Eyes wide open: the cost of commuter rail

There are a number of folk getting quite excited at the prospect of some serious new investment in heavy commuter rail, as part of the Brisbane 2032 Olympics infrastructure legacy, but also with projects like Melbourne’s suburban rail loop. There’s no doubt that improved public transport connections are a good thing. But they can - in the case of heavy rail - be also very costly. The phrase “be careful what you wish for” seems apt.

The Moreton Bay (Redcliffe) line north of Brisbane serves as a recent and useful example of cost and commuter usage. None of what follows suggests it should not have been built. But the costs of the project, relative to its actual use by the community and the ongoing operational costs, are a reality the taxpayer is entitled to be aware of. If they’re unaware, politicians and consultants extolling the upsides should also be honest about the costs, because they are playing with other people’s (taxpayer) money.

The rail line to Redcliffe was first proposed in 1885. It was finally opened 130 years later in October 2016 (proof that trains really don’t run on time?) The line is a 12 kilometre extension of heavy gauge, dual track electrified commuter rail from Petrie to Kippa Ring.

The project cost $1.2 billion and includes six stations – Kippa Ring, Rothwell, Mango Hill East, Mango Hill, Murrumba Downs, and Kallangur. That’s $100 million per kilometre, including the stations, or $100,000 per metre of line distance covered.

According to Queensland Rail data, the total number of passengers boarding the line via the six stations in 2018 is a combined daily weekday average of 4,291. That was pre-Covid, and numbers broadly dropped by a third due to the pandemic. The pre-Covid numbers are a fairer representation of demand so we’ll use them.

The busiest station was Kippa Ring (989 people boarding daily), followed by Mango Hill (895) then Kallangur (703), Mango Hill East (672), Murrumba Downs (562) and Rothwell (470).

Assume that daily boardings are a proxy for the number of users (most people who board also disembark as it’s typically a return journey, and most people do this regularly for their commute). So assume it’s broadly the same 4,291 people using the line each day for commuting. There are also a small number using the line on weekends but this is unlikely to be for work so as a congestion buster, it doesn’t really qualify – heavy commuter rail is never justified as a weekend recreation service but as a weekday commuter service.

If the new line cost around $1.2 billion and is used by around 4,291 regular commuters, that’s a capital cost of $279,655 per passenger. That’s a big number.

Plus there’s the operational costs. Prior to Covid, the Queensland Government subsidy per rail passenger, per trip, was $21.15. (This number doubled when numbers fell in Covid). That subsidy is in addition to the train fare. So every time a passenger pays their fare, the taxpayer pays an additional $21.15. Per person, per trip. That’s $42.30 for a round trip, in addition to the fare.

(Buses were subsidised $4.02 per person per trip – a fraction of the cost of rail. Not only are buses much cheaper as public transport but they also make use of the existing road network, they can be re-routed in response to changing needs, and can be efficient with fewer passengers – none of which you can do with rail).

So for each of the 4,291 regular rail users on this line, the taxpayer is contributing an average $42.30 per day, or $211.50 per week per user. That’s $907,546 in total per week for all users. Call it a million dollar rail fare subsidy per week for travellers on this line alone.

I’m told the famous Sir Leo Hielscher, when head of Queensland Treasury, used to try hose down politicians’ enthusiasm for heavy commuter rail projects by telling them it would be cheaper by far to give every person likely to benefit a free taxi voucher, for life.

Scarily, these projects are often proposed on the assumption that they will meet a commuter demand from frustrated motorists sick of congested roads. Which they would, if most of those motorists were trying to go where the train lines take them – the inner city and CBD. But with more than eight in ten residents working in suburban locations (not typically served by rail), this math was never going to add up.

It didn’t stop the pollies though. Said acting Premier Jackie Trad at the time: “A project like this, that will see 600 new trains go from the Redcliffe Peninsula to the city CBD every week, is a fantastic initiative for workers in this area.” Even though the overwhelming majority of workers in the area don’t work in the CBD.

Opening the rail line, then Prime Minister Malcolm Turnbull also made a raft of silly comments including that “Realistically, someone could jump on a train here in Kippa Ring and use our public transport network to visit the beaches of Gold Coast or Sunshine Coast.” Memo to Malcolm – let’s buy them airfares to visit Fiji. It would be cheaper.

The same story carried this justification for the line: “The Moreton Bay region is one of the fastest-growing in Australia with the population of 375,000 tipped to reach 500,000 by 2031.” True. But this does not mean they’ll be commuting to the Brisbane city centre for work.

It added: “Over 80 per cent of people in the region (83%) use a private car to drive to work each day.” True again, but they do so because the car will take them when and where the train doesn’t.

Are there better ways around this, which could be explored? One might be to rethink the idea that suburban rail stations are mainly a place for CBD-bound commuters to board for their journey to work. Instead, why not explore options for suburban stations as the destination? A place to get off because your work in a mixed-use precinct is within a short walk of the station? There are stations that meet this potential.

Or what about locating schools next to stations? Would it not be a good idea to be able to get off the train when the station is incorporated with the school, for safety and convenience of students and staff and parents. There would seem to be ample land surrounding stations in some cases. Very large on-grade areas set aside for commuter parking aren’t always fully utilised. Like this image of Murrumba Downs station (taken on a weekday in pre-Covid July 2019, with barely a third of spaces taken). Why not a multi-deck park and ride to cater for commuters (the cost to park for the day could include your rail fare) which in turn frees up some of on-grade land for a school?

Whatever the options, it would seem sensible to evaluate future proposals for heavy commuter rail through some different lenses. If we are to add massively expensive heavy rail upgrades to meet future transport needs, it would be nice to think that we are fully informed of likely costs, realistic projected user numbers and that all options to maximise the economic and community value of the stations have been explored.