Facebook Google Menu Linkedin lock Pinterest Search Twitter


Airports & aviation

May 16, 2017

Is an airport train to the CBD the whole story?

The Victorian and Federal governments have different strategies for Melbourne Airport rail but there's a lot more to the transport task at Tullamarine than a train to the CBD


Some of the route options considered over the years for a rail line from Melbourne Airport to the CBD – this one goes back to circa 2000/01!

There’s so much political smoke generated by the debate over a rail line to Melbourne Airport that even the key players can’t see where they’re going.

Last month the Victorian government said it will provide $10 million of Federal Government money to undertake a “detailed assessment” of the best route for an airport rail link, its estimated cost and how best to pay for it. Then last week, the Federal government said it will make $30 million of Victorian Government money (from the Asset Recycling scheme) available for the development of a business case for a rail link.

There’s no fire though. The Andrews government has no intention of starting construction of an airport line in the near future; it’s under political pressure from the Federal Government and wants to be seen to be doing something. The Turnbull government isn’t serious either; it’s playing politics in league with the state Opposition (see What’s Turnbull’s airport rail adventure all about?).

$30 million (it includes the state’s putative $10 million) is a mere wisp compared to a likely cost of $3 – 5 Billion to construct the rail link. But it’s around thirty times more than the former Napthine Government spent on its study of alternative route options for a rail connection (see Trains: should the ‘where’ come before the ‘why’?). It’s more than enough to blow away the smoke but it won’t start a fire.

My view on the airport line is the same as Infrastructure Victoria’s and the Victorian government’s (see Should Melbourne Airport rail be put on the front-burner?). Yes, it will be needed in the medium term (say 15 years); yes, I expect it would have a BCA > 1; and yes, it would be nice to have it now. But it’s not yet a necessity; SkyBus is an acceptable if flawed solution in the short-term.

Critically, there are much higher priorities at present for scarce infrastructure dollars. For example, here’s Gay Alcorn writing in The Guardian on the weekend about Melbourne’s outer northern suburbs:

I’m heading to Epping… (it’s) groaning with growth, desperate for schools and roads and public transport… The big issue here (is) the lack of infrastructure to support a burgeoning population. Its symptom? Grinding, demoralising traffic congestion…

Premier Andrews could do a lot in the outer suburbs with $3 – $5 Billion. He’s right in the short-term to put the everyday interests of Melbourne households above those of interstate and international visitors; above those of Melburnians who holiday once or twice a year; and ahead of business people whose travel is at company expense. He’s right at this time to prioritise projects like Mernda rail and the North East Link over one that, when it’s all boiled down, replaces one form of public transport with another.

As I’ve argued before (e.g. see here and here), it nevertheless makes good sense to start the detailed planning for a mass transit link now. Airport passenger forecasts aren’t fixed; the outcome could be higher or lower and government needs to be ready to respond quickly if necessary. Both the Andrews and Turnbull governments are on the right page there, even if their motivations are suspect.

The business planning process should include an examination of modes and routes, as well as the scope for reforming road pricing and airport parking pricing. But there’s a bigger problem the business case should also address; all those travellers who aren’t going to or from the CBD. While a large proportion of interstate and international visitors to Melbourne are headed for the city centre, most Victorians – including the many meeters and greeters – begin and end their trips from home. That includes residents travelling on business too.

Since over 90% of Melburnians live more than 5 km from the CBD, it’s not surprising that private vehicles dominate airport travel. SkyBus only has around a 9% mode share. It’s unlikely an airport train would change the balance dramatically. Consider that Sydney’s Airport Link has around an 18% mode share (see Is rail on track at Sydney Airport?). Brisbane’s Airtrain, which is a better comparator with Melbourne because it’s also an end-of-line service, wins about 10%.

The ambit of the business case should accordingly be expanded to look at ways of reducing car use by non-CBD airport travellers. Integrating the airport train with Melbourne Metro so that travellers from the South-East can make a single seat journey to the airport will help. But as Sydney and Brisbane show, most will prefer to drive themselves or take a taxi direct to the airport. Business travellers don’t pay their own fares and recreational travellers usually go in pairs or groups where they can share costs.

There’s a range of policy options that should be fully assessed. The fare structure for the train can have a significant impact on demand. But imposing a high tariff on airport parking and/or tolls on access roads, could have a much more substantial effect.

Another potentially far-reaching option would be to decentralise entry to the airport by incentivising parking at two or three satellite locations linked to the terminal by a dedicated airport transit system e.g. in the vicinity of Ballarat Rd, Sydney Rd, or Plenty Rd. It would in effect be a more sophisticated version of the current airport long-term carpark, using light rail or BRT in a dedicated orbital right-of-way (e.g. see What should we do about the airport?).

It could have a significant effect on the demand for road space within a large radius of the airport, as well as reduce total vehicle kilometres of travel. It could make particular sense for travellers from the west and north who might be reluctant to travel “backwards” to the CBD in order to get an airport train. Something similar at a handful of strategic locations on the Dandenong line – priced similarly to the current airport parking tariff – could also make train travel more attractive for south-east residents who would otherwise eschew feeder buses.

No doubt there are other possibilities. It’s important though to avoid the trap of thinking all airport trips must be taken entirely by public transport. That’s just not going to happen within any sort of plausible planning horizon, least of all in a low density city like Melbourne.

While the CBD is the largest single destination by far beyond the airport, it doesn’t account for the majority of airport-related trips. The airport itself, however, must handle all trips, both in and out; it’s a huge generator of traffic concentrated on one very small location. A CBD mass transit link is an important part of the solution but by itself it doesn’t address most of the airport problem in Melbourne or, for that matter, in Brisbane or Sydney.


We recommend

From around the web

Powered by Taboola


Leave a comment

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

11 thoughts on “Is an airport train to the CBD the whole story?

  1. The Ghost of Paul Mees

    Don’t you wish Melbourne had a competent agency to plan and integrate all public transport, including Sky Bus?

    We could call it Transport for Melbourne (because we’re a bit sad like that). Or shoot for the stars and call it Melbourne Verkehrsverbund…

  2. Itsumishi

    A couple of points Alan.

    Firstly you state at various points that all alignments/options/routes etc must be considered; but then proceed on the basis that the airport rail line must be connected to Melbourne Metro. Why? You argue it should be to provide one-seat journey’s from the South-East; but that’s only necessary if the intention is to pair it with the Dandenong line specifically; which I imagine will throw a significant spanner in the intended line-pairing between the Dandenong and Sunbury lines. It may make more sense to connect it to Frankston line through the city loop; or the regional Traralgon services. The last option also means the airport would be serviced with trains that have much better luggage storage.

    You’ve also said multiple times the airport will be an “end of line” station; but again this isn’t necessarily true. There is potential to link back to Sunbury; which then provides regional travelers from the north a rail-link to the airport that doesn’t involve heading to the city first.

    I’m certainly not suggesting this is the best option, just testing some of your assumptions.

    1. Alan Davies

      I don’t say the airport line “must” be connected to the Metro, only that it will make it more attractive to travellers from the south-east relative to driving or using SkyBus. Remember, routing the line via Metro is the government’s explicit policy. None of that contradicts my proposition that all route (and mode) options should be assessed in the new business planning process.

      Not sure what you mean by “spanner”; what’s the issue you’re alluding to?

      Successive governments have planned the line as terminating at the airport i.e. as an end-of-line station like the Brisbane airport station, Hurstbridge station, Alamein station, Lilydale station, and various country lines. The idea of diverting one or more existing regional lines via the airport, as proposed by Rail Futures (see Are regional dormitories the way to grow our cities?), is relatively new. As far as I’m aware it’s never been subjected to independent analysis. At this time I’d say it’s one of many marginal propositions (the former government’s study looked at 22 route options, some quite fanciful). Speaking strictly, I’d amend my words to say it’s very likely that a line terminating at the airport will almost certainly prove to be the best option.

  3. Jacob HSR

    I went to the airport recently, using the 901 or 902 bus, and the bus stopped far away from the terminal.

    Does SkyBus park far away from the terminal too?

    No wonder people want a railway station at the terminal.

    1. Alan Davies

      SkyBus stops at the entrance to the terminal building. It’s closer than the rail station in Brisbane.

  4. Cal

    I’m (still) strongly of the opinion that any solution must have connections to local areas on the way to the airport, such as stations/stops connecting with existing public transport, and also bringing PT to areas that currently don’t get good services. So, if via Albion, it should stop at Sunshine (at least!) and probably Footscray too, as well as at least one or two stations in the Niddrie / Keilor area (loved the above description of it as a “Bermuda triangle”!). If it is some sort of monorail stuck above the motorway via Essendon, then there should be a connection to the train line in the Essendon / Pascoe Vale area, and also to the tram in the Essendon Airport / Airport West area. Without connections closer to the airport, a lot of potential users get effectively locked out.

  5. George V

    My completely uneducated and stab in the dark solution would be a monorail from the Arden Metro Station (interchange platform built in).

    Southern Cross would be ideal, but I’d imagine that going over the airspace in the CBD would be unworkable and tunnelling under the CBD for a dedicated airport line would add billions to the price tag, so piggybacking from the new Arden station should be much easier and cheaper.

    With an interchange at Arden, all of Melbourne’s commuters would still effectively be one or two changeovers from the airport link, which isn’t ideal, but manageable (and how it works in many or the world’s great cities).

    With the Monorail, I’d go with stops at Vic Uni, Highpoint, and the Bermuda Triangle of Melbourne’s rail network, the East Keilor/Niddrie area. Depending on whether the entire length of the airport trip can be kept below 20 minutes, possibly also add another station in Airport West or the suburb of Tullamarine itself.

    Alternatively, the route could also go straight over Citylink/Tulla freeway, but I think the opportunity to service areas like East Keilor and Highpoint with some form of rail, albeit Monorail, is one that should be seriously considered, not forgetting the elimination of needing to “rent” airspace for decades from Citylink. Also, if the environmentalists allow the Monorail to straddle certain creek areas, very few residential areas would be impacted by the “skyrail” effect/look on this alternative line.

    Based on figures in earlier Monorail suggestions put forward by existing companies, a solution like the above should be doable for about $1.2 billion. Add in the inevitable non-partisan waste, stupidity, corruption and unions, let’s round it off to $2 billion.

    Now, financials. I just can’t don’t know where on Earth I’m going wrong, it has to be somewhere, because the entire concept seems an absolute no brainer in terms of return.

    I’m basing my numbers on this:
    “The airport expects passenger demand will exceed 60 million by 2033”

    Let’s say the equivalent of a $20 one way fare today is $25 in 2033. Let’s split the difference between the share numbers Alan has provided (18% Sydney, 10% Brisbane) in the article, and go with 14% share of trips to/from the airport using this new link.

    14% of Tullamarine’s 60 million estimated passengers per year equates to 8.4 million trips on the monorail. At $25 per trip, that’s $210 million per year in revenue. The whole thing pretty much pays itself off within 10 years and pretty much pure profit for the state thereafter (assuming driverless trains and reasonable ongoing admin/maintenance costs – ha!). If a certain percentage of people use the line meet/see off family, it’ll probably be even more revenue.

    And that’s excluding the:
    a) Potentially millions in long term car park fees at the new satellite car parks (I’m pinching Alan’s idea that he raised for stations across the route)
    b)Potentially tens of millions every year in normal, everyday myki trips and fares to/from the other three stations. $25 per person one way is only for anything that involves a trip to/from the airport (VicUni, Highpoint and East Keilor/Niddrie stations would fall under Myki system and charged under normal Myki rates).

    I’m certain I’ve forgotten to carry a 1 somewhere.

  6. Robert Dow

    Alan, the Brisbane Airtrain is not the best example. Poor frequency, poor span and connects with a relatively low frequency network. I think Melbourne could achieve results more comparable to Sydney with an MEL rail link.
    Having said that I agree that there are higher priorities. I regularly use SkyBus and it is fine generally in my opinion too. (I do tend to travel out of peak though).
    Robert Dow

    1. Alan Davies

      But there’s no guarantee a Melbourne airport train would provide a high level of service. After all, Brisbane could increase it’s mode share too if it ran higher frequencies etc but the operator chooses not to.

      Sydney runs 7 minute frequencies in part because the airport stations are “on the way” to suburban commuter destinations, but I’d be surprised if “end of line” Melbourne did better than 15 minutes. It’ll have those big 7-car Metro train sets with the option to move to 10-car train sets (243 metres long!) – the temptation to run at low frequencies could be more than the operator, private or public, can resist.

    2. skippy

      The Skybus service is perfectly acceptable for 23 hours a day. The problem is between 7:30 and 8:30 the freeway slows to a crawl and the 20 min trip can take up to an hour!

  7. Peter Thornton

    (and Ben Sandilands)
    You might like to look at my research on airport rail links which is on my website.


https://www.crikey.com.au/2017/05/16/airport-train-cbd-whole-story/ == https://www.crikey.com.au/free-trial/==https://www.crikey.com.au/subscribe/

Show popup

Telling you what the others don't. FREE for 21 days.