Related to the freeways-as-bypasses fallacy is the claim that most work trips are now ‘cross-suburban’ rather than ‘radial’ towards the city centre. As a result, it is claimed, people cannot use the mainly radial public transport system and must be provided with freeways.
The fact is that successful cities have to adapt to the modern economy. This means greater flexibility in transport to correspond with a greater diversity of home, shopping and working locations. The radial links that characterised cities in the past two centuries are being replaced by greater cross-suburb travel, and public transport is ill-equipped to deal with this.
—Alan Moran, The Age, 24 May 2006
Melbourne is a spread-out, low-density city in which most traffic is driving from one suburb to another. We need freeways to improve inter-suburban access, just as we need a Metro to improve access to the city and inner suburbs.
—Tim Colebatch, The Age, 8 May 2013
The truth is quite the contrary. Official journey-to-work figures from the 2011 Census actually show that:
- 45% of workers work close to home in their local suburban region;
- 31% travel radially (20% to the CBD and adjoining areas, and 11% part-way towards or away from the city, such as Ringwood to Camberwell, or Melton to Footscray); and
- 24% travel across the suburbs, mostly for relatively short distances.
Fewer than one worker in four wishes to travel across the suburbs. When trip length is taken into account, the share drops to one in eight!
At the core of this myth is the mistaken idea that people’s workplaces are now more-or-less equally spread around the urban area, rather than being concentrated in the inner city. It’s a popular idea spread not only by road lobbyists but even by some professional demographers (who really should know better):
Dr Bob Birrell said the Government’s aim to lure many more people to public transport through transit city developments was fanciful. Dr Birrell said the reality was that jobs were widely dispersed throughout Melbourne and most workers preferred the convenience of their cars.
—“Workers still shun public transport over cars”, Herald Sun, 10 October 2007
The influence of this notion is powerful, and (notoriously) counts among its more prominent fans Federal Liberal leader Tony Abbott:
The problem is the economics of public transport in a suburban metropolis. Mostly, there just aren’t enough people wanting to go from a particular place to a particular destination at a particular time to justify any vehicle larger than a car, and cars need roads.
—Tony Abbott, Battlelines, 2009
Meanwhile in the real world, employment statistics show that Melbourne retains a strong central focus for employment and other urban activities. 30 per cent of all jobs in Melbourne are found within the ‘core’ municipalities of Melbourne, Yarra, Port Phllip and Stonnington. If one were to draw a circle enclosing half of all Melbourne’s jobs, it would extend approximately 13km from the GPO: a region corresponding roughly to Myki Zone 1, or to the extent of the inner-city tram network, yet covering just 15 per cent of the total urban area. Most new jobs created within the last decade have also been within the inner area (a big part of the reason for the stunning growth in rail patronage in this time).
What is true is that the CBD itself accounted for a static or declining percentage of jobs over the period from the 1960s to the 1990s. But while the employment focus shifted, much of this shift was to inner suburbs like Richmond, more than to outer suburbs like Rowville. The myth arose from drawing an artificial boundary around the city centre, and classifying any work location outside it as a ‘dispersed suburban’ location, whether it was genuinely out in the suburbs or actually just beyond the boundary (as most in fact were).
And since 2004, the CBD itself has been catching up with the broader inner region. In just two years from 2004 to 2006, while Melbourne’s population grew by 2.4 per cent, the CBD workforce grew by 10 per cent.
Another big piece of contrary evidence is the finding that many ‘circumferential’ arterial roads supposedly failing to cope with cross-suburban traffic are actually carrying less traffic now than they were a decade ago. As urbanist and blogger Alan Davies has observed, Vicroads figures show that on Alexandra Parade and Elliott Avenue – the roads that according to the road lobby are more clogged than ever and need an East West Link to divert the traffic – the traffic volumes have actually been steady or declining for years!
|Alexandra Pde near Smith St, eastbound||33,000||32,000||32,000|
|Alexandra Pde near Smith St, westbound||32,000||30,000||30,000|
|Alexandra Pde near Nicholson St, eastbound||40,000||37,000||34,000|
|Alexandra Pde near Nicholson St, westbound||38,000||35,000||34,000|
|Princes St near Rathdowne St, eastbound||29,000||28,000||28,000|
|Princes St near Rathdowne St, westbound||33,000||30,000||28,000|
|College Cres near Swanston St, eastbound||30,000||28,000||28,000|
|College Cres near Swanston St, westbound||30,000||28,000||28,000|
|Elliott Ave, eastbound||19,000||18,000||18,000|
|Elliott Ave, westbound||17,000||17,000||17,000|
Source: Vicroads, Arterial Road Traffic Volumes (spreadsheet download), February 2013.
Even if one looks further afield, there is scant evidence for huge growth in east-west traffic through Melbourne’s inner and mid-northern suburbs.
|Bell St, Coburg (Nicholson St), eastbound||22,000||22,000||22,000|
|Bell St, Coburg (Nicholson St), westbound||25,000||24,000||24,000|
|Bell St, Preston (Albert St), eastbound||20,000||21,000||21,000|
|Bell St, Preston (Albert St), westbound||24,000||23,000||24,000|
|Brunswick Rd, Brunswick (Lygon St), eastbound||11,000||11,000||11,000|
|Brunswick Rd, Brunswick (Lygon St), westbound||9,000||8,500||8,400|
|Victoria Pde, Fitzroy (Brunswick St), eastbound||24,000||21,000||22,000|
|Victoria Pde, Fitzroy (Brunswick St), westbound||24,000||21,000||22,000|
Source: Vicroads, Arterial Road Traffic Volumes (spreadsheet download), February 2013.
The upshot is that although there’s certainly a significant minority of travel that’s over long distances to dispersed locations, the reality is contrary to what Moran, Birrell and Abbott suggest. Cities draw their strength from strong centres that allow people to do business in proximity to one another. And for travel in cities, there is a great deal of commonality in routes and destinations. As many have noticed, this makes travel patterns very predictable from one day or one year to the next.
For the large part on any given weekday, most of the populace leave the same place (home) at the same time and travel the same route via the same means to the same place (work). In short….traffic is the same every day.
—Peter Downey, National Times, 17 August 2010
For non-work trips, the tendency to remain within one’s local area is more pronounced than for work trips. Figures from the major Victorian travel surveys (VATS in the 1990s and VISTA in the last decade) indicate that trips within local areas account for more than 50% of all travel.
So even when all travel is considered, most of it is local and radial, not across great numbers of suburbs. Even most travel on freeways is local: a Vicroads study (reported in the Herald Sun on 11 July 2005) found that 70% of all journeys on the Monash Freeway were within one council area or to an adjoining council area. Removing all those local trips (assuming one could do so) would remove two-thirds of the traffic, and with it all traces of congestion.
But what about all that traffic on Springvale Road?
An object lesson in the ‘cross-suburban’ travel myth is the almost complete failure of EastLink to fix congestion in Springvale Road and Stud Road, the main north-south arterial routes in Melbourne’s outer east and the two nearest parallel routes to EastLink.
In the 1990s when planning for EastLink began (and it was known as the Scoresby Freeway), pointing to this congestion was a favourite lobbying tactic. It was argued that this congestion was caused by long-distance north-south travel in the eastern suburbs, and that building the Scoresby Freeway would take the bulk of the traffic off these roads.
With EastLink now built, the argument has served its purpose. But the building of EastLink has also shown it up to be completely false.
SPRINGVALE Rd is becoming a nightmare again – and toll road bosses at EastLink are crowing. The morning peak along Springvale Rd is now busier than before EastLink was built, despite promises the toll road would solve the traffic problems.
—Springvale Rd traffic jams as bad as ever, Herald Sun, 19 August 2010
Let’s look more closely at why this seemingly logical idea was, after all, wrong. There is no doubt that traffic has been heavy on these roads for a very long time, both before and after EastLink, but the reason is not long-distance north-south travel. We know this from the official figures. For example, of the 15,000 workers who live in Dandenong, fewer than 200 travel to the City of Maroondah to work. Similarly, of the 20,000 workers in Ringwood, fewer than 600 work in Greater Dandenong.
Traffic on Springvale and Stud Roads is comprised mainly of the same flows that are heavy elsewhere in the region: radial and local traffic. On Stud Road, a key local traffic generator is Knox City shopping centre at the Burwood Highway intersection. Any local traveller approaching Knox City from the south needed to use Stud Road, and still needs to even after construction of EastLink. In addition, travellers to Glen Waverley from almost anywhere in Knox must use Stud Road to reach High Street, Ferntree Gully Road or Wellington Road. The local centre for Rowville (Stud Park) is also located on Stud Road, adding to traffic pressure. In a similar way, people travel locally along Springvale Road to reach shopping centres, workplaces and other destinations, and to access major east-west roads.
Several earlier ‘red spot’ studies by the RACV confirmed this fact. In the first study conducted back in 1994, Stud Road in Knox featured in the top 20 locations where traffic congestion and delays to motorists were most severe. But complaints were not about through (North-South) traffic being delayed. People instead complained of the great difficulty in turning out of Stud Road into Ferntree Gully and Wellington Roads, both of which lead towards the inner eastern area and city centre. The difficulties were experienced heading toward the city in the morning peak, and away from it in the evening peak.
The important point is that none of these local traffic problems are of the type that EastLink can assist with. EastLink does not eliminate the need to access Stud Park shopping centre via Stud Road, or for local residents to travel some distance along Springvale or Stud Roads in order to turn west. And it is foolish to expect that people travelling a few blocks down Springvale Road will make a three-kilometre detour in order to travel those few blocks on a motorway. This is why even road engineers themselves, in their more honest moments, admitted that EastLink would take hardly any traffic off Springvale Road.
With EastLink open and operating since 2008, events have now confirmed these predictions. As numerous commentators have observed, traffic on EastLink has been disappointingly low, with only the northern section (which acts as a radial feeder to the city centre) close to achieving its forecast traffic level.
Investors yesterday dumped shares in the tollway owner, ConnectEast, after the company said traffic volumes had fallen to 28% below its projections. The numbers – and the corresponding scenes of sparsely occupied traffic lanes along the 37-kilometre route – present a stark contrast to what has been happening on the rail network.
—Drivers desert EastLink as rail network overflows, The Age, 8 August 2008
While the road might provide smooth travelling for motorists, EastLink is proving harder to navigate for ConnectEast. About 149,000 drivers a day are using the road – well down on the 260,000 forecast before the road opened. The operator suffered a trading halt on its shares last week after announcing plans to raise $450 million to repay debt in a move it insists owes more to the global credit crisis than poor traffic volumes.
—A tale of two city highways, The Age, 30 November 2008
Toll road operator ConnectEast Group says it is working to grow traffic and revenue on its EastLink tollroad, after reporting an annual net loss of $531.58 million…. The result included a writedown on the EastLink toll road concession of $400 million – to about $2.9 billion – after an independent review of traffic. The review, released last week, revealed a major shortfall between observed volumes and revenues and projections made in 2004.
—ConnectEast books annual loss of $531.5m, The Age, 24 August 2009
PRIVATE investor funding of Australian toll roads is all but dead: the game is up. We thought the bankers and builders would cream off fees, but we never knew that the traffic forecasts were baloney…. The EastLink road that runs from Ringwood to Frankston opened in 2008 with a projection of 256,000 vehicles a month (after 15 months), but here we are long after that deadline and the average daily count is 178,000. Ouch! And the stock price is less than half it was on the day it listed at $1. It closed on Friday at 43.5¢.
—Toll roads on a collision course with their investors, The Age, 5 September 2010
In conclusion, local and radial traffic can cause as much congestion as cross-suburban traffic, when there is enough of it. Big ‘bypass’ roads, that do not cater for local or radial traffic, have proven ineffective as a remedy. Evidently, the cure for this kind of congestion is not more roads, but palatable alternatives to car travel.
[Eastlink] is a terrible disaster and I think it should never have been built. We got lumped with that thing because the Liberals went hard on the matter and forced the Bracks government’s hand. It was an unviable project and the economic tragedy of all this is that in building it we have forgotten opportunities to build much more urgent projects such as the Rowville rail.
—Nick Economou (Senior Lecturer in Politics, Monash University), Monash Journal, 1 December 2008
Last modified: 4 March 2015