Australian high-speed rail has barely left the station – some experts say a new US project shows a better way

The Guardian

01-06-2025

Progress has been slow on the proposed high-speed rail line between Sydney and Newcastle since the establishment of the High Speed Rail Authority (HSRA) in 2023, and the federal government is yet to commit to building the megaproject.
But some experts say there may be a cheaper and easier way to do it, pointing to a US example of what can be achieved.
The Newcastle-Sydney high-speed rail line is estimated to cost at least $30bn and take well over a decade to build.
The HSRA has spent its infancy developing yet another business case, starting with Sydney-Newcastle, to ultimately form part of a Melbourne-Brisbane line in the second half of this century.
Sign up for Guardian Australia's breaking news email
Geotechnical drilling has begun to determine how to tackle the mammoth task of traversing the Hawkesbury River with services that can reach a top speed of 320km/h. Meanwhile, Infrastructure Australia is evaluating its business case and will shortly make a recommendation.
By contrast, a longer line linking the eastern outskirts of Los Angeles with Las Vegas at similar speeds is being built for a fraction of the cost and time.
The private-led Brightline West represents an approach that is in many ways the polar opposite of how Labor is pursuing the technology in Australia, experts say.
With construction estimates soaring, experts are now questioning if Australia is letting perfect be the enemy of good and risking a cost blowout that would crush the chance of the line ever being built.
The HSRA plans to establish a service between central Sydney and Newcastle's Broadmeadow, running end-to-end in an hour, with a stop at Gosford.
The journey on the existing Newcastle-Sydney line takes about 2.5 hours, almost half an hour longer than an express service achieved in the mid-20th century.
The HSRA chief executive, Tim Parker, wants to build dedicated dual tracks so that high-speed trains won't have to interact with slower trains, a contrasting approach to how high-speed rail expanded in most countries. Spain's high-speed train, for example, shares sections of track with slower services. While it must slow down in certain sections, this philosophy presents an opportunity to roll out fast train technology in stages and more cheaply.
Not only is building a dedicated line more expensive, the Sydney-Newcastle corridor will require an engineering feat to construct what would be the world's longest rail tunnel to allow trains to cross the Hawkesbury River without compromising speed.
'They've picked a starting stretch that is the hardest and most expensive part of the entire east coast,' says Garry Glazebrook, an associate professor at the University of Technology Sydney.
The business case has not yet been made public, but the price tag is expected to be eye-watering.
Modelling for the New South Wales government's 2019 plan to build a line without federal help reportedly predicted it would cost about $30bn and take 12 years to build just the first stage from Gosford through the Hawkesbury to a station at Sydney Olympic Park – a cheaper option than Central. The costs proved too great. The then Perrottet government abandoned the plan at the end of 2022.
Brightline West broke ground in April 2024. Despite covering a 315km point-to-point distance – nearly three times as long as Newcastle-Sydney – construction is expected to be completed swiftly, with services for the 2hr 10min trip to be operating in time for the 2028 Los Angeles Olympics.
The project's most up-to-date budget is US$12.5bn (about A$19.5bn), already overrun from a 2020 estimate of US$8bn, but still a fraction of the cost of the Australian proposal.
Brightline West – owned by Fortress Investment Group – is funded by a mix of equity, private finance, federal government grants and California and Nevada state-issued tax-exempt bonds for private ventures in the public interest.
It is entirely unrelated to the California state government's plan for an LA-San Francisco line, which has been held up as an example of disastrous infrastructure projects in the US. Approved in 2008, no track has yet been laid and cost estimates have more than tripled to US$128bn.
Donald Trump has vowed to block federal funds previously promised to the LA-San Francisco project, but administration officials appear satisfied with Biden-era commitments to the Brightline model.
Those costs and timeframe may appear ambitious, but Brightline has already established a Miami-Orlando line that reaches 200km/h – just shy of the high-speed definition. The first section, between Miami and West Palm Beach, broke ground in 2014 and was in service by 2018.
Brightline West's impressive price tag and timeframe are only achievable through several compromises.
Most of the line will initially run along a single track with passing loops , reducing speed accordingly.
It will also share its corridor with the Interstate 15 Highway. The track will be built down the I-15's median strip, with trains travelling alongside road traffic, similar to a number of existing rail lines in Perth.
This delivers serious savings. Building just one track is cheaper, and land acquisition and permissions are streamlined.
Perhaps the greatest compromise is that Brightline West will terminate in Rancho Cucamonga, on the eastern outskirts of LA. While avoiding costly construction in built-up urban areas, passengers will need to connect to existing, slower rail to make the 67km journey to Union station in the heart of the city's downtown.
Building anything in Australia is expensive, and rail even more so.
Alon Levy is the co-lead of the transportation and land use program at New York University's Marron Institute, and researches why construction costs differ between countries. He believes Newcastle-Sydney already has unnecessary costs baked in thanks to overly prescriptive demands from government.
'It's a foolish requirement this early on to not share any tracks with commuter lines,' Levy says. 'You leave less space for the experts to find cheaper workarounds and drive innovation.'
Australia's rail industry is another problem, Levy believes.
Australian governments rely heavily on consultants for infrastructure planning, a symptom of the hollowing out of departments of expertise.
'This problem is common in Anglosphere governments,' Levy says. They bring in outside consultants who work on one thing then disband and the expertise vanishes.'
Levy believes an Australian government – either the commonwealth or NSW – should establish a 'public sector consultant unit' for rail expertise, which it can then provide to other states to work on their projects at cost, a model successfully proven in France and Italy.
The profit motive of companies such as Brightline can also bring down costs by pushing the limits of what is possible.
'They go out to the market and challenge their suppliers … to push the technical limits, so trains can move through difficult stretches and save the company from spending billions moving dirt around,' one rail source who requested anonymity says.
'But in Australia, governments want already proven technology, they want a safer bet that is future-proofed, and that often means being overengineered and double the cost necessary. It's always got to be the gold standard.'
Rail unions also increase construction and operating costs.
'Unions have enormous say, it's just the way we work in Australia,' Glazebrook says.
Union pressure is broadly seen as having pushed NSW to build new metro tracks for driverless trains with a private operator.
A separate rail source, unable to be named due to their work, said the new metro extension under Sydney harbour could not have been built had it been subject to existing Sydney Trains union agreements, due to the line's gradient.
Herein lies a catch-22, Glazebrook says.
By adopting the proven European approach – staged building using existing tracks already subject to union agreements – construction and operation are beholden to stringent standards that increase the cost and timeframes of projects.
Building dedicated tracks is a way to bypass such pressures and can allow for the automation of driving and fewer staffing requirements, saving money on operations once the line is running. But construction costs will soar.
'You throw everything together and you get a real bloody mess,' Glazebrook says.
Experts are now questioning if Newcastle-Sydney is the smartest place to start.
Glazebrook still backs the proposal put by his group, Fastrack Australia, for the HSRA to begin by upgrading the Sydney-Canberra line in stages to high-speed capability. Its easier terrain makes it lower-hanging fruit, even if trains don't run along the Federal Highway's median strip.
Others say Canberra-Sydney more closely resembles the value proposition of Brightline West – the potential to capture market share where air or car trips have dominated.
Taxpayers currently foot the huge travel bill of bureaucrats and politicians flying between Canberra and Sydney. The existing train service takes about four and a half hours – a 90-minute high-speed alternative could attract that business to rail instead.
The existing Newcastle-Sydney service – ticketed at $11 – takes only an hour and a half longer than the proposed high-speed trip.
While local rail experts are not proposing to copy Brightline's model exactly, most agree there are attitudes worth adopting to avoid Australia's high-speed rail ambition being shelved.
'We don't want to let perfect be the enemy of the good,' Glazebrook says.