Facebook Google Menu Linkedin lock Pinterest Search Twitter


HSR High Speed Rail

Dec 19, 2012

Will High Speed Rail save the cities and develop the regions?

Taking growth pressures off the city fringe while simultaneously developing the regions is one of the key arguments used to support proposals for a High Speed Rail network on Australia’s east coast. But is it a cost-effective way of doing either?


Proposed east-coast HSR alignment

Late last month the Federal Minister for Transport, Anthony Albanese, told Parliament that HSR would be costly, but had to be considered in the context of the regional development opportunities it would provide.

The recent discussion paper on Melbourne’s metropolitan strategy emphasises the potential of developing regional centres. It says land should be reserved for a very fast train service between Melbourne and regional centres.

In its new report on the economic benefits of HSR, the Greens emphasise the potential of fast trains to relieve growth pressure on the fringe of our big cities by promoting regional centres as alternative locations for commuters.

HSR users will be able to live in Albury or Shepparton and commute on a daily basis to the Melbourne CBD. This will take pressure off Melbourne’s straining outer suburbs while also boosting regional economies. Perhaps more importantly, it will help get people into the property market by enabling them to buy property in regional Australia while commuting to where the jobs are.

According to the Federal Government’s Phase One HSR Report, the estimated travel time by HSR from the centre of Shepparton to Southern Cross station in Melbourne would be 55 minutes and from Albury/Wodonga it would be one hour and five minutes.

Whether or not HSR-led regional development is as sensible and attractive as it sounds is an important public policy question. The benefits and achievability of regional development – including avoiding fringe growth in major cities – are often asserted but seldom explained.

It’s a pertinent question because past efforts at regional development in Australia were underwhelming. This time regional development is being cited to justify an investment of around $125 billion – that’s enough to pay for 25 years of Gonski’s proposed reforms in education.

In earlier eras like the decentralisation push of the Whitlam Government, it was hoped new regional jobs would be created by inducing major manufacturers to locate in growth centres. This was plausible at the time because the low cost of transporting goods (as distinct from transporting people) meant that many manufacturers were relatively location-independent.

Thus the centrepiece of the Whitlam initiative, Albury-Wodonga, was able to induce (with large subsidies) some significant firms like Borg Warner and Uncle Bens to settle on the Murray.

But employment in manufacturing has declined precipitously in Australia since the 1970s. It was expected Albury-Wodonga would grow to 300,000, but today its population is only around 105,000.

Canberra wasn’t the subject of an endorsed decentralisation program, but it also provides a salutary lesson. It’s the national capital, but has a population of just 350,000.

That’s despite good air services, a freeway connection to Sydney and a very high standard of living. It’s also despite specialising in knowledge-intensive industries and having one of the best-educated workforces in the country.

However what the Greens advocate isn’t even regional development in the sense it was conceived in the Whitlam era i.e. self-containment of jobs and residents.

Rather, the Greens envisage regional centres would simply be dormitory suburbs connected by HSR to the jobs and economic activity in the capital cities.

New residents have to eat, so dormitory suburbs in the regions would certainly generate a lot of jobs in sectors like retail and personal services. Indeed, that’s what happens now when suburbs on the outskirts of capital cities are settled.

But it wouldn’t create self-sustaining growth in regional centres in the way it was hoped manufacturing might in the 1970s. Nor would it drive growth in the same way that a few regional centres have benefitted from expansion of the mining and tourism industries.

It’s also highly unlikely a centre like Albury-Wodonga would attract CBD-type businesses from capital cities in the event air services were replaced by HSR. Knowledge economy firms already have the choice of moving to the outer suburbs but they usually don’t because they prefer the benefits of agglomeration.

But if they wanted to move, the outer suburbs of capital cities would give them access to a far larger workforce and greater proximity to other firms than any regional centre could provide. And from a worker’s point of view, the median journey to work time is relatively short e.g. in Melbourne’s outer suburbs it’s 30-35 minutes door-to-door.

The cost of commuting from regional centres to capital cities by HSR would also be a major impediment to the Green’s plan.

There’s no estimate for Albury-Wodonga or Shepparton in the Phase One HSR report, but as a guide, note that it assumed the one-way fare from Gosford to Sydney would be $26 (51 km by air; 75 km by road). That assumes no capital cost is recovered in fares.

It proposed an additional subsidy for Gosford commuters to reduce that fare to $14.25 one-way. Albury-Wodonga is 280 km from Melbourne, so they’d pay somewhat more.

There’s an equity issue here – only those on decent incomes could afford to pay upwards of $30 a day in fares. Only those working in CBD jobs could make the station-to-station trip from Albury-Wodonga to Southern Cross Station in Melbourne’s CBD in the assumed one hour and five minutes journey time.

A further key issue is there’s no great advantage in building new suburbs on the edge of Melbourne versus building them on the edge of a regional centre. The required levels of infrastructure and community facilities are the same.

Like capital cities, regional cities also have problems with water supply, especially inland cities. Nor is there any reason why broadacre land should be significantly cheaper in a major regional centre than in the immediate hinterland of a city like Melbourne.

Moreover, the impact on the environment and on agricultural land would be little different on average between the two locations. In fact the pressure to provide smaller lots might be lower in a regional centre.

One potentially important difference, though, is transport infrastructure has to be upgraded to accommodate growth in capital city fringe suburbs. That expense has to be compared against the cost of HSR.

But the calculation also needs to account for the cost of consequential upgrades to transport systems within regional centres as population grows. That needs to be studied but my guess is the HSR option would be much more expensive.

Of course it doesn’t necessarily follow that all regional HSR commuters would be diverted from the urban fringe. Around half of all new dwellings in Melbourne are constructed within established suburbs. The proportion is considerably higher in Sydney but lower in other capitals.

Given the likely level of fares, a significant proportion of HSR commuters might be comparatively well-heeled residents of middle and even inner ring suburbs. They might be tempted to offset their long and expensive commute by living in much larger properties than they could afford if they stayed put.

So I don’t think there’s a compelling argument that HSR would promote regional development. Nor do I think it would be a cost-effective way of responding to growth pressures in capital cities.

The case has not been made that Melbourne and Sydney are anywhere near being “too big”. Their problems are a failure of politics and planning, not scale or urban form.

And I think the idea of subsidising extra-long commutes at a time when many observers think long commutes are bad for health and community cohesion deserves further consideration. Indeed, I’m surprised the Greens, of all parties, ignore this issue.

As a related aside, Anthony Albanese quietly let it be known last month (in answering a question in Parliament from Craig Thomson) that the Phase 2 HSR report which was due this month now won’t be “handed down until early next year”. I haven’t seen this reported anywhere.

HSR boosters won’t take much comfort from his final sentence:

We know that in Europe and Asia High Speed Rail is a growing part of the transport solution. Of course, they have much denser populations than Australia does across the vast expanse of this continent. So the challenge here is much greater.


We recommend

From around the web

Powered by Taboola


Leave a comment

Your email address will not be published. Required fields are marked *

17 thoughts on “Will High Speed Rail save the cities and develop the regions?

  1. Socrates

    HSR will NOT save regions. If anything, it will lead to greater business concentration. See any big towns cropping up between Tokyo and Osaka? Nope. The growth is at the ends of the line.

    If the train stops frequently it will be too slow to beat air travel, and hence uneconomic. Infrequent stops mean towns bypassed. There may be a benefit to places like Newcastle, wool ongoing and Geelong in making commuting to the CBDs viable, but that is it.

  2. Peter Connell

    sydney cheap & cheerful 2/

    yes, the topic is hsr, but the thrust is green travel

    sydney has fantastic rail assets way under used.

    I cant cope with assembling the precise numbers

    but in short, sydney is a stupid radial system unlike london or any sensible city

    3 main lines are unconnected, yet only ~2km apart ~20k from cbd & 12km~ from nearest link (sydenham or wolli creek)between them.

    try going cross town in the southern half of sydney. Its hopeless & so are the private buses.

    ~ punchbowl (bankstown) to riverwood (east hills line) 2km over flat cheap land

    then after a few stations – narwee to penshurst or beverly hills to hurstville (illawarra line/south) is only another 2km – may need a tunnel tho.

    simple & cheap

    maximise what you have before the mega projects & it would save way more emissions

  3. Peter Connell

    Here is a cheap & cheerful suggestion for sydney

    central/airport/wolli creek/glenfield/campbelltown. 100mph should be easy on that new track

    same to the south – some good track. city,sydenham, wolli creek, sutherland, waterfall

    there is an upside to unambitious top speeds

    big roomy carriages!

    In the US n/e corridor

    amtrack is booming despite crummy track & woeful speed

    DC to NYC is now their domain & the jets are hurting a lot.

    nyc to boston, much the same.

    its the default business mode of travel Not cheaper, just better.

  4. Russ

    Duke, true, but the discussion above was concerning total demand and the relevance of population density.

    There are some up-to-date figures here on the mode share vs. air for different times (time being more relevant than distance). Until recently there weren’t many lines longer than 600km, but the graph there (with the back-to-front axis) indicates that journey times of 4 hours attract 40% mode share. On a corridor as busy as Melbourne-Sydney that is still a lot of passengers. Note that Paris-Marseille is on the long side, which is why I chose it above.

  5. duke the lost engine


    A simple gravity model is appropriate for overall demand on a corridor, but not for estimating demand for a particular mode. i don’t know the figure for paris-marseille specifically, but generally air tends to be chosen over hsr for distances over around 500km

  6. Russ

    Does noone use gravity models anymore? Obviously there are several caveats (the enhanced number of pulls on French cities vs. the networking effect of extending those lines), but as a rough approximation of relative worth pop1xpop2/distance^2 is better than hand-waving.

    For the record, using metro areas on similar lines (Melbourne-Sydney 870km vs Paris-Marseille 780km):
    Paris-Lyon: 79.3, Paris-Marseille: 31.5, Marseille-Lyon: 23.3
    Melbourne-Sydney: 25.5, Sydney-Canberra: 20.2, Melbourne-Canberra: 3.5

    Thus a Melbourne-Canberra-Sydney line has roughly the same value under the model as the Lyon-Marseille extension. Canberra being the national capital it is probably worth adding a multiplier for it versus a city of the same size. The model would also work better with time than distance, but that makes it more dependent on the number of stops and route.

    Which brings up an interesting point related to the article. One of the questions should be: is it worth stopping a high speed line in regional centres at all? The total added value of Albury-Wodonga (by far the largest centre on that route) is only 6.2. The value lost from additional time/distance trying to service places off the line like Shepparton and Wagga offsets the expected value gained from regional passengers.

    If the HSR was used to anchor a tenant to a regional centre – the oft discussed here idea for an international airport would probably increase population by ~100k in the area – then there might be an expectation of regional growth. Otherwise the HSR would be much better assessed purely as a connection between the major Australian cities.

  7. Alan Davies

    Saugoof #10:

    From memory the Sydney-Melbourne route is the second busiest airline route in the world

    It’s the fifth busiest airline route in the world according to Wiki, so that’s very busy.

    But a technical point. The Wiki ranking is on an airport-to-airport basis. A lot of cities have multiple airports, so I suspect this whole “busiest route” in the world stuff is over-cooked. Sydney (metro)-Melbourne (metro) is busy, but probably not “world top five” busy.

  8. Saugoof

    The difference being that if you’d build a high speed train line from Brisbane to Melbourne (no one is proposing Noosa to Warrnambool) then you’d have to build one single train line, totaling about 1800km.

    France currently only has a bit over 2000km of high-speed tracks in place, but a vast network of “low” speed lines that these same trains can use. And by low speed we’re still taking speeds far in excess of what Australian trains do. Once the French network is completed to current plans the country will end up with about 4000km of high speed lines. That still tips the population vs. line length needed in favour of France. But what in my opinion tips it back in Australia’s favour is that Australians travel far more between Sydney, Melbourne and Brisbane than the French travel anywhere in the country. From memory the Sydney-Melbourne route is the second busiest airline route in the world.

  9. Steve777

    If you draw a line on the map of Australia from about Noosa to Warrnambool, then the area to the East and South of that line will be roughly comparable to that of France but the population would be about 14 million, about 22% of that of France. That area would also be comparable to the size of Japan, but with about 10% of Japan’s population. I would have to be very sceptical of the viability of a VFT for SE Australia.

  10. duke the lost engine

    Saugoof #7,

    The relatively high concentration of the Sydney/Newcastle/Wollongong cluster would support the case for investing in high quality infrastructure within that cluster.

    The same applies for Melbourne/Geelong/Ballarat and Brisbane/Gold Coast/Sunshine Coast.

    However, these clusters are very distant from each other, and the population density between them (eg between Newcastle and Gold Coast) is really very low compared to Europe.

  11. Saugoof

    In a way you could almost make the claim of France being a concentrated country too. Paris has a population that is twelve times the size of the second largest city Lyon. I used NSW for my example because the relationship between Sydney and Newcastle/Wollongong would be roughly the same.

    France is definitely more densely populated than the whole of NSW, and has a much higher overall population to support building infrastructure like this. But the fact that NSW is a concentrated state with the vast majority of the population living with about 100km from the coast, you’d think this would actually make it easier building a rail infrastructure that serves, say, 80% of the population. The population density of that NSW coastal strip (outside of Sydney) would be quite similar to France (outside of Paris).

  12. Alan Davies

    Saugoof #5:

    I think the argument is Australia’s a concentrated country, with a large proportion of the population living in a small number of big primate cities. Those cities are distant from each other and already well-connected by air services.

  13. Saugoof

    I spent a bit of time in France over the last couple of years. This is a country that has built a high-speed network connecting regional centres and large cities with very fast and efficient trains. From having used these and seeing how they are used I can only say “bring them on!”

    Although they are somewhat expensive to use, the ease and speed with which you can travel the country is amazing. The argument about Australia being a low-density country is only partially applicable here. France is remarkably thinly populated, at least by European standards, and like, say NSW, consists of a huge city and many much smaller regional centres. For all the massive cost that it would have taken to build this network, I love the fact that I can take public transport, as I did earlier in the year, from Lyon to Paris’ CDG airport that was 500kms away and it took less time than to go from Mornington to Melbourne’s airport.

    I do find it funny that it seems like we only ever look at whether public transport solutions are cost-effective but have no problems building freeways on a whim.

    Shifting goods transportation to rail is another subject that never seems to come up for discussion. Just out of curiosity, is this something that, in the past, wise people have looked into and found that it’s not feasible or is it just that no one is interested or that transport companies want to protect their turf? The amount of trucks that rattle up and down Hume Highway is staggering.

  14. Steve777

    I agree with David R. The current public transport systems are nowhere near up to scratch and money would be much more effectively spent upgrading them. The ‘High Speed Rail’, if it were to be built, would be either: the ‘Concorde’ of the public transport word – hugely expensive and a viable option for a very few; or a black hole swallowing huge subsidies that could be better spent elsewhere.

  15. Krammer56

    You’ve hit the key point Alan.

    It costs the pretty much the same to put a house, schools, parks, water, local roads, bus services, etc in a country town as on the edge of Melbourne. Land is still used up, wherever a house is built. The raw land cost may currently be less in the country due to less competition, but that would quickly change if real growth pressures occurred (lots of studies show improving accessibility drives prices up – that is the main reason why central city prices are higher than the burbs!)

    I was surprised that the Greens are promoting HSR for regional commuting. The energy costs of moving a body 300 km at 300 km/hr from Wodonga to Melbourne via HSR must be many times higher than moving one from, say Craigieburn to Melbourne (30 km). This seems to be a knee-jerk reaction to the simplistic “burbs are bad” mantra that hasn’t been thought through!

  16. David R

    I think it would be much better to focus on improving existing rail links to regional cities. Key routes would be Melbourne to Geelong, Ballarat and Bendigo and Sydney to Newcastle and Woolongong. I know this policy has already been adopted in Victoria. I think it could still be improved on.

    It utilises existing rail corridors so the cost of land acquisition is avoided. Also the distances are shorter, less than 150km, which can be covered in one hour by conventional rail rather then requiring the more expensive high speed rail technology.

  17. mook schanker

    Some thought provoking comments about regional centres and also regional business clusters so to speak Alan. The two are different aspects and I would suspect the major attribute for HSR would be the former. Also Melbourne is almost exclusively radial PT wise and the heart is the CBD which is growing faster than probably any other business cluster.

    I am unsure myself about the value of regional centres in the scheme of HSR say when compared to a fast rail service, though this a feature in the overall HSR scheme that must be accounted for. Would we build HSR just for regions? Of course not but it adds to the +/- in the BC. In saying this though, London has oodles of rail based benchmark quantitative data on regional centres, productivity and the efficiency of utilising existing infrastructure, especially around the ring of 1-1.5hrs from London town. Surrounding rural towns as a whole are ‘valued’ according to a string of qualitative metrics, they have it down to an art.

    A lot of middle class folk in the UK would probably kill to only pay $30 a day as well 🙂


https://www.crikey.com.au/2012/12/19/will-high-speed-rail-save-the-cities-and-develop-the-regions/ == https://www.crikey.com.au/free-trial/==https://www.crikey.com.au/subscribe/

Show popup

Telling you what the others don't. FREE for 21 days.