Airports & aviation

Feb 23, 2015

Is Daniel Andrews right to ignore airport rail?

A train to Melbourne Airport would be more comfortable than SkyBus and one day it'll no doubt be needed, but the Victorian Premier is right; there are much more pressing priorities for the scarce public dollar

Alan Davies — Editor of The Urbanist

Alan Davies

Editor of The Urbanist

The original 2001 route remains the preferred alignment for a rail line from Melbourne Airport to the CBD (PTV Feasibility Study)

Just when I thought the change of government in Victoria had put the notion of an airport rail line to bed, The Age insists the idea is going to stay up and keep banging loudly on the walls.

Late last week the paper reported a warning from the CEO of Melbourne Airport, Chris Woodruff, that the city risks drowning in traffic if it doesn’t build a rail line between the CBD and the airport. According to Mr Woodruff:

In 10 years time Melbourne Airport will be handling 46 million passengers a year from the current 32 million. There is no other airport in the world at that level that doesn’t have a rail link.

The private owner of the airport is hardly a disinterested player and isn’t offering to foot the bill, so it’s important to look at the key issues related to this proposal. I discussed them at length last year, so this time I’ll just summarise the main points:

  • It would cost a lot to build a line, at least $2-3 billion for the route recommended in the former government’s feasibility study. That money would have to come from taxpayers.
  • It would be a risky project. The two existing Australian airport lines – Brisbane and Sydney – were privately financed but both struggled financially in their early years despite only costing $0.4 billion and $1.2 billion respectively (2012 $$); the initial Sydney operator went broke.
  • It would mostly replace existing public transport services. Most passengers would come from buses and taxis, not from private cars. The environmental benefits would consequently be relatively small (or put another way, it would be an enormously expensive way of reducing emissions and pollution).
  • It would replace a service that costs Victorian taxpayers nothing (SkyBus) with one that doesn’t cover any of its capital costs.
  • Tickets wouldn’t be cheaper than SkyBus. It’s virtually certain the government would insist on charging circa $17 one way as happens with the Sydney and Brisbane airport rail services.
  • The key advantage of rail is capacity but that’s not relevant yet; SkyBus has ample scope to scale up capacity in the near to medium term by increasing frequency.
  • It wouldn’t reduce congestion on the Tullamarine freeway any more than building new roads would; the road space liberated by any motorists who switch to rail would quickly be filled by other drivers.
  • It would provide faster trips than SkyBus during Melbourne’s (landside) peak periods, but it’d be slower in the off-peak (because the recommended route for rail isn’t direct). But SkyBus could be made faster in the peak by giving it a dedicated road lane at considerably lower cost than building a new rail line.
  • It won’t reduce the cost of parking at Melbourne Airport or the airport’s revenue from this source; that’s the result of anti-competitive practices, not the absence of alternatives to driving.

Please read my previous article where I discuss these issues in more detail, Is it time Melbourne got a rail line to the airport?

I suspect much of the enthusiasm for airport rail is based on the positive experience Victorians have with Airport rail lines when they’re visiting other cities; and no doubt most visitors to Melbourne would appreciate the greater comfort of rail compared to SkyBus.

But it’s different for the vast majority of local residents who, after all, would be the ones paying the bill; they mostly drive to the airport because they live and work in the suburbs (over 90% of Melburnians live more than 5 km from the CBD) and live in households with a car (90% of Melbourne households have at least one car). (1)

While there’d be social benefits if they did, residents who currently choose to drive or take a taxi to the airport rather than use SkyBus aren’t likely to switch to rail en masse.

They’ll mostly continue to travel to the airport by car or taxi because it’s faster and more convenient than taking public transport to the CBD and then changing to the airport service (whether train or bus). For many air travel is only occasional; for others the cost of a taxi or airport parking is paid by work. (2)

If airline passenger numbers grow at Melbourne Airport in line with the owner’s forecast, there’ll inevitably come a time when buses can’t cope and some form of mass transit will be required; but that’s not likely to be for some decades yet. When it does, the owner should contribute to the cost. (3)

As I noted last time I discussed this issue, there’s no shortage of competing infrastructure needs like upgrading signalling that make a lot more sense at this time than airport rail. Fortunately Victoria’s new Premier, Daniel Andrews, understands this; he says most Melburnians would only use an airport train once or twice a year:

No one for a moment says that an airport rail link is not a worthy project. But be very clear: the services that people use every single day are my priority.

Of course with the government having to find the funds to meet its existing public transport promises like building Melbourne Metro, extending rail to Mernda, and eliminating 50 level crossings, there’s no way it could contemplate spending at least $2-3 billion on a rail line to the airport even if it regarded it as a priority.

I expect The Age will continue running stories framed in a way that amps up airport rail; its demographic loves the idea. However I’d like to see it put greater effort into giving its readers a more balanced bedtime story.


  1. 70% of Melbourne’s jobs are more than 5 km from the CBD; only 15% are in the CBD i.e. within a 1.5 km radius of the Town Hall. At airports, a high proportion of “drivers” don’t park but either ‘kiss and ride’ or are driven by ‘meeters and greeters’.
  2. Consider that my wife and I paid $76 to park in the long-term car park for four days last week. We had to pay for fuel too, but it would’ve cost us $74 by public transport (train from home to SXS then SkyBus, return) but taken much longer than driving (the cost of parking could’ve been lower if we’d had the wit to pay for it in advance).
  3. Most existing airport lines in other cities were built for reasons of prestige or political opportunity, not because there was a practical need at the time.


Leave a comment

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

24 thoughts on “Is Daniel Andrews right to ignore airport rail?

  1. Joe Vernossi

    The corridor through which the airport line passes makes no sense for additional stations.

  2. Alan Davies

    Michael r James #21:

    I think the key point is that SkyBus contributed the other $7 million for the road works.

  3. john doe

    Australia does have XPT trains which are based on the 1982 British HSR. But in operation they are speed restricted to 160km/h because of inferior railway track in Australia compared to the UK in which the trains operate at 200km/h.

  4. michael r james

    A Davies:
    [It would replace a service that costs Victorian taxpayers nothing (SkyBus) with one that doesn’t cover any of its capital costs.]

    [The freeway is used by Skybus Super Shuttle services to Melbourne Airport, and in 2002 the Victorian government contributed $3 million to a $10 million plan to expand and improve these services, after a feasibility study into an airport rail link found the number of passengers using a train would not make the scheme economically viable]

    Further, the bus uses the Tullamarine Freeway which is being upgraded with two additional lanes at a cost of $800m to the tax-payer (though it is apparently at risk with the cancellation of the EWL, tant pis!). And obviously this is yet another short-term fix as widening freeways never, ever, solves the problem of congestion.

    As to Kennett’s nonsense, it is pure distraction. Obviously the privatization of the airports has made sensible policies almost impossible. For the owner they make huge profits from car-parking (and the retailing monopoly, $100m last year) which is very cheap to provide, yet of course relies upon the billions of dollars of publicly-funded roads, not to mention that $800m ugrade.

    [Proposals in January 2013 to improve the bus service to the airport involving turning emergency lanes into bus lanes on the freeway and the Bolte Bridge and putting SkyBus on a myki fare, were challenged by CityLink operator Transurban, because it would limit its toll revenue, and by Melbourne Airport, because it would reduce its car parking profits.]

    However Kennett’s empty suggestion that Melbourne Airport (ie. Australia Pacific Airports Corporation Ltd) fund it does suggest another method: a levy on every airport passenger. A $2.50 or $5.00 levy would generate $80m or $160m today. This is the mechanism that NYC is using to help fund their AirTrain JFK and likewise the creation of an express RER line to Paris-CDG. On top of the train fare (which should not be exorbitant so as to encourage use) from both the airport and local pax, this is a mechanism to fund a government loan without using any tax-payers money. This is a mechanism that is not available to other rail lines–ticket prices cannot be so high as to repay capital costs. It is a reason why it could be given priority while not impinging on any other rail projects.
    AD Note 3.
    [Most existing airport lines in other cities were built for reasons of prestige or political opportunity, not because there was a practical need at the time.]

    Verging on the fantastic!
    No, it’s called vision or planning for the future when it is the easiest and cheapest to do–ie. at the beginning.
    #20 john doe at 11:17 am

    The fact that Australia has no HSR of any type makes it more likely to have MagLev. If we wait long enough it will make sense and its relative cost will be less and less of a difference to the standard option. As I have argued here and elsewhere Australia’s long distances and sparsely-inhabited interstitial bits (between cities) make MagLev sensible.
    Oh, and while we are on the subject, Shanghai’s Maglev cost half of AD’s inflated cost of a stock standard rail line (two thirds uses existing line); that was US$1.2bn for 30km versus 23 km for Tullamarine. And the Chinese and Japanese claim to have made Maglev between 30% to 50% cheaper to build.

  5. john doe

    Robert Merkel #19

    Japan first introduced high-speed railway lines in 1964. It is unlikely Australia will construct a Maglev train anytime soon considering we do not even have HSR train lines capable of moderate speeds.

  6. Robert Merkel

    In 1915, the notion that people on median wages would take holidays in Bali via aircraft would have been thought laughable.

    In 1965, at least in some circles it was thought that tourist trips to the moon would be routine by 2015.

    To give a for-instance of something that might dramatically reduce the demand for air travel, what happens if something like Elon Musk’s Hyperloop proposal turns out to be economically viable? It would likely see a dramatic reduction in air travel between Australia’s east coast capitals.

    Sure, it would be unwise to totally discount our descendants a century hence (particularly on environmental issues) but making infrastructure demand forecasts a century out is an exercise in futility.

  7. Socrates

    John Doe

    It would be possible to extend a rail line in the corridor you describe to service the suburbs you mention. Fine. But that is my point – calling it an airport line then is just a label. There are many areas of outer suburban Melbourne that would like rail line extensions, and this one should have no higher priority than the others, maybe less, depending on cost and demand. It is a long way from the highest priority rail project proposed for Melbourne.

  8. john doe

    Avalon airport could be used as a second airport I guess. But I believe Tullamarine is adding the extra third runway so large passenger aircraft such as the A380 airbus can use it. Tullamarine will increasingly compete with Sydney for international flight traffic.

  9. Dylan Nicholson

    #15, I’d agree that’s the most likely scenario, but 35 years is a long enough time that a number of possible developments could lead to significant reduction in existing airport demand (*), whereas it’s virtually certain that more people will be living along various possible corridors between the city and airport by then.

    (*) One wild notion of the Elon Musk variety – that near-vertical and silent take-off becomes commercially viable, and hence minimal need for airports to be situated so far from city centres.

  10. john doe

    Dylan Nicholson #12

    ‘Personally I’d look at building a train line that would be useful even if capacity didn’t grow massively in the next 20-30 years’

    According to forecasts Melbourne’s population is expected be six million in 2031 and almost eight million by 2051. Melbourne will have a similar population to what London’s population is today. I would assume the airports passenger numbers and employee numbers will continue to increase.

  11. john doe

    Dylan Nicholson #10

    Someone also suggested extending the railway beyond Tullarmarine to Mickleham.

  12. Dylan Nicholson

    100 years? Nobody could possibly have predicted how airports today are used 100 years ago. It wouldn’t even surprise me (outside of the sheer fact of still being alive!) if today’s airports are largely abandoned in 100 years.
    Personally I’d look at building a train line that would be useful even if capacity didn’t grow massively in the next 20-30 years – i.e. that services parts of Melbourne not currently well-serviced by PT. The two obvious candidates are Maribyrnong and Keilor. You could in principal take the existing Racecourse line and extend it across the river through both those suburbs to the airport, though I don’t deny it would be a pretty expensive option.

  13. john doe

    Jacob HSR #9

    Infrastructure needs to be planned for the next 100 years otherwise it will become obsolescent. Considering the Tullamarine Airport growth projections the cheaper option would be a mistake. Also when the T4 terminal masterplan is complete the length of the airport will be 1.4km. For a passenger tired from a flight with baggage this walking distance would be difficult. Skybus also has two stops at T1 and T3. And train stations on the Melbourne network are on average about 800m apart.

  14. Jacob HSR

    john doe #6,

    Absolutely. People should stop thinking of this a merely a railway to the airport. Look at the KLIA Ekspress, the Delhi Airport Express, and Gautrain in South Africa.

    Sure they all get you to the airport within 30 minutes, but they also have other stations. So, have a station outside the airport! Maybe even 2 stations outside the airport!

    OR Tambo Airport station is overground.

  15. Jacob HSR

    john doe #7,

    Dubai Metro has air conditioned elevated stations. Hell, they have air conditioned bus stops!

    In Melbourne we have air conditioned waiting rooms at train stations. The whole platform does not need to be AC.

    KLIA has 2 airport train stations, but it is much busier than any airport in Oceania.

    If you build 2 underground stations at MEL, it would have to avoid the foundation pillars upon which the terminals sit. Therefore the stations might have to be deep underground?

    Underground stations are more costly to build, maintain, operate.

    If you build 1 overground station at MEL and have travelators or a PRT system to take passengers right in front of the check in counters, it might be quicker.

    The 901 Bus stops in front of T1.

  16. john doe

    Socrates said – “Melbourne is a busy airport, but has less than half the annual passengers of somewhere like Heathrow (30m VS 72m)”

    London Heathrow airport has five underground train stations one for each airport terminal. Melbourne should at least have two underground stations if you use the percentage quoted.

  17. john doe

    Considering the current expansion of the T4 terminal to Francis-Briggs road and the adding of a 3 runway. Currently 31 million passengers travel through Melbourne Airport, with more than 64 million expected by 2033. 14,300 employees and contractors work at Melbourne Airport. The airport’s infrastructure needs to be improved. Melbourne international airport will probably need two underground railway stations in my opinion. International and Domestic with the platforms 8 hundred metres apart. Directly underneath the terminals is preferable. It would allow an easier transition for commuters from train to a flight. Underground stations are also air-conditioned and more user-friendly.

  18. john doe

    Socrates 5

    They could add additional stations along the Albion alignment. The list would be as follows – Sunshine North – Keilor East – Airport West – Gowanbrae – Gladstone Park – Tullamarine. Adding these stations would make the Airport Rail-link more viable. Some trains would stop all stations and other would run express.

  19. Socrates

    A good article Alan. Even if the Melbourne airport rail link were sensible, why would you do it ahead of the many other public transport investments the city needs? At best this project will simply divert attention and funds, causing delay to what the city really needs.

    Melbourne is a busy airport, but has less than half the annual passengers of somewhere like Heathrow (30m VS 72m). Sydney has more arrivals (38m VS 30m) but the rail link there still nearly went broke.

    The history of airport rail lines in Australia is not encouraging. Sydney airport rail line now is only viable because of the amount of redevelopment and population living along the route since the Green Square area went residential. Less than 20% of rail passengers are airline passengers headded for the airport. The rest are commuters.

  20. William Holliday

    What about using Broadmeadows station? Only 5km away and from memory there is a bus that goes there from the northern end of the bus rank at the airport.
    In Sydney, we get off at Mascot, one station before Domestic to avoid the $12 charge, and take the 400 bus from around the corner (or walk – only 15 minutes).
    Of course the NSW government should remove the fee but then they wouldn’t have an excuse to feed a nice juicy expressway project to their developer mates.

  21. Jacob HSR

    The bit from Sunshine to Southern Cross has already been built as part of the regional rail link.

    All that remains is a 17km section from Sunshine to Airport.

    In Europe the 55km long Brenner Base Tunnel will cost €8.59 billion. About €156m/km.

    That will allow trains to go through them at 250km/h.

    So even if the bit from Sunshine to Tulla were a 200km/h tunnel, it should cost no more than A$3.5b.

    Or we could have the world’s busiest airport without a rail link. Another case of “Victoria is right, the rest of the world is wrong”.

  22. Alan Davies

    Former Vic Premier, Jeff Kennett, tweeted tonight that the owner of Melbourne Airport, “should build and operate the (rail) link”. The government’s role he says should be to provide access to the land and assist with approvals.

  23. hk

    Many in my cohort of The Age readers agree with your statements, but are also aware that The Age will continue running stories framed in a way that amps up airport rail; its demographic loves the idea. However I’d like to see it put greater effort into giving its readers a more balanced bedtime story.”

Share this article with a friend

Just fill out the fields below and we'll send your friend a link to this article along with a message from you.

Your details

Your friend's details