Would an airport rail link take us for a ride?

Possible corridors for an airport rail link identified by DoT

A senior economist at Essential Economics, Sean Stephens, has joined the debate about a rail link from the CBD to Melbourne Airport, arguing in The Age last week that “a rail link would help Melbourne maintain its world-class status, where visitors and locals could access Melbourne Airport, our gateway to the world, with ease and convenience”.

There are some misconceptions in the article and an evident misunderstanding of existing public transport services between the airport and the CBD.

The main supplier is Skybus, a privately operated and profitable operation that carries two million passengers per annum, or about 8% of airport passenger traffic. Skybus operates a 24 hour service (with 10-15 minute frequencies between 4am and 11.45pm). Because Skybus makes use of the emergency lane on the freeway, it takes 20 minutes from the airport to the CBD in the off-peak and up to 40 minutes in the peak.

Also, from next year, the Government will extend operation of the existing Frankston to Ringwood Yellow Smartbus service to connect with Melbourne Airport via Broadmeadows station. Buses will operate every 15 minutes with ticket prices based on the standard Metlink fare structure.

Mr Stephen’s only concrete criticism of Skybus is that the fares are too expensive and “well above those of comparable cities that provide a rail link”. In fact Skybus tickets cost $16 one way, much the same as those on the Sydney ($15) and Brisbane ($15) airport rail systems, notwithstanding Melbourne airport’s greater distance from the CBD. Skybus offers airport workers a discounted fare.

If there’s a case for pricing Skybus in accordance with the standard Metlink fare structure, as Mr Stephens appears to think, then the Government could subsidise Skybus in the same way that it subsidises other public transport services in Melbourne. As it happens, Skybus pays the Government a “dividend” of over a $1 million p.a. It’s probably the only on-ground public transport service in Melbourne that makes a profit.

Mr Stephens also criticises the forthcoming Smartbus service on the grounds that it will be “inconvenient for many travellers who need to get to the city quickly and would be of marginal overall benefit”. The reality of course is it’s not designed for time-sensitive travellers (almost entirely people on business) as they can use Skybus or take a taxi at company expense.

Smartbus will provide budget-conscious travellers who want to get to the CBD or other parts of Melbourne with the option of catching a train to the city from Broadmeadows station. It will also enable some travellers destined for the northern, eastern and southern suburbs to avoid the CBD altogether, as well as provide an alternative to driving for the many airport workers who live within the northern region.

Another of Mr Stephens’s contentions is that there are lower cost ways of providing rail. He argues that the 50 km Regional Rail Link project in Melbourne’s west, which is now in the early planning stages, has the potential to reduce significantly the cost of a link to the airport compared to previous estimates.

Just why this would cost significantly less than the Albion option considered in previous feasibility studies is not explained, but the key issue with using the existing rail network is that trip times would very likely be slower than Skybus. This would reduce patronage and hence the financial viability of rail. And the cost is still likely to be in the hundreds of millions, so there actually does need to be a business case for rail.

Mr Stephens also argues that rail can pay its way. He says that Brisbane’s Airtrain “announced a profit of $4.8 million in 2008 and now services nearly 2 million passengers a year”. This of course contradicts his claim that Skybus is too expensive. The fact is Airtrain is profitable because it charges $15 for a one way trip i.e. much the same as Skybus.

He also fails to mention that the operators of Brisbane’s Airtrain minimise costs by offering limited services. Compared to Skybus’s 24 hour operation, the last Airtrain service departs the airport at 8pm even though, like Melbourne, Brisbane Airport does not have a curfew. Services are also less frequent than Skybus. No wonder Airtrain now turns a profit if it doesn’t have to offer off-peak services – but what do travellers in Brisbane do after 8pm?

Another argument Mr Stephens presents is that analyses undertaken by his firm “show that a (rail) link would alleviate pressure on the freeway system”. It’s a pity we can’t review this study as it would be useful to know by how much traffic would be reduced (1%? 5%?), but there’s a more fundamental problem here.

The idea that public transport can provide lasting relief from traffic congestion is a common misconception. It has no more lasting effect than building more freeways or widening existing ones. Latent demand will inevitably fill up any spare capacity liberated by a train. Managing traffic demand on Melbourne’s freeway system requires smarter approaches, such as congestion pricing.

Like the arguments of so many proponents, Mr Stephens ultimately falls back on populism and “us too” logic. He says Melbourne needs an airport rail line because some (unnamed and unreferenced) report examined 52 cities around the globe and “found that 34 have airport rail links and 10 have rail links under construction or as a committed policy”.

I suspect that was the sort of thinking that started cities like Melbourne and Brisbane down the Wilbur Smith and Assocs path of building freeways in the 1960s.

We already have a good public transport system connecting Melbourne Airport with the CBD and it costs taxpayers nothing. It makes no sense to replace it with a new one that will very likely either provide an inferior timetable in order to be profitable or will require a public subsidy. In either scenario it is unlikely ticket prices will be any less than those charged by Skybus.

As I’ve pointed out before, the savings in carbon emissions from substituting a train for Skybus could be bought for vastly less than the cost of a new rail line.

The only real deficiency of Skybus seems to be that it’s a bus and most people with an interest in this issue prefer to travel in trains and trams (I know I do). But that’s a very poor excuse for building a rail line. The opportunity cost would be high – there are many other projects like providing improved public transport in the outer suburbs that would actually have real positive economic, social and environmental benefits.

In the long run it is likely that a rail line will be justified, but not yet.

(I’ve previously discussed the issue of a Melbourne Airport-CBD rail link here, here, here, here and here).


15 Comments on “Would an airport rail link take us for a ride?”

  1. Matthew says:

    The video they show on the Skybus is also highly annoying. Queuing for the Skybus at the airport is also not guaranteed to be smoke free, and the last thing you want after a long flight is to be queuing for an infrequent, uncomfortable ride, with an annoyingly loud video, in the presence of one of those twits who thinks that their right to take drugs in public outweighs my right to clean air.

    I get a cab to Broadmeadows and catch a train. Bugger the costs. And do the taxi drivers have a big whinge that they aren’t getting a big fare from the airport.

    Skybus and taxis are rubbish public transport. Trains are quality public transport. And the Skybus is hugely overpriced.

    I’d say build both one of the Albion options plus a link to the Broadmeadows line, and run it as a suburban loop service. The NW suburbs would have far better PT options. The train is a normal suburban line, with normal prices, and not rip off airport prices like Brisbane and Sydney. And Skybus and Citylink can go to hell. Win, win, win.

  2. Dave says:

    Alan, good rational arguments that I’ve similarly used. There’s a pre-occupation with ‘me-too’ grandstanding projects, to have the same as some other overseas city, or better than some other Australian city. Billions get wasted on poor value projects, like Airport rail links and stadiums. Meanwhile, schools, hospitals, community health facilities, parks, etc remain under-funded.

    However, raising them in an election year, in the hope that a party will pick one up as a pre-election announcement, is an effective strategy for those that stand to gain, but not have to directly pay, for the announcement.

  3. Q.Maisie says:

    When I read comments like Matthew’s that “Skybus and taxis are rubbish public transport” I realise why facts and figures just won’t sway people with emotional responses. I fail to see how rail is superior if it runs relatively infrequently and not when I need it (see my sad cold trip home from the CBD after 11pm – 40 mins wait for a train to come, so I can sit on dirty seats. yeah, quality – not.
    Alan, you say that in the long run the rail line will be justified. When? What’s the tipping point?

    • Alan Davies says:

      Rail is generally more efficient when there are very large numbers of people to be moved e.g. workers travelling to the CBD in the AM peak. As airport passenger numbers increase, the case for rail will improve. Buses can be made more frequent, more like light rail in their design and they can be made larger, but eventually economies of scale will favour rail if demand continues to increase.

      However the tipping point could be later for the airport than it would be in the case of a CBD-like situation, because air arrivals/departures are not as “spiky” as city centre commuting.

  4. Moss says:

    Hmm, interesting. It seems you totally debunk your entire argument with your very last sentence. If it certainly needs to be built in the future, then why not build it now? Why wait until all of the existing infrastructure is straining under the weight of increased passengers/cars/congestion? Very short sighted.

    As mentioned before in a previous post, building the airport rail link has to be done intelligently to work – such as laying down dedicated high speed track for an eventual extension north towards Sydney. It all depends on the terms of reference for any project.

    • Alan Davies says:

      It shouldn’t be built now because it is inefficient to provide infrastructure before the conditions that justify its construction have been fully realised. Otherwise it will be under-utilised for a period and, moreover, the Government will carry the risk that the forecast demand might not eventuate.

      It makes better sense in the interim to invest in alternative, more productive projects and build the infrastructure when it’s justified.

      I agree with you that any rail link would have to be done intelligently. It would of course be a no-brainer if a Sydney-Melbourne HSR were to be built, but who’s holding their breath on that one?

  5. Matty T says:

    Those emotional response people, like me, will actually be some of the people using or not using the services. If coming from Adelaide I may as well get the Firefly bus from Adelaide, as fly to Tullamarine and get on the Skybus. Ok it’s a slight exaggeration, but I am going to avoid the Skybus at all cost, including avoiding Melbourne. Melbourne these days for me is walking between the domestic and international gates (to save a bus trip between the terminals at Kingsford Smith). Life’s too short to actually bother with Melbourne.

    I do think this website thinks too small. A VFT station at Tullamarine with a direct service to Southern Cross, with the Upfield line to loop through the Broadmeadows line, and hit the Sydenham line at Albion, would make Tullamarine a northwestern hub for Melbourne. The website is called the Melbourne Urbanist, not the Melbourne Suburbanist after all. I wouldn’t have subscribed to this blog if I didn’t like reading Alan’s analysis (or even maybe deep down like Melbourne (no that can’t be right)) but a lot of it is too conservative for my tastes.

    • Alan Davies says:

      Ouch! I can see I’m going to have to write something shortly to set the record straight on this ‘big thinking’ stuff. For the record, the Melbourne Urbanist is about all of Melbourne (although I get the joke :-)).

      BTW, your VFT via the existing rail lines is not going to be a very fast train because those lines are pretty busy.

  6. david roman says:

    alan yes you appear to have made a reasonable case for the”current” argument of bus over proposed airport rail link. No mention of social costs of increasing frequency of bus journeys as demand increases/airport users over time etc (ie increased air pollution/traffic noise etc but probably minimal in the big picture)
    The Syd airport experience of rail access is not a pretty story!!
    original proponent went bust, then taken over by govt. Not to sure how profitable current operations are but the fare cost acts as a big disincentive for its use.

  7. I’ve been in this business for 20 years on three continents, and can assure you that rail to the airport’s political popularity is always several orders of magnitude above its actual ridership.

    As someone who flies frequently interstate in Australia, I can assure you that Skybus is far superior to Brisbane Airtrain, both in its frequency and its span. The only problem with that upon arriving at Southern Cross it’s not always clear how to use the rail system to complete an intra-CBD trip, due to the infernal complexity of the Loop routing patterns and the lack of clear information at the northern end of the station. That, however, is hardly Skybus’s fault.

  8. jack horner says:

    I support Alan. Points like ‘Skybus shows annoying videos’ are details that can be fixed. They are not a reason to spend billions on duplicate infrastructure for slight gain. You have to think opportunity cost – what else of more value could be done for Melbourne’s PT with the same money? Heaps, undoubtedly.

  9. […] a comment » Yesterday’s discussion of an airport-CBD rail link prompts me to look further at the idea that airport users have a […]

  10. […] I wouldn’t know where to start with this shopping list. Each proposal throws up issues that would require a separate post to consider properly (although I’ve previously discussed bicycle licensing and the airport rail link). […]

  11. […] that time hasn’t come. Not yet. I’ve outlined the case against an airport rail link before (here, here, here, here, here, here and here), but in summary the key objections […]

  12. […] too – it only carries 10% of all travellers (and a further 2% access the airport by bus). As Jarrett Walker observes, the political popularity of airport rail “is always several orders of magnitude above its actual […]


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