a higher belief in value capture — in France it does help small cities outside Paris, although not to the extent that it would capture large amounts of money compared to city rail stations). This changes the sweet spot to “places I can get to easily and back in one day”. This is probably especially true in Aus where land is very expensive.

    Also, lots of journeys are in your two to three hour sweet spot, and it is probably even longer than this (Paris->Marseille used to take 3.5 hours but it was a very busy route).

    You can see why just by looking at times. For example, as somewhat that has flown a one hour trip any number of times between Melbourne and Adelaide, the overall amount of time it takes from city centre to city centre is about three hours (longer if you don’t take a taxi in Adelaide). You also can’t work on the plane, they are often late, and you are doomed if you are late. Trains are generally very reliable, have quiet carriages, and power points for your laptop. They are also cheaper than flying, so there is no reason not to take them if they take a similar amount of time, or even slightly longer.

    This means many of the major routes like Melbourne to Sydney are in the sweet spot (I don’t see why it couldn’t be done in around 3 hours).

    • David Walker says:

      Bear in mind these points:
      * A lot of people don’t just travel to or from the CBD. I live near a Melbourne suburban railway station, and on most trips my door-to-door time would only be about 10 minutes lower if I was travelling on an HSR from Spencer Street Southern Cross Station.
      * Sydney-Melbourne isn’t really in the sweet spot with existing technology. Tokyo-Osaka is 550km and most of the trips take more than three hours. Future rail technology, cheaper tunelling and larger populations at each end could bring that line toward the sweet spot.
      * Melbourne-Adelaide won’t get an HSR. There just aren’t enough people doing the trip to justify the line.

  3. Pappinbarra Fox says:

    What if we put freight into specially designed HSR trains. Does that alter the economic equation? Think of the uneconomic trucks that would be taken off the inter-city routes. I only ever see passenger HSR discussed. What if a HSR freight train left for Melbourne/Sydney?Brisbane every hour? With maybe only 1 or 2 stops along the way at freight transhipping points?

    • David Walker says:

      Short answer: freight is rarely significant in HSR. From KPMG’s Phase 2 report in the federal government’s High Speed Rail Study, looking at a high-speed rail line from Melbourne to Sydney and on to Brisbane:

      International experience demonstrates that the only freight carried on dedicated HSR networks is transported in vehicles similar to high speed passenger rolling stock. ‘Light freight’ trains carrying items such as high value, parcel-type goods may have some potential, although there would be some additional cost involved to cater for these services. Additionally, freight services on the HSR line would have ramifications for speed and also for track wear from heavy haulage. Overall, this opportunity is minor when compared to the HSR passenger services and has not been considered as part of the preferred HSR system.

      The report does note that if you can get passenger services out of the conventional rail system, you’ll free up capacity for conventional rail freight. But bear in mind that rail and road freight services are far from perfect substitutes.

      I suspect we may get radical innovation in road freight before we get high-speed rail freight services.

  4. john Walker says:

    The routes that look reasonably plausible are Sydney and airport (s) to Canberra airport and probably Brisbane to Gold Coast and on to Lismore .

    There is a fair bit of population, and travel, growth happening in both corridors anyway. Engineering etc should not be too hard. And flying around about about 250 ks to 400ks is expensive and time consuming- as it is there are two big buses ,most hours from Canberra to Sydney and about half of the passengers get on or off at Sydney international, its cheaper and not ( all up) that much slower.

    PS Pappinbarra Fox
    HSR trains are by definition pretty light – they need very good power to weight ratios and ( think) that the sort of track that could cope with a very heavy freight train doing 250 kph would be a big ask to build.

  5. derrida derider says:

    The reason a new SYD-MLB or (worse) SYD-BRI line is so frighteningly expensive is basically geology and suburbia – there would be no way to avoid LOTS of tunnelling. But building it across open farmland is surprisingly cheap.

    Even outside dreams of a VFT this is why a new MLB-BRI inland freight line stacks up, while a SYD-MLB fast freight line does not, despite volumes for the latter being potentially much larger than the former.

    So who knows – a MLB-ADL VFT might be more viable than we think. We’d probably waste less money keeping Adelaide alive that way than by trying to build cars or warships there.

    • john Walker says:

      Yes NSW had for a fairly long time two, totally separate train systems . One was centered on Newcastle and the other was centred on Sydney ( NSW actually had fir a while two engines” number one”). Linking Sydney to Newcastle was too expensive.

  6. I am and will always be Not Trampis says:

    ah Utopia a great series!

  7. Pappinbarra Fox says:

    So freight is not viable because of engineering problems. Here is a thought – build a third robust line at the same time as the passenger line – the course would be much straighter than currently and therefore shorter. Since a new line corridor is being built anyway for passengers the additional cost would not be that great and even if the freight was limited to 140km it would still be a lot more efficient that a hundred B Doubles on the road. And it might bring the whole proposal into economic status.

  8. Ken Parish says:

    See my new post on high speed rail (much more positive about it than David).

Leave a Reply

Your email address will not be published. Required fields are marked *

Notify me of followup comments via e-mail. You can also subscribe without commenting.