Is high-speed rail too good to be true?

After 30 years of reports on High Speed Rail (HSR), it’s hard to believe the country needs yet another one. After all, only last year we got the second volume of Anthony Albanese’s exhaustive $20 million study by AECOM into the viability of HSR from Brisbane to Melbourne.

But last Wednesday a consortium of HSR advocates led by think tank Beyond Zero Emissions (BZE) published a new report titled Zero Carbon Australia: High Speed Rail. I’m surprised only The Guardian and the ABC appear to have picked up the story, because the report has a number of extraordinary propositions.

Claims

The first claim is especially interesting; BZE estimates the entire HSR system from Brisbane to Melbourne could be constructed for $79 Billion (excluding rolling stock). Costs would be limited by smart design e.g. sharing corridors with urban services in some locations. Even so, the express travel time between Sydney and Melbourne CBDs would still be “less than three hours”.

In contrast, AECOM estimated the capital cost of the system at $114 Billion i.e. 44% more than BZE. AECOM assumed a higher level of service (e.g. 144 km of tunnel compared to BZE’s 81 km) and calculated the express trip from Sydney to Melbourne would take two hours forty five minutes.

The second proposition is astonishing; BZE reckons the entire 1800 km network could be built and operating by 2025 i.e. in ten years.

That optimism contrasts sharply with AECOM, who concluded the network would need to be staged to keep economic benefits positive and wouldn’t be fully complete until 2060 (i.e. for 45 years), although the more promising routes would be operational earlier i.e. Sydney-Canberra in 2035 and Sydney-Melbourne in 2040.

Indeed, AECOM recommended construction shouldn’t start until 2027 to allow for planning, consultation and environmental approvals, as well as pre-construction and procurement activities. In other words, BZE’s timetable envisages its network would be fully built and operational well before AECOM’s sees the first sod turned! (1)

Just as astounding is BZE’s claim that HSR would be profitable within five years of commencement (i.e. by 2030) and would earn enough to repay both its operating and capital costs from revenue (essentially ticket sales). It assumes average business and leisure ticket prices would be significantly lower than the current air fares (see exhibit).

AECOM, on the other hand, assumed ticket prices would be the same as current air fares. Revenue would cover operating costs but virtually all of the capital cost – $98 Billion – would have to come from taxpayers.

Scepticism

BZE’s contentions seem breathtaking given the conclusions of the AECOM study. Unfortunately, it’s very difficult to evaluate them because the numbers lack transparency; there’re supposedly various technical appendices that support the calculations, but they aren’t available to the public.

The different time frames for construction also mean it’s hard to compare important numbers like patronage. That’s not helped though by BZE making little effort to facilitate meaningful assessments against the obvious comparator i.e. the AECOM study.

Nevertheless some assumptions invite scepticism; for example, BZE’s conclusion that “HSR will pay for itself financially” is based on a 4% discount rate. I’d like to see some explanation for how costs can be limited by lowering speeds (e.g. by sharing existing rail corridors) yet the project will ostensibly still deliver a trip time that’s reliably and consistently “less than three hours”.

The claim that patronage will be 68 Billion in 2030 also requires explanation given AECOM’s estimate is for only 84 Billion in 2065 (35 years later!), notwithstanding a near doubling of total travel demand in the corridor over the intervening period.

Both estimates refer to patronage five years after commissioning of the complete system, but under AECOM’s timetable the  most productive elements of the network would’ve already been operating for many years by 2065 i.e. Sydney-Canberra for 30 years and Sydney-Melbourne for 25 years).

The notion that 1,800 kms of high speed rail line and 21 stations could be designed, built and start operating commercially in ten years is probably the hardest of all to swallow. Especially as it’s assumed there’ll be no implications for construction cost due to factors like potential skilled labour shortages.

The failure to plan properly in advance is one of the key reasons major infrastructure projects exceed estimated costs by nearly 50% on average, with overruns of 100% not uncommon. A technology like HSR that’s new to Australia is likely to be at the upper end of the risk curve and hence demands very careful pre-planning. (2)

Flexibility

What really incites scepticism though is that BZE has “form”; it has shown a willingness in the past to be “flexible” with the facts.

For example, I pointed out last September (Infrastructure: does getting the facts right matter anymore?) that the organisation’s assertion that Melbourne Metro could be built for a mere $3-4 billion is way below the accepted estimate of  at least $9 billion for this project.

BZE arrived at that figure by grossly under-estimating many costs. Here’s part of what I said about its methodology in a follow-up article (What does urban rail really cost to build?):

Nothing has been allowed for Land Acquisition; just $1.6 million for Earthworks and Substructure; $30.6 million for Superstructure; and a mere $9 million for Signalling and Communications…The estimates also make no provision for train maintenance and stabling facilities, for design and contractors overheads, or for provision of a train control centre.

The explanation for this “flexibility” was captured succinctly by a commenter on the first article, who described BZE as “the flip-side of the propaganda outfits of the extreme right.” The organisation seems quite happy to pick and choose the facts in pursuit of its mission.

Credibility

So I think the credibility of this report is beyond zero. It’s a marketing exercise and isn’t intended to be open to critical analysis; it’s purpose is political.

It doesn’t change the key facts about HSR i.e. that it would cost taxpayers a fortune; overwhelmingly benefit capital city business and regional leisure travellers; replace one form of public transport with another; and facilitate regional sprawl. It would also be a very expensive way of dealing with GHGs.

As I’ve noted before (High Speed Rail: would it run down Australia?):

There are many other ways that public funds on this scale could be applied for a bigger social, economic and environmental pay-off and a fairer outcome than HSR would provide. For example, just looking at infrastructure, public subsidies could be directed at improving public transport in the nation’s cities or shifting electricity generation to renewable sources.

_________________________________________

1. AECOM says an accelerated program could see a start on construction by 2022 (although the economic benefits would be lower and, presumably, the risks would be higher). Sydney-Canberra would commence in 2030 and Sydney-Melbourne in 2035.
2. There are other issues that require further explanation too e.g. assuming 100% renewable energy as a “free” benefit; assuming HSR would substantially weaken the case for a second Sydney airport.