High Speed Trains between Canberra and Sydney

Current Issues Brief 17 1996-97

Matthew James, Technology Adviser,
Science, Technology, Environment and Resources Group

Denis James, Transport Adviser,
Economics, Commerce and Industrial Relations Group


Major Issues


Train Technology

  • TGV
  • Tilt trains
  • Maglev
  • Seraphim



  • Maglev
  • Tilt trains
  • State Rail Authority
  • Road and Air Transport


  • Reviews
  • Impacts

Economic Analysis


Major Issues

On December 4, 1996, the Prime Minister announced to Parliament that the Commonwealth had decided to proceed with a joint Commonwealth-New South Wales-ACT approach to the investigation of options to provide a commercially viable high speed train service. The Prime Minister stated that any future Commonwealth involvement in the project will be on the basis of no net cost to the taxpayer and the project demonstrating its commercial viability. He further stated that the Commonwealth would establish a competitive tender process and seek expressions of interest from interested parties. The Commonwealth would participate in a joint project control group with New South Wales and the ACT to oversight the tender process, according to the Prime Minister's statement.

The idea of a very fast train operation between Sydney and Canberra has a long history, especially in the context of earlier feasibility studies for a service between Melbourne and Sydney. A general lack of governments' willingness to commit to rail resources has not assisted plans. Most serious effort has involved consideration of the use of French high speed train technology based on the 'TGV' system to provide an 80 minute service, currently the subject of the Speedrail proposal. The TGV is an established high speed train technology, which commenced operations in France in the early 1980s, but is now in widespread use in many west European countries. The proven French technology however requires a high-density corridor and wide financial support for the concept to proceed, given the considerable investment required.

Other possible train technologies include overseas high speed railway operations, proven tilting trains and prototype magnetic levitation schemes. However, the former types do at least require an existing right of way with few curves plus quality trackwork and signalling. The Canberra-Sydney corridor does not possess these features, as evidenced by its current operations. Upgrade of the existing railway to enable tilt train operation would require expenditure of the order of $400 million. Some observers have questioned the viability of more advanced rail technologies such as magnetic levitation on the grounds that they are both technologically and commercially unproven at this stage. Another issue is the mix of passenger and any freight traffic, along with existing services.

The Speedrail consortium proposes the completion of a largely new right of way, of high standard, between the capitals, with a planned expenditure of from $1.6 to $2.6 billion. With hourly services, priced at about $75 for economy fares, it aims to capture passengers from existing rail, air, car and bus markets to ensure the project's feasibility. While French train builder GEC Alsthom has committed $200 million for the project, it appears that this amount is for the vehicles only, which the company could sell if the venture failed. Speedrail proposes operation by 2001, but only with the approval of all governments under a set regulatory and standards regime.

Previous government reviews of very fast trains provide guidance on matters of community concerns, land use and development, environmental issues and economic effects. Controls over land acquisition, route selection and noise and safety standards appear warranted, along with a coordinated Environmental Impact Statement by all involved governments. It remains unclear whether Speedrail can gain the private finance that it requires.

Finally, the economic viability of the project remains uncertain and governments require indemnity from claims should the venture make a loss. Assessments of viability are very much dependent upon estimates of the number of journeys likely to be made each year. Figures quoted by Speedrail show that initial patronage could be as high as 3.84 million journeys per annum. On the basis of such figures, Speedrail would be self financing. However, a study commissioned by a NSW, ACT and Commonwealth secretariat has indicated that patronage could be as low as 1.82 million journeys. At this level of demand, establishing the service would require a capital injection of public funds in excess of $1 billion, with recurrent public funding of around $400 million per annum for the first five or six years of operation. Speedrail has announced a further $7 million feasibility study to clarify such issues.


Current proposals for an advanced passenger railway operation between Sydney and Canberra stem from earlier efforts directed towards establishing such a system between State Capital Cities. In October 1981 the Institution of Engineers, Australia, proposed a Bicentennial High Speed Railway Project linking the five capitals of south-eastern Australia. The first stage involved improvements to the Canberra-Sydney route and the use of diesel-electric locomotives, to reduce train times from the prevailing five hours to three hours. The proposal did not gain government acceptance, but studies continued in CSIRO.

Out of this work grew a proposal for a Very Fast Train (VFT): a totally new, purpose-built rail link between Melbourne, Canberra and Sydney, using proven French 'TGV' rail technology. A 'joint venture' comprising BHP, Elders IXL, Kumagai and TNT promoted the scheme as a passenger railway to compete with all existing transport services. CSIRO studies emphasised the need for new track, as fast trains of the TGV type can negotiate hills with relatively steeper gradients than existing railways, thus minimising earthwork costs. A 1984 report stated that the Sydney-Canberra VFT journey would take an hour.

In 1986, the VFT Joint Venture emerged, comprising TNT, Elders IXL and Japan's Kumagai Gumi. It completed a pre-feasibility study in June 1987, a passenger market analysis in August 1988, a concept report in December 1988, a progress report in October 1989 and an economic assessment and evaluation in October and November 1990. Under considerable scrutiny, the Joint Venture initially promoted a coastal route via Canberra, Cooma and the Cann River Valley, leading to strong environmental objections; but it finally accepted an inland route via Canberra, Wagga and Albury, as a precursor to a greater network. A favourable Senate Committee review ended in mid 1991.

The VFT partners promoted the scheme as a private sector venture, subject to special taxation arrangements which recognised the long pay back periods associated with major infrastructure projects. In August 1991, Federal Cabinet rejected the joint venture's taxation treatment proposals, despite VFT arguments noting the long term $10 billion investment required. The VFT Joint Venture subsequently folded, but in 1993 plans for a new Sydney-Canberra line re-emerged from the Speedrail Pty Ltd Joint Venture, comprising an engineering consultancy, project managers and Anglo-French company GEC-Alsthom. Speedrail's proposal also relied on TGV technology, suggesting a travel time of 80 minutes. A feasibility study began in June 1994, receiving multi-government financial support, while alternative proposals and train technologies also began to emerge.

Train Technology

High speed trains operate at speeds above 200 km/hr, sometimes on upgraded conventional track designed to handle mixed freight and passenger traffic and sometimes on purpose built, dedicated high speed track. They tend to be most economic over distances from 160 km to 800 km, above which aircraft have significant travel time advantages for passengers. Where very high speed trains operate at speeds of 300 km/hr or above, they require separate rights of way of precise tolerance and clearance from slower freight traffic. The chief technological limitations on train speed are from aerodynamic effects, sound shocks, and wheel friction. These problems are overcome by the TGV, tilt train, Maglev and Seraphim designs.


The French train à grande vitesse (TGV) system incorporates a number of innovations including line design, aerodynamic rolling stock and reliable network control. It utilises overhead power supply and specialised continuous signalling and relatively steep gradients to avoid costly tunnels and earthworks. Train operation permits four minute spacing at peak periods at speeds of over 270 km/hr. While the TGV system was initially quite an operational and financial success, in recent years some planned extensions in France have met with resistance from governments and anti-development groups. Nonetheless, the TGV system forms part of a wider network of high speed European trains extending from the Netherlands to Belgium, France, Germany, Britain, Italy and Switzerland.

Recent TGV upgrades have featured onboard computerised control systems, fewer motors and better braking for use on new cross-country routes. Note that TGV can operate at lower speeds on existing railway track although some track modifications may be necessary. A derivative of the TGV provides Channel Tunnel service as the Eurostar train, while Spain has purchased TGV rolling stock for use on its special high speed routes. The Republic of Korea is constructing a TGV line to serve the Seoul-Pusan corridor. However, TGV only remains suitable for countries with available credit, high-density corridors and valid reasons for building a high speed line mainly for passenger use (UN 1988, 23).

It is also necessary to mention other counterparts such as the Japanese Shinkansen and German ICE trains. The latest Japanese electric units enable operational travel at 300 km/hr, while the German ICE trains travel at 250 km/hr. High speed trains also operate in Belgium, Italy, Spain and Britain, for example the British Intercity 125 diesel units. All of these systems have successfully provided safe, rapid and consistent passenger service for many years, although not always without considerable capital and ongoing operational public investment support. There has been some interest in Canada and the United States in building high speed rail systems, with various proposals underway in Florida, Ontario and Pennsylvania. Nonetheless, the TGV holds the world rail speed record of 515 km/hr.

Tilt trains

Tilt trains have suspension systems that allow them to take curves at greater speeds than usual. They have operated for some years in Europe using traction bogies with radially self-steering axles and a car body tilt control system, all under computer control. Tilt trains have extra brakes to cope with the higher speeds involved compared to standard units. The European companies Siemens and Adtranz (formerly Asea Brown Boveri; ABB) offer tilt train versions using Italian and Swedish technology.

In those countries, tilt trains have offered an economic solution to allow at least 25 per cent faster services on existing curvilinear routes without having to invest in new high speed train rights of way. Germany is presently introducing tilt trains on a number of regional routes which cannot justify upgrading to ICE standard. France is investigating use of tilt train services to supplement its existing TGV network.


Maglev is a generic term for a family of technologies in which magnetic forces suspend, guide and propel a vehicle, unlike the electromechanical techniques used by normal trains. Although conceived over thirty years ago, research and development has taken some time, with virtually all effort in Germany and Japan. At this stage, there are no long distance Maglev train services in commercial operation, so the economics of the technology remain unclear. A Maglev vehicle's magnets lock into a magnetic wave created by an alternating current fed to windings along the guideway. Variation of the electrical energy along the guideway controls speed to provide direct linear motion rather than rotational, conventional wheel motion to speeds up to 400 km/hr.

Maglev utilises an advanced system in which trains float above a guide rail due to a powerful magnetic force created by a linear induction motor. The linear motor operates like an ordinary cylindrical motor split down the middle and stretched out along the track. The train wraps around the guideway so that it cannot fly off and, as there are no wheels, noise is minimal and mostly due to air flows. Maglev trains use around thirty per cent less energy than standard trains at the same speed, and far less than cars or planes. Magnetic fields inside train cabins equal the Earth's field, but do change rapidly; a possible health concern. Maglev trains can reach 500 km/hr speeds and negotiate grades as steep as ten per cent with a track radius of 2.2 km for a speed of 300 km/hr; half the radius for TGV.

The levitation system uses the attractive and repulsive magnetic forces of electromagnets in the vehicle for suspension and guidance, while a synchronous long-stator motor provides propulsion and braking along the guideway. An electric travelling wave field generated by current in the windings of the stator cores pulls the vehicle along by attracting its suspension magnets, which also act as the exciter section of the linear motor. A given section is energised only when the train is crossing it to minimise energy losses. The first operational Maglev system operated at Birmingham Airport as a shuttle in 1984.

There are, however, two types of magnetic levitation: electromagnetic as used by Maglev and electrodynamic as applied by Japan. The electromagnetic system (EMS) relies on attraction between vehicle mounted electromagnets and others on the underside of the guideway to counter the weight of the train. The electrodynamic system (EDS) uses a greater gap with repelling superconducting magnets to counter vehicle weight, allowing less precision in guideway construction giving poorer ride quality. An EDS version in Japan came to grief in 1995 when a superconducting magnet failed, sending the train crashing onto its track. Another newer Japanese proposal involves an EMS configuration of magnets. Meanwhile, the German Maglev EMS has completed 150 000 kilometres of trips in experimental service.


Recently, engineers at the Sandia National Laboratories at Albuquerque, New Mexico have developed a way to operate an electromagnetically propelled train on conventional tracks (Kleiner 1996, 32-6). This segmented rail phased induction motor (Seraphim) operates on wheels at 300 km/hr over conventional but modified tracks. Conventional rail propulsion depends crucially on the friction between train wheels and the rail to drive the train forward. As train speed increases so does wind resistance, such that, by 300 km/hr, it is stronger than the frictional force, requiring elevated tracks to enable the air to pass below the train.

The principle of electromagnetic induction is that a current is induced in a conductor when a changing magnetic field is applied. The current sets up a magnetic field in opposition to the original field, resulting in a force that pushes the magnet away. To utilise this principle, rail tracks would have a series of aluminium plates or an adjacent segmented aluminium rail along the track, costing $0.2 million per kilometre. Seraphim's motors would consist of pairs of electromagnets arranged so that the segmented rail passed through the gap between them as the train moved along the track. Complex switching on the electromagnets as each segment passed between them would propel the train forward. Each locomotive would have up to thirty electromagnets powered by an on-board gas turbine generator.

However, large development problems remain including the complex switching, shielding passengers from the magnetic fields, resonance noise, cooling and wheel friction. Without levitation, the Seraphim train would subject track beds and rails to considerable wear. The gas turbine generator would add weight, mobile pollution and energy costs. Its linear induction motor would also add weight, unlike Maglev linear synchronous motors. Matters of brakes for safety also arise. Unlike Maglev, Seraphim could use existing tracks if ever proven.


Speedrail is a proposal for a new, very high speed, electric TGV rail service between Canberra and Sydney and major centres along its route, taking 80 minutes for an hourly service. The 270 km Speedrail route runs from near Canberra Airport, through North Canberra towards Goulburn and on to a Southern Highlands stop, then to Campbelltown or Glenfield, Kingsford Smith Airport and Central. It utilises the existing rail network in Sydney from Central to Campbelltown and 220 km of purpose-built, dedicated 350 km/hr alignment built close to the Hume and Federal Highways. The route avoids towns and has few major bridges, with tunnels only within Eastern Sydney.

In 1995, the Commonwealth, ACT and NSW Governments joined with Speedrail to fund project feasibility studies which appear to confirm its viability under current taxation law. The studies found that in 1994, the corridor attracted 10.2 million one-way trips, at a growth rate of ten per cent annually. Speedrail predicts its service to provide 4.3 million one-way trips in 2000, after a planned expenditure of $1.6 to 2.6 billion. Speedrail proposes economy fares of around $75, set at about half the equivalent air fares, along with discount categories. It sees benefits in fast, comfortable and affordable transport to boost tourism and regional growth. Another possible benefit is the use of Canberra as an international airport to alleviate stress of congestion at Sydney Airport. Air passengers arriving in Canberra could complete customs and immigration formalities on the train, just as occurs now on the Channel Tunnel Eurostar service. This would also help avoid any airport delays due to fog or curfews.

Among technical factors considered in the proposal are safety, noise, earthworks, curves and operations. Speedrail notes the safe operating record of high speed trains in other nations which could apply here. It claims that noise effects are no worse than for existing trains and would have less duration, albeit with greater service frequency. Construction costs would not have to include the deep cuttings and high embankments required by standard railways. Such earthworks standards would preclude shared operation with conventional freight trains although it would not preclude freight services for relatively non-bulky, high value-to-weight freight and time critical freight items. The line would have very high radii for both horizontal and vertical curves compared with current values. It would use a design speed of 350 km/hr to cater for future speed improvements.

In September 1996, media reports suggested that Speedrail sought $7 million for a comprehensive project viability study to investigate routes, environmental impact and social needs. This was despite a preliminary tri-Government study that reputedly suggests a need for $1.3 billion of public funds, plus a further $406 million over five years to subsidise operating costs. This figure arises from a patronage estimate of 1.8 million one-way trips, compared to other estimates of 3.8 to 2.8 million single trips. Nonetheless, GEC Alsthom has undertaken to invest more than $200 million in the project. Of the total project cost, about 15 per cent would be spend on trains such as GEC Alsthom supplies. Should the venture fail, they would be in a position to sell the trains to other countries.

The three elements of expected revenues, construction and operating costs were the subject of a financial analysis that incorporated forecasts of future interest and inflation rates. Following a travel survey, the Institute of Transport Studies at the University of Sydney prepared demand and revenue forecasts for Speedrail. Preparation of the capital cost estimate has included consideration of land acquisition, services relocation, materials, power, network controls, stations and support areas, bridges, earthworks, track laying, trainsets and management. The operating cost analysis includes train and support crews, control centre, customer service, power, maintenance and administrative costs.

According to Speedrail, following final approvals, construction employing 13,000 people could commence in 1998, for completion by late 2001. It would serve almost four million people, with a majority of the project funding spent in Australia. Further engineering, environmental, economic and heritage studies await completion by late 1996, followed by finance procurement. There is a specific opportunity to improve the existing SRA railway between Menangle and Mittagong through joint development with Speedrail. This 30 km shared section would save about 18 km compared to the current circuitous route via Picton. In the future, Speedrail believes that its network could extend to Melbourne, Adelaide, Newcastle, the Gold Coast and Brisbane.

Before then, Speedrail claims that it requires both government approval and support in order to attract and retain major private sector investment. While it believes in the financial viability of the project, Speedrail claims some unique requirements pertaining to the need for clear and unchanging approval processes, straightforward regulations, and endorsed standards for the construction and operation of high speed railways within Australia. None of these apparently exist at present and the project's future rests with the government agencies investigating the proposal, as do the futures of projects put forward by other competing interests.


Once Speedrail announced its proposal, other interested parties announced their own plans for the Sydney-Canberra corridor. The most visible competitor was probably the Maglev consortium, while a rival group demonstrated operation of a tilt train prototype on the route. As yet, there has been no suggestion for a Seraphim service, but this may come out of Maglev proposals.


The Transrapid Maglev in Germany aims to serve 14.5 million passengers a year by 2010 with a ten-minute frequency over a distance similar to Canberra-Sydney. Transrapid International is a consortium of major West German corporations: Thyssen Henschel (TH), Messerschmitt-Bolkow-Blohm (MBB) and Krauss Maffei (KM) that have developed Maglev over twenty-five years. Transrapid claims that its Australian system could complete 400 km/hr trips in 58 minutes, with three stops, at a cost of $4 billion. It would use overhead tracks sitting on 20 metre high columns, perhaps along the existing highway.

However, at this time, Maglev is not in commercial operation, nor totally technologically proven, but only runs on a 31 km test track in Germany. The German Government has promoted construction of a 283 km track Maglev between Berlin and Hamburg under a $4.6 billion guiderail grant plus $2.6 billion of private funding. A planned 20 km service between Orlando Airport and Disneyworld in Florida was cancelled when the cost estimate reached US$1 billion. A 427 km Las Vegas-Anaheim proposal has also failed. The Japanese Maglev has operated only on a 7 km test track, with no commercial use as yet.

Tilt trains

During 1995, the NSW State Rail Authority trialed a Swedish X2000 tilting train on the Canberra corridor, claiming notable travel time savings. However, most of the savings came from the deletion of normal station stops and special operational arrangements in the Sydney network to avoid conflicts with suburban trains. The X2000 actually achieved standard trip times of three hours and twenty minutes at best. The existing one-hundred-year-old rail corridor has gross inadequacies resulting from high frequency of tight curves and a very circuitous alignment, especially in the Southern Highlands region, when compared to the Hume and Federal Highways. Existing poor rail signalling, trackwork and management do not help efforts to improve safe running times on the route.

According to press reports, various tilt train proposals are also in the offing. ABB Daimler-Benz Transportation (Australia) Pty Ltd (Adtranz) reportedly promotes a tilt train using existing track. The Spanish company Talgo estimates that improvements to sections of the existing railway to enable 220 km/hr travel could reduce travel time to 150 minutes still competitive with aircraft, but at a cost of around $400 million. Improving the track could involve expenditures from $794 to $1,350 million, says Speedrail, while Adtranz claims only $200 million for a 150 minute trip. Presumably these tilt trains would all be diesel, not electric and would have to mix with slower freight haul mainline traffic, perhaps taking between 200 and 250 minutes for the one-way trip, little better than existing travel times.

State Rail Authority

At present, standard SRA passenger trains run along the corridor at speeds averaging 80 km/hr taking four hours for the twice-daily 270 km trip. This is a result of the line and operational deficiencies, well evident when the 160 km/hr XPT trains, a British HST derivative, ran along the route in the 1980s. SRA 'Countrylink' passenger services run at a deficit, with common media reports of booking difficulties and service problems experienced by users. The SRA has plans to provide improved services over upgraded track in due course, but this will depend on NSW State Government budget allocation which remain uncertain.

Road and Air Transport

By the year 2000, a dual carriageway highway will operate between Canberra and Sydney, the first time such a road has existed between Australian capital cities. Work is now underway on completing duplication of the last two sections of single carriageway on the Federal Highway. When completed, this work should allow a consistent three-hour travel time for private cars (80 per cent of trips) and buses (8 per cent) taking passengers between the capitals. Hourly bus services already provide the bulk of cheap surface public transport movement, rather than the slower and more costly rail services which only handle 2 per cent of intercity trips at present. Speedrail predicts a 20 per cent shift of car and bus passengers to its services, plus all SRA train travellers.

Australian airlines carry 6 per cent of corridor travellers, taking total journey times, including allowance for check-in plus Sydney airport delays and city connections, of 2 to 3 hours. Speedrail expects to take two-thirds of airline market share, due to the provision of more reliable and cheaper services at similar timeframes. Airlines will, however, no doubt respond.


Analysis of the earlier VFT proposal provides some guidance in assessing the feasibility of the current Speedrail and other claims. The fact that the VFT idea foundered serves to show that such schemes are not necessarily guaranteed of success, whether for good or bad reasons. There are wider issues than high speed rail proponents tend to flag that bear on their proposals, as appeared during VFT scoping studies and government reviews of them.


In 1990-91, the Victorian Government Very Fast Train Review Panel inquired into the social, environmental and economic issues relevant to the VFT. This included matters of land acquisition and compensation relating to development, infrastructure requirements, land use and access, economic and employment impacts, community amenity and social impacts, as well as a range of environmental issues. While supporting the proposal, it was cautious about the development of the VFT and its effects on the factors listed above. In the event, suspicions of a dubious land acquisition practices appear, in part, responsible for a strong, public anti-VFT lobbying campaign that no doubt assisted the proposal's demise.

The Senate Standing Committee on Transport, Communications and Infrastructure conducted an inquiry, from 1989 to 1991, into aspects of the proposal for a very fast train. Its final report made recommendations on the need for public comment on environmental impacts, protection for the Commonwealth in the event of project failure, study of the relationship between a VFT and a Fast Freight Train (FFT) plus the shared use of the VFT corridor, as well as scrutiny of noise and safety standards. In its favourable view of the project, the Committee did not assess the matter of economic viability, but it did rule out use of Maglev over the longer corridor route. However, possible joint usage of the shorter corridor was a matter of prime importance to the Committee.

In 1989, the ACT Government established a VFT Advisory Committee to undertake its own analysis of the issues. The Committee addressed matters of routing, urban and rural impacts, economic analysis, social effects, commercial viability, public consultation, noise and government finances. In June 1990, the Committee presented its final report having many specific recommendations concerning environmental impacts, financial implications, public interest aspects and social impacts. While generally supportive of the VFT concept, the Committee urged further in depth study and community consultation on matters such as fair dealing practice, noise effects, safety matters, land acquisition and tax incentives.

It is not clear to what extent the NSW Government has reviewed any VFT proposals. Recent press reports suggest that the NSW SRA has acted to delay and frustrate Speedrail's proposal as the latter project would undoubtedly serve to cancel existing rail passenger services. This would leave the SRA providing only freight services on the route, unless these too were offered by Speedrail, as vaguely proposed at this stage. The NSW Government has tended not to make clear statements either way on any proposals. It would be foolish for Speedrail to proceed without clear NSW, ACT and Commonwealth Government approval along with resolution of the freight and FFT issue.


Analysis of the aforementioned government reviews allows some assessment of probable impacts of Speedrail's proposal. An economic analysis appears in the next section. Among the issues considered here are land acquisition, infrastructure effects, community issues and environmental aspects. None of these matters seem to offer insurmountable barriers.

The ACT and Victorian reviews recommended that leasing of land along the corridor may offer a fair deal for landowners, the public interest and project operators. Nonetheless, government power over land acquisition would still be necessary, along with a means to share equitably any changes in land value between operators and the community, noting the leasehold system in the ACT. A joint government approach to the project is necessary.

The ACT review indicated a need for considerable work on route definition to avoid local disruption. As well, it cited a need for further investigation of greater cross-track or underpass access for emergency use, humans, fauna, native and farm animals. The Senate and ACT Reviews urged the imposition of national noise and safety standards for high speed rail. Matters of visual amenity, electricity supply and technology transfer also arise.

Among social impacts, we might expect a boost to the ACT economy along with better linkages to Sydney and the region. However, in the absence of appropriate land use planning controls, the project might also encourage unwanted corridor development and urban sprawl. The need for public consultation throughout the project remains. The Senate and Victorian reviews note the need to preclude financial intervention by Government if the project fails.

Among environmental effects, apart from severance factors mentioned above, are the need for control of weeds, soil erosion, siltation, drainage and dust. Heritage and archaeological sites will also require protection. The Victorian Review notes the need to avoid reserves, remnant native vegetation and wetlands. The ACT Review found a need for coordinated Environmental Impact Statement by all governments concerned with the project.

Economic Analysis

It is difficult at this stage to fully assess the economic viability of the various high speed rail proposals put forward for the Canberra-Sydney corridor. While there would appear to be a reasonable amount of certainty about the costs of providing such services, the potential revenues involved are far from easy to predict. Furthermore, the attractiveness of any proposal to the Governments of the Commonwealth, NSW and the ACT will reflect the financial demands that a project might impose upon the public purse. These issues, among others, have been canvassed in the report prepared by a secretariat appointed by the three Governments. The report was presented to the Governments concerned at the end of September 1996, but since it contains information that is 'commercial in confidence', it has not been publicly released. However, it would appear that the report has cast doubt on the economic viability of both the very high speed train and the tilt train proposals.

The Speedrail consortium has suggested that the cost of providing its high speed train link between Sydney and Canberra will be between $2 and $2.6 billion. Of the estimated capital cost of the Speedrail proposal, around 85 per cent would be spent on civil engineering and fixed installations with the remaining 15 per cent relating to the trains. The cost of the Transrapid project has been reported as being as high as $4 billion, although even this figure is somewhat surprising as this would imply that the Canberra-Sydney project would be costing only around one-half of the similar-length system between Berlin and Hamburg.

Both Speedrail and Transrapid are estimating the standard, one-way economy fare between Sydney and Canberra to be approximately $75. However, Speedrail has indicated, as one might expect, that a range of discount fares would also be made available. At these fare levels, both systems would expect to generate a substantial amount of patronage. This patronage would have several sources.

Firstly, there would be some diversion of travellers from existing alternatives. Such a diversion is likely to be at the expense of the airlines, coach services and existing train services. Many travellers who use private transport may weigh up the convenience of having a vehicle to get around in once they have reached their destination. Given that a car can hold several passengers, the cost per passenger of a car journey may also be relatively low. Coach and train services are currently cheaper than the mooted high speed train fare, but journey times are considerably longer.

The train would be expected to compete mainly with the airlines. In terms of total travel time (including the journeys to and from the airport), the train would essentially match the airlines. The problems of curfews and delays at Sydney Kingsford Smith Airport, the impact of fog and other weather conditions on airline services between Canberra and Sydney and possible user perceptions of greater safety with rail transport would further enhance the attractiveness of high speed rail. Furthermore, the proposed high speed train fare is around one-half of the airline economy fare. It might be noted that Speedrail has tended to play down the possible impact of a high speed rail service on the provision of freight services between the two cities, but this could well prove to be a further area of competition with other transport modes.

A second group of potential users are those who may be induced to undertake travel they might previously not have undertaken. This could occur if the time/price mix offered by the high speed train encourages travel that would not have been undertaken at the higher time/price mix on offer from other transport modes. Some travellers may even be attracted to the service because of its special or 'novel' characteristics. If the train service were to complement increased use of Canberra airport by international and domestic travellers (and perhaps even freight consignors), this could be a further source of induced demand.

Speedrail also points to the rapid growth in demand for travel. It notes that travel to Canberra has been growing at between 15 and 20 per cent per annum, while the number of international visitors to Australia has been estimated to increase by between 8 and 15 per cent per annum over the period to 2000. The existence of a high speed train service is seen as attracting a significant proportion of this travel market. On the basis of its market surveys to date, Speedrail estimates that by the year 2000, as many as one quarter of all persons making journeys between Sydney and Canberra would use the high speed train. It is even estimated that 14 per cent of private vehicle users would be likely to switch to rail. Two studies have been undertaken of initial potential passenger numbers. One study, undertaken by French National Railway System analysts (Sofrerail), put likely patronage at 2.82 million per annum, while Sydney University's Institute of Transport Studies put the figure at 3.84 million.

The Report of the three-State secretariat apparently questions the likelihood that such patronage will be generated. A study commissioned by the secretariat indicates that as few as 1.82 million trips per year would be made on the train. Whilst acknowledging that the project would be economically viable if patronage figures were as high as those put forward by Speedrail, the secretariat allegedly claims that, on its lower figures, the project would not be attractive to private investors. Premier Bob Carr is reported as saying that the project would require over $1 billion of public money to get it started and recurrent annual injections of around $400 million over a five to six year period to run it. These estimates do not include the cost to the Commonwealth of tax concessions associated with infrastructure bonds. Given the crucial importance of patronage estimates in judging the viability of the project, Speedrail has stated that it intends to undertake a $7 million market research and viability assessment study to clarify this issue, hopefully once and for all.

The policy position taken by all Governments concerned is that no direct public funding should be expected for the project. However, the Minister for Transport, Mr Sharp, has been quoted as stating that the Speedrail project would be eligible to raise finance through infrastructure bonds and that this could cost the Commonwealth many millions of dollars. Speedrail claims that, if its estimates of patronage are correct, the only direct government financial input it would require would be the resumption of any private land required for the right of way and for the construction of two platforms. It is estimated that this would cost $100 million. Speedrail has suggested that government (mainly the NSW Government) should borrow the funds to purchase this land, with Speedrail paying them back at one percent over the government borrowing rate. This proposal has reportedly not found favour with the NSW Government.

The Governments concerned have stated that they would leave any high speed rail proposal up to the private sector to implement. In the case of Speedrail and Maglev, these projects could well be managed privately since they would involve their own tracks and rolling stock. However, it is difficult to see how a tilt train proposal could operate without a substantial amount of public sector involvement. The major cost advantage of the tilt train scheme is that it can utilise existing track, which implies quite a deal of coordination with NSW State Rail. In fact, unlike the two very high speed rail proposals, there would appear to be little interest on the part of the tilt train manufacturers to operate services between Sydney and Canberra. These manufacturers seem content simply to sell the appropriate train sets to State Rail.

However, even the tilt train proposal is not without its costs. Apart from the cost of the train sets themselves, around 70 km of track would have to be substantially upgraded simply to enable the journey time to be reduced to around 2 hours. The capital cost of establishing a tilt train service could be as high as $1.3 billion. While this cost is lower than that of the two very high speed train proposals, it would be false economy if the tilt train were to generate a lower rate of return than the other contenders. Given the major improvements which have been made to the Hume Highway, travel times on the tilt train would not be significantly less than road travel times and certainly would not be competitive with the airlines. Whether the tilt train would attract sufficient patronage to justify its cost would appear doubtful.

Certainly, there might be a case for operating tilt trains on other routes within NSW where extensive track upgrading would not be required. State Rail might therefore see some advantages in acquiring tilt trains and using them more extensively over their rail network, including the Canberra-Sydney route. The tilt train option might appear even more attractive to State Rail if the purchase of such trains could be financed by infrastructure bonds, although at this stage all applications for infrastructure bond approval from the Development Allowance Authority have been put on hold.

Addendum 16 January 1997

The risk guarantee aspects of the project include wider commercial considerations than discussed in the paper. There are certain commercial requirements for projects of this nature.

There is new information to hand on a TGV-type high speed system linking Miami, Orlando and Tampa. However, the Florida project remains a proposal at this stage.

The CIB paragraph on Maglev noise states that it "is minimal and mostly due to air flows"; in other words the principal source of noise is aerodynamic flow. A video provided by Transrapid Maglev clearly demonstrates that its noise levels are lower than comparable high speed trains. An independent noise measurement alongside railway right-of-ways provided the results that, at a speed of 300 km/hr, Maglev was 84 dB, ICE high speed train 87 dB and TGV type train 92 dB, with the latter two figures rising with track wear. Note that a reading of 3 extra decibels (dB) means a doubling of the sound pressure level.

While GEC ALSTHOM has indicated a willingness to invest "up to $200 million", it has already spent some extra funds on feasibility studies. In any event, most press reports stated that the company would spend "more than $200 million". This is a choice of words really.

Above the CIB states that "Speedrail expects to take two-thirds of airline market share" plus a 20 per cent shift of car and bus passengers to its services. However, it appears that Speedrail will actually draw most patronage from road travel, yet take pressure off Sydney Airport and contribute to the further development of Canberra Airport.


ABB 1990, The Swedish High Speed Train: A New Era of Inter-City Travel, Asea Brown Boveri Traction, Vasteras, Sweden.

ACTVFTAC 1990, Report on the Very Fast Train Project: Public Consultation and Major Issues, ACT VFT Advisory Committee, Report to Chief Minister, Canberra, April.

Budd, D., 1994, 'Speedrail: a new High Speed Railway between Sydney and Canberra', Reforms in Urban Transport Conference, Sydney, March.

Budd, D., 1995, 'Speedrail: high-speed rail in time for the Olympics?, Transport Engineering in Australia, vol.1, no.1, Institution of Engineers, Australia, December.

Budd, D., Bosci, F. 1995, 'Speedrail: a new High Speed Railway between Sydney and Canberra', Proceedings 2nd National Rail 2000 Conference, Adelaide, April.

Maglev Transit 1990?, Magnetic Levitation Demonstration Project: Executive Summary, Maglev Transit Inc, Tallahassee.

SSCTCI 1991, Aspects of the Proposal for a Very Fast Train, Senate Standing Committee on Transport, Communications and Infrastructure, Australian Senate, Final Report, March.

UN 1988, New Transport Technologies: review of recent technological developments in the land, air and maritime transport sectors, United Nations, ST/ESA/196:E.88.II.A.14, Department of International Economic and Social Affairs, New York.

VFTRP 1991, Final Report, Very Fast Train Review Panel, Report to the Victorian Government, Melbourne, June.