What does the new HSR feasibility study say?

HSR short-listed station sites, Melbourne (DIT/AECOM)

I’ve had an admittedly rushed look at the Executive Summary of the High Speed Rail Study – Phase One, released today by the Minister for Infrastructure and Transport, Anthony Albanese. I’ll have a closer look at the full report shortly, but for now here are a few initial thoughts.

Today’s report is Phase One. It looks at infrastructure costs and forecast patronage. The really important bit – the analysis of the benefits and costs and how the project might be financed – must wait for completion of Phase Two. So for the moment difficult questions like “is HSR a good idea?” are side-stepped.

This study was done partly because The Greens and Independents required it be done. But given how popular the idea of HSR is, it could be up there with the NBN as one of the Government’s smarter moves politically. One way or another the Government’s going to support it and find a way to make it look plausible. After all, it reeks of ‘vision’ and no significant outlays will probably be required for at least two terms. But the risk is it won’t be examined seriously.

Mr Albanese is certainly talking up HSR. He is quoted as saying high-speed rail would be an “attractive alternative” for many, particularly those fed up with airport scanners introduced after the 9/11 attacks. It’s a pity he didn’t see recent reports of Al Qaeda’s interest in trains or recall the Madrid train bombing.

As is now seemingly obligatory, Mr Albanese also cites the success of the AVE system in Spain in support of HSR. ”In Spain, the line between Madrid and Seville is so popular, it carries more people between those cities than cars and airplanes combined”, he says. I’ve pointed out before that AVE is a questionable analogy, at least for routes like Sydney-Melbourne  and Sydney-Brisbane – Madrid has a population of 6.5 million and is only 391 km from Sevilla.

The report says the estimated cost for the most likely route between Brisbane and Melbourne is a cool $108 billion – and there’s a 10% chance it could be higher (ignore the lower $61 billion figure published in the media as there’s a 90% chance it’s too low). These estimates don’t include planning and procurement costs – so add another 15% – and nor do they include the cost of buying and operating the rolling stock.

The estimated cost for the Sydney-Canberra-Melbourne leg is a whopping $45-$50 billion, depending on whether it goes via Wollongong. And of course, add procurement and operating costs.

The study is upfront in making it clear the capital cost can’t be recovered from revenue. International experience, it says, “suggests it is unrealistic to expect the capital cost of a HSR network to be recovered”. Of course that’s par for the course with public transport, but in this case we already have a competitive airline system transporting the public between Brisbane, Sydney, Canberra and Melbourne. So the cost to the taxpayer of replacing one form of public transport with another is no idle matter.

One reason the capital cost is so high is because the investigators have concluded an HSR network is only sensible if it provides for speeds as fast as 350 km/hr in non-urban areas and 200 km/hr within cities. They have assumed a dedicated two track right of way, with tunnels from the urban periphery to the CBD in Sydney, Melbourne and Brisbane.

Based on these speeds, they estimate the travel time between Sydney and Melbourne CBDs at around three hours, making it competitive with air for city centre workers. That seems ambitious – I’ve noted before the maximum permitted speed of Spain’s new AVE system is 300 km/hr. China’s extensive HSR system is also limited to a maximum speed of 300 km/hr for reasons of safety.

Although other candidates are being considered, the most likely city centre station in Sydney is Central and in Melbourne Southern Cross (the alternative is North Melbourne). Suburban stations are also being examined e.g. Parramatta. Stations deep underground are ruled out, so it could be a challenge to accommodate new works in the CBD.

It will come as a surprise to many that HSR is not capable of serving either Sydney or Melbourne airports due to differing operational requirements. It’s possible however than a Melbourne Airport train and HSR could share the same infrastructure, e.g. own tracks but same tunnel. One of the most interesting aspects of the report is that a second Sydney Airport doesn’t appear to even be mentioned. Regional stations are assumed to be located at approx 70-100 km intervals.

The study assumes inter-city HSR fares would be the same as air fares – it doesn’t calculate a fare that recovers costs, not even a portion of operating costs. On this basis it predicts eight million passengers would use HSR between Sydney and Melbourne in 2036, equivalent to about half the forecast air travel at that time. Half of these would be business travellers.

Demand for travel to regional centres is assumed to make up an extraordinary 50% of all patronage, although 85% of these passengers would be cost-sensitive non-business travellers. The study assumes a high proportion of this demand will either be induced or come from cars. Two tracks could be a significant constraint given there might need to be a complex array of stoppers and expresses to serve this large regional market.

The plausibility of the regional demand assumptions and the value to the community of these regional  trips – given how much they the cost the community – is an area I’d like to look at more closely. I’d also like to know more about the assumptions underlying the estimated number of CBD to CBD business travellers. HSR is a very politically charged issue – I note an independent review commissioned by the California High Speed Rail Authority following criticisms of its patronage estimates has just found the Authority’s forecasts are over-optimistic.

As I said at the outset this is just a quick look. The really interesting stuff will be in the Phase Two report, which will also refine much of what’s in the current report. It’s nevertheless important to be clear about what this proposal means. For example, in the case of the Sydney-Canberra-Melbourne corridor, what’s being proposed is a public spend of $45-$50 billion (plus 15% and plus probably a half to two thirds of operating costs) to provide a new monopoly train operator to compete against the four established airlines, without reducing either fares or  average travel times. Phase Two will want to set out some pretty good reasons why that makes sense. I’ve looked at what some of these benefits might be before and not found anything compelling (also see HSR in Categories list in sidepanel).

I plan to look more closely at this (Phase One) report over the weekend.


BOOK GIVEAWAYfollow this link to be in the running for a copy of Sophie Cunningham’s fabulous book, Melbourne. Entries close Saturday 13 August.

17 Comments on “What does the new HSR feasibility study say?”

  1. My biggest concerns with your arguments regarding the HSR project proposals are:

    Ignoring that peak oil will drastically increase the cost of flying in the next few decades, and that in fact currently the cost of flying is already subsidised
    Belief that ‘competition’ will produce better outcomes for people travelling than a monopoly
    Ignoring that a decent HSL line is a very long term investment. Of course maintenance will be an ongoing issue, but once the money is spent once, upgrading the line to new advancements is far cheaper than laying a new line down in say, 100 years.

    • Alan Davies says:

      “Peak coal” is also driving the cost of electricity on an upward trajectory as lower carbon sources obtain a larger share of generation. Also, aviation expert Ben Sandilands reckons “by 2036….jet fuel is realistically predicted to be at least 50% derived from algal or biological fuel substitutes….”. Higher prices drive innovation and the search for substitutes.

      All forms of transport are subsidised to some degree (in fact you can find a subsidy in almost any economic activity if you look hard enough). You need to elaborate on the extent to which air travel is subsidised on the Brisbane-Sydney-Canberra-Melbourne route (financially, not just in terms of unpaid externalities) compared to what would be involved with HSR.

      Who doesn’t think a competitive market is better than a monopoly? That’s not to say competition is always ‘perfect’ but most times it’s far better than monopoly.

      It doesn’t matter if HSR is a short or long term investment if there’s no sensible social rationale for providing it in the first place. Far better to spend that $50 billion (plus!) required for a Sydney-Canberra-Melbourne HSR route on improving urban public transport within Sydney and Melbourne.

      • by 2036….jet fuel is realistically predicted to be at least 50% derived from algal or biological fuel substitutes….

        I agree, jet fuel will almost certainly be derived from bio-fuel substitutes. Unfortunately, these will never be remotely as cheap as the oil we use today. Just as off-shore drilling will never be as cheap as obtaining oil from the few key mammoth on land reserves that have supplied the bulk of oil for the better part of the last 100 years. Once these massive on land oil reserves are depleted, oil will never be close to as cheap as it currently is ever again.

        Your second point is fair, and I don’t have the data on hand to respond at the moment.

        “Who doesn’t think a competitive market is better than a monopoly? That’s not to say competition is always ‘perfect’ but most times it’s far better than monopoly.”

        I’ve got news for you Alan, a lot of people disagree with Neo-Liberalism.

        If you’re able to point to a single good example of a competitive market producing an optimal urban public transport system then I’ll take my hat off to you (and please don’t mention ‘the Bangkok Model’). The other obvious one is Health Care. Nationalised systems produce better results than “competitive markets” just ask anyone that’s gone broke in the US after needing some surgery. There are many examples of natural monopolies, services that function far better as monopolies for various reasons. These almost always should be state owned, state managed (with some commercial outsourcing) or at the very least heavily regulated, again by the state.

        Far better to spend that $50 billion (plus!) required for a Sydney-Canberra-Melbourne HSR route on improving urban public transport within Sydney and Melbourne.

        I’m sure there are plenty of good ways to spend $50billion, however I think throwing it at public transport in either of those cities with the current management in place would result in gross wastes of money. Put some competent agencies in place and then it might be a different matter.

        Hopefully if $50billion was to be spent on HSR some competent overseas experience would be brought in to help manage the project.

        • Alan Davies says:

          Preferring competition to monopoly does not necessarily make one a ‘neo-liberal‘. There are indeed situations where markets don’t work, but aviation isn’t one of them.

          Your public transport system analogy is flawed. It’s next to impossible to get real competition with systems like urban rail and hence monopoly is the ‘natural’ outcome, whether private or public. However my neighbours and I would be much, much better off if we had four rail operators sharing the local station — but with their own lines to the CBD — vying for our business. Unfortunately that’s highly unlikely with rail, but it’s effectively what we’ve got with aviation. It’s not ‘perfect competition’ but it’s a lot, lot better than the Government-created two airline duopoly of my youth.

          • “However my neighbours and I would be much, much better off if we had four rail operators sharing the local station — but with their own lines to the CBD — vying for our busines”

            Sorry Allen but I’m still disagreeing with you. It may be impossible to get real competition with urban rail, but it’s easy to do with buses and plenty have tried. Look at the results of bus deregulation in England. Lower customer satisfaction, more confusion, higher and incompatible fares between companies, lower total patronage, less integration between different routes and services. A few of the more profitable routes may have gotten better services, but its been entirely at the expense of making the rest of the network less usable.

            Even Thatcher had the foresight not to deregulate the buses in London at the time. I’m guessing someone would have done it since had the results been different.

            You are right though, Aviation isn’t suited to a monopoly, but I never said it was. I said the fact that the aviation industry industry works as a market while rail would be a monopoly does not make aviation better by default. Travelling by Shinkansen from Tokyo to Osaka for example is certainly a superior experience to flying the leg for about the same price.

          • Alan Davies says:

            Julian, you’re stretching the analogies again! Remember we’re talking about the Brisbane-Sydney-Melbourne corridor here – a very busy trunk route. Buses don’t usually operate in competitive markets because they’re typically relegated to marginal routes – you don’t find competition on marginal airline routes either. However on busy trunk routes suited to buses, it would be possible to have several bus operators competing. Have a look at the very vigorous market in inter-city bus services in the US.

      • The production of biofuels on the scale required to feed the aviation market of 2036 is…um…speculative.

        Furthermore, radiative forcings from contrails are the great uncertain global warming bogey for aviation. Once they get better quantified, it’s likely that the effective carbon price faced by aviation will increase significantly.

        Look, I’m a HSR skeptic too, but it would be unwise to underestimate the economic/environmental challenge aviation will face over the next few decades

  2. Aussie83 says:

    Its interesting if comparing with say a project that actually saves people time and I must say $50b on a project that only services 8M people in 2036 seems like an extravagent waste when you think of all the projects for which that money could buy.

    For example eastlink was around the $2b dollar mark, actually saves people time and is used at around 200,000 vpd, which equates to 73M trips per anum.

    Wouldnt we be better off investing $50b in 25 no. x $2b urban rail or freeway type projects in Melbourne and Sydney?

    Why does everything have to be in a tunnel in urban areas?

    I also hate how they add the travel time, taxi time, waiting time to the airport trip when calculating the air travel time. But then dont go and do the same to the train trip with connections etc. (last time I checked not that many people lived or have final destinations in waliking distance to Southern Cross or Central.)

    Hell even spending money on finally duplicating the intercapital road networks which must have low BCRs due to the very low traffic volume would be money better spent wouldnt it? i.e. Hume 6,000 vpd is 2.2M trips a year … You could get a damn lot on rural roads with $50b would improve safety, travel time, amenity etc.etc.

    Anyway rant over heres hopeing this white elephant gets seen for the political stunt it is and common sense prevails.

  3. heritagepoliceman says:

    Tunnells ! Wondered how they would get into the city centres fast. Still, HSR has a lot of supporters I think basically because people like trains ! Emotional attachment or something. I like them ! I would love to see an HSR to Canberra and Sydney and would prefer that to flying; I can get to Sthn Cross much more easily than Tullamarine. But yes it may be a load of investment for not much social return – may be better to spend the $ on higher speed rail country trains for Melb, Syd and Brisb, allowing some decentralisation instead of ever-growing sprawl. In Vic, basically making sure the Reg Rail project actually makes Ballarat, Beidigo, Castlemaine, Kilmore, and perhaps more sensibly Geelong, Warragul, Gisbourne, etc. a workable commuting distance – going for Albury instead does seem a bit of a stretch.

  4. Johnyboy says:

    I think the HSR project is wrong. It will not work. I think that its just not practical. The biggest problem with the whole thing is that we don’t need it. The population of australia is way to small. This has been talked about before and backed by BHP I think in the late 80s. It was not practical then and is not now. If the money is wasted on this , then the opportunity costs need to be considered.

    What we really need is better public transport within the metro areas of each of our capital cities. The congestion we all see with the cars on the road is a result of bad planning. I probably would need to put up graphs and mathematical formulats as proof of this. To keep it simple . Just realise this. There is 20km of road per car in australia. I cant remmeber the exact amount. Thats alot of road not being used. I know its simple but i hope it explains things .

    Most of the debate seems to be ideologically drive between two views of thought. The socialist people who believe in government and that it can solve all our problems. Then there is the other side that believe that government get in the way and that the best way forward is private enterprise. This ideological drivel that the only way forward is the market and competition will give us the best outcomes is a fallacy. You can not just blanketly apply ideology.

    This ideological driven ideals are for the most part driven by people who don’t want to know the facts or have special interests and are usually just selfish people. That goes for both sides of the argument.

    If anything, we need to manage the ability to move within the cities we have. I am not in favor on restrictions on cars or trucks. I am sure that tunnels for cars is wrong and the government and reports that back road tunnels are just junk. They are not worth the paper they are written on. I think the obvious things are never taken into consideration because the government is essentially corrupt and inept and driven by special interests. The media is essentially dead in australia. I will say there are some enlightening articles within the newspaper but for the most part they are taken from people far away and reprinted here. Though I still read the opinion sections and forums I find them moderated to the extent that most of them are just following the politically correct ideals. Anyone who doesnt conform is shown contempt and made to look as foolish as possible my the moderators.

    The congestion on the roads is and has been created by a corrupt , inept government in the pocket of special interests. Thats it. The cynical public get what they deserve. I can type my thoughts here effortlessly but I feel that it is all in vain.

    I know that i am not some super human or genius to see the solutions to the problems. I suspect that if i can easily see them , then others can too.

    Again. NO HSR. fix the cities we are in.

  5. Daniel says:

    We have to consider the purpose of public transport. For urban public transport, the answer is obvious. Consider the 300,000+ people who commute to the city each day in Melbourne by train and then try to imagine what it would cost to do it by single occupant transport. You have the cost of car parking, roads, injuries, pollution, extra cars, car running costs, etc. If you consider that a single parking bay in the CBD can be worth up to $70,000 and do the sums you’ll see that getting people into the city by train/tram/bus is a bargain. On the other hand, public transport also has a social purpose, it lets people who can’t drive get around.

    Now consider HSR. What is its purpose? It is not cheaper than air travel. It is not faster than air travel. Given its cost it is not greener than air travel (using the money to buy carbon offsets would cause a much greater reduction in pollution). It doesn’t increase GDP (by raising taxes over what they would otherwise need to be, it could even subtract from GDP). It is not more socially inclusive than air travel.

    The Greens and other HSR supporters don’t seem to understand opportunity cost or even, ironically, that resources are limited. One hundred billion would be enough to get trains, buses and trams running every five to ten minutes or more in each of Melbourne, Sydney and Brisbane. Who’d argue that that wouldn’t be a better outcome?

    • wizofaus says:

      Good response. In a decade or two HSR may well be cheaper and/or faster than air travel, but there doesn’t seem be a good argument for using precious tax dollars for building it before it’s truly justified.

  6. Peter Hill says:


    Your summary analysis is spot on.

    The same problem of political dream-visioning arose after 1996, when the then Victorian ALP looked to broadening its electoral support via preferences from left-leaning parties by proposing a rail link between Melbourne Airport and the CBD. This policy was put into effect when the newly-elected Bracks ALP government instructed the Department of Infrastructure (DoI) in 2000 to conduct a feasibility study on rail link options. The defects in the study outcomes were over-estimation of modal share of trips by a rail link, and under-estimated construction and running costs.

    I had the brief to do market research for Skybus’s parent company, which showed more realistic patronage forecasts for rail and for express bus. A direct rail option would not have lifted patronage significantly above the Skybus service as it was. This was submitted to the government’s Ministerial Planning Panel investigation of the DoI proposal. Thence, the government’s decision was to choose between massive outlays on an airport rail link for little net incremental benefit, versus similar outlays in the general metropolitan public transport systems for much larger incremental benefits arising from much larger shifts of travel from cars to public transport.

  7. […] and development issues with a particular focus on Melbourne, Australia What does the new HSR feasibility study say? […]

  8. […] predicted to be at least 50% derived from algal or biological fuel substitutes….”. Others like Robert Merkel are not as convinced of the prospects for biofuels on the scale required. There’re plenty of […]

  9. Aitor says:

    When comparing the AVE in Spain with a possible future train in Australia I find that often the wrong figures are given as to population etc. So, here goes.

    Madrid population: 3.3 million (the over 6 million figure is for the region of Madrid)
    Barcelona population: 1.6 million
    Sevilla population: 704.000
    Distance Madrid-Barcelona: 497 km
    Distance Madrid-Sevilla: 390 km
    Also popular is the Barcelona-Sevilla fast train. Distance: 830 km

    Sidney population: 4.34 million
    Melbourne population: 3.8 million
    Distance Sidney to Melbourne: 712 km

    • Alan Davies says:

      That’s not right Aitor. Population of Madrid metro area is 6.3 million and it covers an area of 4,610 sq km. That’s less than the area of the Melbourne metro area (usually taken as the Melbourne Statistical Division) which covers an area of circa 8,000 sq km and has a population of 4 million. The bit of Madrid you’re citing is essentially what we’d call the inner city/inner suburbs in Australia.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s