WestConnex: Drawbacks and Alternatives

The NSW Roads Minister Duncan Gay asserts that we have no choice but to build WestConnex; a project which government regards as a congestion buster.
Other commentators see it as a necessary evil; a more nuanced view that recognises there will be drawbacks.
The advocacy group Action for Public Transport (NSW) believes the drawbacks are so significant that a deeper examination of alternatives is warranted.

The first disadvantage is cost.
This is expected to be around $450 million/km for the yet-to-be-contracted stages of WestConnex to provide for an ultimate people carrying capacity of about 8,400/hr on each three lane carriageway.
Two rail-based projects currently under way in Sydney demonstrate that public transport can deliver much lower costs per unit of passenger capacity.
In comparison with the remainder of WestConnex, the CBD and South East Light Rail project will provide a similar ultimate passenger capacity on just one surface lane in each direction at less than half the capital cost/km.
Similarly, the North West Sydney Rapid Transport project will provide many times the passenger capacity of WestConnex and at a somewhat lower capital cost/km.
The unit cost comparison with public transport, based on project cost and capacity data that is attached, would be even more unfavourable if an allowance for the capital cost of cars is added to the WestConnex figure.

The second disadvantage is traffic growth.
It is inevitable that there will be more traffic with WestConnex than without.
It was the early Macquarie Street Mint economist William Stanley Jevons who first observed that more efficient production (in this case coal) led to higher consumption, rather than to lower costs as intuitively expected.
150 years later in transport today, this effect is referred to as demand elasticity with respect to generalised cost, of which time is the major component, and can be related to behavioural tendencies to maintain a constant travel time budget.

Assuming that the traffic demand forecast with WestConnex is realised, and even this may be questionable given trends in VKT (vehicle kilometres travelled) growth, the above elasticity means that the future demand without WestConnex must turn out to be less.
Travel mechanisms to achieve this can include time shifting, shorter distances, ride sharing, alternative modes and trip purpose amalgamation.
Consequently the predict-and-provide technique that appears to have been used to evaluate WestConnex, whereby the demand forecast with WestConnex is applied to the network without it, to make a with/without congestion comparison, can seriously overstate the benefits.
This is because the high congestion costs calculated without WestConnex would not be realised in the real world due to the aforesaid elasticity and travel time budget considerations.

Other areas of disadvantage include the social and environmental impacts from an increase in car dependency.
Offsetting all these disadvantages is a need to consider freight, trade services and car travel involving diverse origins and destinations.
Starting with the last, the sheer mass of VKT implied, with at least 60% having only one occupant, suggests that opportunities to realise metropolitan transport economies of scale, through complementary public transport and land use changes, are not being realised.
This, along with the Household Travel Survey statistics, showing that the commute travel distance over all modes is twice that for the average of each other trip purpose, indicate that there are major structural inefficiencies in Sydney that should be tackled instead, rather than being wastefully reinforced by proceeding with the remainder of WestConnex.

Such structural changes that over time enable the volume of diverse origin and destination car use to be reduced would improve road travel conditions for those that remain, including the providers of freight and trade services that make up a minor, but growing, share of the total traffic.
Only then can the needs of these users be properly considered, with the possibility of dedicated infrastructure provided on commercial terms being part of the mix.

Planning NSW has already implicitly signalled the need for structural change with the visionary 2004 announcement of a “city of cities” future for Sydney.
The most recent strategic announcements continue on this theme.
What has been, and still is, lacking is planning for the evolution of public transport from just a radial commute network into also a rapid grid to make suburban nodes more accessible.
By making cross-town travel as convenient as accessing the CBD, nodes of knowledge based employment can agglomerate in a self reinforcing process for patronage on the network.
An extensive and well patronised rapid public transport grid network providing comprehensive accessibility seems to be the mark of most, if not all, great liveable global cities.

The passing of ten years since the “city of cities” announcement without major progress towards a supportive public transport infrastructure can be linked, for example by the increasing Western Sydney employment deficit, to the worsening traffic congestion being experienced in Sydney.
Thus WestConnex can be seen as a politically motivated panic reaction to this neglect, despite the disadvantages mentioned above.
However it may not be too late to instead urgently pursue the construction of a public transport rapid grid network if some short-term measures to ease road congestion are also introduced.
An example would be state-sanctioned ride sharing on selected road commutes until reductions in per-capita metrics such as commute distance and transport pecuniary cost are being achieved.

Public transport funding mechanisms also need to be better structured as a complementary measure.
While public transport and land use benefits, and road use externalities, are already recognised through community-accepted subsidies, rates and user charges, a formal linkage of these cash flows is needed to put urban road and rail funding on a more equal footing. In conclusion, while there may be no single PROJECT alternative to the remainder of WestConnex, there is a STRATEGY alternative that better suits a nation struggling to live within its means, and therefore deserves greater consideration.

Project Cost and Capacity
WestConnex Unit Cost
2012 announcement$11.5 billion for 33 km
Adjustment to 2014 dollars$12.0 billion for 33 km
Contracted M4 Widening stage$0.5 billion for 7.5 km
Yet-to- be-contracted remainder for the original length$11.5 billion for 25.5 km
Initial unit cost for the yet-to-be-contracted stages$450 million/km
WestConnex Later Stage Capacity (approx.)
Initial capacity with two lanes and 1.4* occupancy5,600 people/hour
Ultimate capacity with three lanes and 1.4* occupancy8,400 people/hour
CBD and South East Light Rail Unit Cost
2014 contract announcement$2.1 billion for 12 km
Initial unit cost$175 million/km
CBD and South East Light Rail Capacity (approx.)
Initial capacity with four minute intervals7,000 passengers/hour
Expanded capacity with 3.25 minute intervals8,500 passengers/hour
North West Sydney Rapid Transit Unit Cost
2014 contract announcement$8.3 billion for 23 km
Estimated allowance for Epping-Chatswood conversion$0.2 billion
Project remainder for the length of the new route,
ignoring a possible reduction from deducting
the Epping-Chatswood share of trains
$8.1 billion for 23 km
Initial unit cost$350 million/km
North West Sydney Rapid Transport Capacity (approx.)
Initial capacity with six car trains at four minute intervals17,000 passengers/hour
Ultimate capacity with 8 car trains at 2 minute intervals45,000** passengers/hour

20 January 2015


* Road capacity is 2,000 light vehicles per hour per lane multiplied by the average light vehicle occupancy of 1.4 taken from the Transport Supply and Demand Forecasts for the Sydney Region published by BTS. An occupancy of 1.4 implies that at least 60% of vehicles have only one occupant
** The present maximum Sydney rail capacity (existing double-deck trains) is lower at about 25,000 passengers/hour (due to less people standing), with an increase being planned. web counter