Why is rail patronage 50% higher in Sydney than Melbourne?

There’s a Wikipedia page “Commuter rail in Australia” which has a quick snapshot of the country’s heavy rail systems.

(The term “commuter rail” is problematic. In a North American context it often refers to systems that really cater only for peak commuter/work trips. Australian rail systems sit somewhere between here and metro systems, and are generally referred in Australia to as suburban rail. Confusingly the term sometimes used in North America is “regional rail”, as in serving an entire city/region. But I digress.)

Given Melbourne and Sydney now have similar populations, I thought a comparison of the two cities and their rail networks might be interesting.

(excluding Sydney Metro)
(excluding V/Line
but including Stony Point line)
Metropolitan population
Urban density423 per sq km [1]508 per sq km [1]
Train operatorSydney TrainsMetro Trains
Number of lines8 plus branches15 including all branches
Track length1588 km
(Route length 721 km)
998 km
(Route length 430 km)
Fleet size230 eight-car trains
= 1840 carriages
226 six-car trains
= 1356 carriages
Usual all day frequency15 minutes10-20 minutes
Usual evening frequency15 minutes30 minutes
Timetabled stops per weekdayabout 46,000
(average 262 per station)
about 44,000
(average 198 per station)
All-night servicesNo
But there are Night Ride buses every night
Yes, hourly on weekends
But no all-night service during the week
Car parksMore than 30,000 spacesMore than 40,000 spaces
Annual ridership377.1 million (Source) [2]244.1 million (Source)

[1] These density numbers are probably unreliable. Better measures take into account statistical areas which might actually be rural or not available for settlement. Charting Transport has some good discussion on this.

[2] Sydney Metro patronage was about 2 million per month in late-2019, so extrapolated out would be about 24 million per year.

So then, how can we explain why Sydney’s ridership is more than 50% higher than Melbourne’s, when the population is only 5% larger?

It’s not station car parking. Melbourne has more car spaces at stations than Sydney.

Craigieburn line at Parliament Station, 2:57pm on a Monday

Longer routes probably help a bit, giving Sydney’s train network a bigger footprint across the city, though Melbourne actually has more railway stations.

Fares potentially might be a factor, though it’s really swings and roundabouts. Some (especially shorter trips, and off-peak) are cheaper in Sydney. Others (especially longer trips and those involving multiple modes) are likely to be cheaper in Melbourne – though Sydney’s $50 weekly cap makes even long trips pretty price competitive for everyday users.

Until now, Sydney has had a ridiculously cheap Sunday cap of $2.80 – compared to $16.10 Monday to Saturday. That changes from next week to $8.05, but it will also apply on Saturdays and public holidays.

Better train frequencies in Sydney would certainly be helping, making Sydney’s trains a more attractive proposition for more trips, especially when connecting from other lines. Melbourne has a few 10 minute lines, but these are in the minority, and drop back to 30 minutes in the evenings.

Remember, transport is supply-led. The better the service, the more people use it.

What about train journeys to the CBD? City of Sydney has 246,343 residents plus (under normal circumstances) 615,000 visitors/workers. City of Melbourne in contrast has 148,000 residents plus 763,000 visitors/workers. So in theory, Melbourne has a busier CBD.

Sydney train at Museum station

Broader urban planning is almost certainly a factor. Large suburban hubs such as Parramatta serve as alternatives to Sydney’s CBD, and importantly, are located on the heavy rail network.

One way of comparing is to look at each city’s biggest (suburban) shopping centres:

1Macquarie CentreY (Metro)ChadstoneN
2Westfield ParramattaYWestfield Fountain GateN
3Westfield Warringah MallNHighpointN*
4Westfield Bondi JunctionYWestfield KnoxN
5Westfield MirandaYEastlandY
6Castle TowersY (Metro)Westfield SouthlandY
7Macarthur SquareYWatergardensY
8Westfield HornsbyYWestfield DoncasterN
9Westfield PenrithYPacific WerribeeN
10Bankstown CentralYNorthlandN
Heavy rail access9/10Heavy rail access3/10

*We could quibble about Melbourne’s Highpoint – it has tram/light rail access. But this doesn’t provide the speed or capacity of heavy rail. (Knox almost got tram access… but the Bracks government extension fell short.)

In terms of heavy rail access, Sydney is clearly ahead. (And I didn’t count Sydney’s biggest shopping centre: Westfield Sydney in the CBD, which also is close to rail.)

This makes a huge difference not just to mode share for a substantial number of journeys, but also to heavy rail patronage.

Are there other factors in Sydney’s success? Almost certainly. Some more digging might involve looking at:

  • the spread of peak vs off-peak journeys
  • mode share/feeder service provision for easy station access
  • travel times including express trains
  • CBD commuter train mode share, and the role of higher capacity double-deck trains
Town Hall Station, Sydney

Planning for success

Sydney’s rail network might not be perfect, but they seem to be doing a lot of things right – patronage 50% higher than Melbourne’s doesn’t happen by accident.

Can Melbourne do better? I’m sure we can.

Urban planning outcomes can take decades to achieve. Are we heading in the right direction? Hopefully – it’s hard to say. We’re seeing urban consolidation in some areas, but the State is still investing in huge motorways (Sydney too, mind you) and the big centres like Chadstone and Fountain Gate keep expanding, with little or no effort into improving their public transport access, despite strong demand.

But if we can’t instantly fix urban planning, rail service frequencies are a lot easier to improve in the shorter term. It’s high time the Victorian government got serious about providing rail services that meet the expectations of a 21st century city of 5 million people.

Melbourne transport

High-density around railway stations: a good idea, if done well (but that’s a big If)

I think it’d be true to say that Melbourne hasn’t done high-density development in the suburbs very well.

For example, this monolith in Camberwell, a bit too far away from the railway station, out of scale with (some of) the buildings around it, and I’m sure not well liked by many of the locals.

Camberwell Junction, July 2013

But that doesn’t mean high-density around railway stations is a bad idea.

The topic came up yesterday in an Age report that Metro’s plan to upgrade the Dandenong includes such development around stations such as Murrumbeena.

Residents of Melbourne’s politically sensitive south-east face the possibility of high-rise development at their rail stations including Murrumbeena, under a confidential deal between the Napthine government and a consortium led by the city’s private rail operator.

The deal for the proposed multibillion-dollar upgrade of the Pakenham-Cranbourne rail corridor – contained in documents leaked to The Sunday Age – includes a specific clause about development around sites identified for level crossing removals.

In some ways this shouldn’t be any great surprise — Metro’s parent company in Hong Kong makes a lot of money from development around stations, and there’s been a lot of talk in the past few years about development around stations helping to pay for grade separation. The tiny (in comparison) development of a cafe at Caulfield was a flop, but a grade separation, new station and re-development of the whole precinct would actually work… if done well.

Population growth is happening. Planned, targeted in-fill development is better than never-ending sprawl, and better than a free-for-all that destroys local streets and leads to more car dependence because you get lots more people living where public transport isn’t convenient.

I’ve lived in Murrumbeena twice — for a couple of years last decade, as well as in the 80s when I was a teenager. In that time the shopping centre has always moribund. To an extent, the railway line split it east-west, and the busy road split it north-south, and it could never compete with Chadstone and Carnegie, both nearby. Getting a lot more residents in the immediately vicinity of the station could re-vitalise it, and make much better use of the land currently used for parking.

I’m not sure about how high they should go. Chris Hale proposes 15 stories in The Age article; having seen the blocks go up around Footscray, and the Camberwell example above, I don’t think in most suburbs (outside the inner city, at least) you’d want to go above 8-10 for now, staggered downwards as you get further from the centre/station.

There are provisos to all of this, of course:

  • as Chris Hale says, good design, including green space — Melbourne seems to be lacking good examples, but experts cite cities such as Vancouver as having got this right
  • mixed use development so people can do much of their daily shopping without going elsewhere
  • in a some areas, particularly inner-city, you’d want to be sensitive to the heritage strip shopping streetscape
  • upgrades to rail services to 10 minutes, 7-days, so it really is an option
  • ditto, upgrade to local bus services to other major nearby destinations (in the case of Murrumbeena, the obvious one is bus 822 to Chadstone and Southland)
  • bike paths/lanes on nearby corridors/roads
  • limit of one car park per residence, with the option of none
  • in fact, set up a car share pod or two to further reduce car ownership
  • given much of the most obvious land for development is currently station carparks, I’d imagine it might be politically courageous to end up with a net reduction in car spaces, though improvements in bus services could counter this. A reduction in spaces is perhaps avoided via development such as in Elsternwick, where existing parking was converted to multi-storey.
  • and of course in the case of the Dandenong line, the grade separation should include provision for future track amplification and platform extensions

The fact that this one is being planned in secret is obviously a concern. And is Metro’s “value capture” going to actually save the government (and taxpayers) any money? It’s not clear.

But that doesn’t mean that the concept high-density around transport hubs is a bad idea, provided the community get some say, and if it’s done well.

Based on past Melbourne experiences, that’s a big if.

What do you think?