Will networking make public transport the mode of choice?

What's the biggest danger?

According to a report in The Age last month, new research published in the latest issue of Australian Planner shows that higher suburban densities are not a precondition for vastly better public transport. Reporter Andrew West says:

City dwellers have been presented with a false choice – live in apartments and enjoy good public transport or retain the house and land and rely on cars

The research by Dr John Stone and Dr Paul Mees contends that it is not necessary to intensify land-use across the whole city before significant improvement in both patronage and economic efficiency of public transport becomes possible.

They say the contribution made by urban consolidation “to recent public transport patronage growth is modest and makes little impact on the density of the whole urban region”. Most residents of Australian cities will continue to live in houses and suburban subdivisions that are already built so “alternatives to the car will need to be effective at existing urban residential densities”.

They argue instead for a ‘networked’ model of public transport. Improving the way existing public transport resources are managed – especially by providing higher frequencies and improving coordination between services and between modes – will yield significantly higher transit patronage in the suburbs without the need for broadbrush increases in density.

I’ve argued before that increasing residential density, by itself, will not necessarily increase public transport patronage significantly, much less shift travellers out of their cars in large numbers.

I’ve also argued that there are generally better gains to be had from using existing resources more efficiently rather than relying on strategies based around huge new infrastructure investments or massive land use changes.

And I think the idea of networking public transport is absolutely critical. By embracing transfers, networking provides faster travel paths to all parts of the metropolitan area than is possible by radial routing.

However it’s not obvious to me that ‘networked’ public transport, by itself, would have the sort of major impact on mode share in the suburbs implied by The Age’s report. I can see that it would make public transport much better for existing users and I’ve no doubt it would increase patronage, but I’m not persuaded that it would be enough to address the ‘false choice’ that The Age says Melburnites have been presented with.

The key problem for public transport in the suburbs is competition from the car. For the vast bulk of trips – especially the non-work trips which comprise the lion’s share of all journeys – cars are simply faster and more convenient. Suburbanites recognise this and are prepared to pay more to get that speed and flexibility.

The ‘networked’ model relies on transfers. Even with the sorts of ten minute frequencies proposed by the authors, a journey involving even one modal change will in the great bulk of cases take longer than one taken by car. A properly networked system is a huge boon for ‘captive’ public transport users but doesn’t do enough to redress the advantages of the car for those who have a choice of mode.

The only situation in which public transport really comes in to its own is for trips where the car has been neutered. In Australian cities that’s primarily work journeys to the CBD.

The reason public transport wins a high share of trips to the CBD is largely because the car is throttled by factors like high parking charges and traffic congestion. Unlike in the suburbs, public transport in the CBD and surrounding areas can out-compete cars. And the CBD has that advantage principally because high density non-residential activities, especially jobs, create conditions that are antithetical to cars.

With public transport’s share of CBD commutes now approaching 70% in Melbourne, there’s probably not a lot of room left for further increases in its mode share (even assuming adequate spare capacity). Most of the remaining car travel to the CBD is by workers whose vehicles are subsidised in some way by their employers e.g. cheap or free parking. They will be very hard to shift.

Work journeys, however, only account for a minority of all trips. And the CBD only has a minority of all metropolitan jobs (in Melbourne, around 15%).

The great majority of suburban residents work in dispersed locations within the suburbs where the obstacles to the car are nowhere near as formidable as they are in or close to the city centre. And the great majority of trips are in any case accounted for by non-work trips, which are the least amenable to public transport.

Astronomically higher petrol prices could possibly throttle the car in the suburbs, but it’s very likely drivers would respond by switching to more fuel-efficient cars and alternative fuels.

Spreading traffic congestion will be a major disincentive to car travel in the suburbs, at least for work journeys, but that’s an inefficient way to promote mode shift. More importantly though, jobs and people are likely to continue to do what they’ve always done in Australian cities when confronted with congestion – spread further out to escape it.

Road pricing is the obvious way to make public transport more competitive with cars in the suburbs.

Drs Stone and Mees are right to advocate better networking of public transport – it’s essential. We’re not going to get lots of new rail lines in the foreseeable future, so the only feasible way to get a decent public transport system is if we maximise the interconnectedness of the existing infrastructure. That will make public transport users better off and win some trips away from cars, but it won’t have a really dramatic impact on mode share unless simultaneous and complementary action is taken to make cars less competitive.


9 Comments on “Will networking make public transport the mode of choice?”

  1. Matthew says:

    @Alan when you say “However it’s not obvious to me that ‘networked’ public transport, by itself, would have the sort of major impact on mode share in the suburbs implied by The Age’s report. I can see that it would make public transport much better for existing users and I’ve no doubt it would increase patronage, but I’m not persuaded that it would be enough to address the ‘false choice’ that The Age says Melburnites have been presented with.” it makes me wonder if you have read Paul Mees’ book, A Very Public Solution. I recommend it highly.

    After reading that book, with a day to spare in Toronto one time (the book compares Melbourne and Toronto) I decided to have a day catching the subway and streetcars exploring Toronto, and it was kind of great having a walk up network, with very little wait times, and I was transferring from one service to another with abandon. Transfers are not scary, and allow for better quality, more frequent services on fewer (but interconnecting) routes. Single seat rides into the CBD from 25km away on a bus are not practical. The “network effect” beats Victorian-aged radial network hands-down.

    I’m convinced.

    • Alan Davies says:

      Thanks Matthew, I have read it and think highly of it (haven’t read his new one though – when I enquired at the time of release, Borders were asking $130 for it!!!).

      As I say above, networked systems are indeed superior to radial systems. But travellers don’t like transfers. The authors themselves acknowledge that “while transfers create many new travel opportunities, they also impose inconvenience”. They also cite other researchers who say “surveys asking what passengers . . . dislike about transit find that transferring is at or near the top of the list . . . [So, traditionally], transfers are avoided”.

      Bear in mind the question I’m specifically addressing is whether networking the system is enough to attract suburban drivers out of their cars and on to public transport on a large scale. I don’t think it is – it’s desirable, but the car also has to be tackled directly.

      • Dave says:

        How much of the fear of transfers in rooted in historical (or current) experience of transfers being painful?

        Some transfers are easy – getting a tram to your local station, in which the tram stops right near the station entrance, and might not even be thought of as a transfer by the traveller (it’s a single trip to work/destination, even with two modes). Alternatively, bad transfers (the unfortunately still common type here in Australia) usually involve some walking between modes, usually a wait of around 10 minutes, often little realtime information on the connection (inducing uncertainty and anxiety) and poor waiting conditions (rarely sheltered or in a vibrant environment).

        Once you break down some of these barriers, the network and transfers become more useful and less feared by planners and travellers alike. However, you are right that the car has to be tackled too – nothing exists in splendid isolation…

      • For what it’s worth Amazon have Mees’ Transport for Suburbia at $US55 which may end up more reasonable, even with postage. Not sure what local bookshops are now stocking it at, but I suspect it won’t be much less as the publisher’s site lists it at $87.

  2. Luke says:

    “For the vast bulk of trips – especially the non-work trips which comprise the lion’s share of all journeys – cars are simply faster and more convenient. Suburbanites recognise this and are prepared to pay more to get that speed and flexibility.”

    I think that it could be worth noting here that a significant proportion of suburbanites’ non-work trips will involve more than one person. It may be true that some single-person car trips have a cost higher than the equivalent trip on public transport, but the cost of a private trip for two or more is almost the same as the cost for one, whilst the costs of the public option multipy.

    • Mahyar says:

      Provided one has a car and car ownership remains pretty cheap. Otherwise, the multiplier effect of PT is still cheaper than all the costs associated with car ownership and maintenance.

    • Chris says:

      If the infrastructure was provided, and the car issue was tackled head on i.e. 30km speed limits in designated areas, then the short non work trips would be frequented by bicycle.

  3. […] is taken from the same article that I mentioned in my last post. The authors, Dr John Stone and Dr Paul Mees, find there is only a modest relationship between […]

  4. I think you and Dr. Mees are in agreement regarding the need to also alter policies on cars to cause a mode shift.

    He certainly mentions in Transport for Suburbia (which is worth the read, I agree with Stephen, check Amazon. I think I paid about $75 including shipping).


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