Where workers from the municipality of Casey commute. Very few workers living in outer suburbs like Cranbourne work in the city centre where public transport is very competitive with cars for the journey to work.

The Age published an article on Monday proposing Five ways to tackle Melbourne’s worsening peak-hour traffic. It starts, as usual, by extrapolating from the “personal” angle:

The latest census data shows that the sleepy suburb of Skye in Melbourne’s southeast has the highest proportion of people driving to work (82.4 per cent), and the reason is perhaps unsurprising: it is a public transport desert. The closest train station is kilometres away, forcing residents to rely on the delay-prone bus network.

As the paper sees it, the problem is the major transport projects the Andrews Government is building, such as the $11 Billion Melbourne Metro and $6.9 Billion level crossing removal program, won’t be completed for five to ten years, and hence won’t do anything to tackle congestion in the interim.

So The Age asked some transport experts for ideas on what could be done in the short-term to reduce congestion. It’s come up with five ideas:

  1. Improve buses e.g. higher frequencies, more priority lanes, redirecting inefficient routes
  2. Encouraging car-pooling e.g. priority lanes, zero tolls, preferred parking for cars with three occupants
  3. Ten-minute train services e.g. 10-minute frequencies across entire network. Doesn’t require additional trains
  4. Road charges e.g. discounted rego and fuel tax charges for low income earners
  5. Improve bike access to trains e.g. secure parking at stations and tram stops.

It’s a fair question and these suggestions are all worth doing to improve transport in Melbourne and other cities, but whether they’d have a noticeable impact on traffic congestion in the short term is doubtful. Here are some matters to consider.

First, contrary to The Age’s assumption, the various major projects currently under construction or in planning will have negligible impact on traffic congestion when they’re finished. Melbourne Metro, for example, adds 39,000 extra seats. They’re needed to bolster the train system, but they won’t have much impact on the 1.76 million car trips in the morning peak, or the 1.83 million in the evening peak.

If and when they’re eventually built, the road projects the Government’s planning – like the West Gate Tunnel and the North-East Link – will reduce localised congestion in peak periods, but it’ll be a short-lived benefit. It won’t be sustained because latent demand will fill up the new roadspace.

The Age is wrong to frame the issue as a short-term problem that will eventually be solved by the Government’s capital works program. Nothing the Government is building will “solve” traffic congestion; note that it’s still a serious problem in New York and Paris. Projects like Melbourne Metro will provide an alternative to congested streets, but only for an additional few.

Second, only one of the five suggested ideas – road charging – would be effective in managing traffic congestion. Like new motorways, the other four will be undermined by the phenomenon of latent demand. Any road space liberated by motorists who switch to bus, train, bicycle or car-pooling will in relatively short order be taken by another motorist. The advantage of congestion charging when properly designed is it deters both existing and prospective motorists.

Third, the time frame specified by The Age is way too short for congestion charging to have a significant impact. The Government’s capital program is already largely determined for the next four years. More importantly, a controversial initiative like this would require a long lead time for politicking, planning, procurement, construction and, most of all, engaging with the public. It’s devilishly hard to develop a congestion charging system that both works efficiently and is acceptable to citizens.

Fourth, The Age has chosen to dramatise its report by highlighting the suburb with the highest car use in Melbourne for the journey to work i.e. Skye, where 82% of commuters travel by car. Yet in nearby Cranbourne, which has a rail line through the middle, the paper’s figures show the corresponding mode share for cars is 79%. It’s 79% in Cranbourne East too, which also straddles the rail line.

That’s a tiny difference, about three percentage points, suggesting the absence of a rail line explains only a small part of mode choice in this part of Melbourne. Even that small difference is more plausibly explained by a selection effect i.e. those settling in this region who want to use public transport select a location close to the rail line e.g. Cranbourne rather than Skye. The key reason most workers in this part of the world drive is very few work in the city centre where travelling by train easily trumps driving.


There are no quick and easy fixes for congestion. It’s to their discredit that politicians and some advocates push new roads and rail lines as the solution. They’re not, although a new rail line provides an alternative to congested streets for some. Other than eliminating cars entirely, the only plausible answer to traffic congestion is to make driving less attractive in some way, so that at least circa 5 – 10% of motorists are deterred from driving in peak periods. Congestion charging is the most obvious way of rationing demand, although there are others e.g. some Chinese cities use odd/even number plates (without success). But there’s no way congestion charging is a short-term solution.