Total vehicle kilometres of travel on an average weekday in Sydney, kilometres (000). Ssource: BTS Household Travel Survey 2012/13

“Don’t panic!” wrote Curtin University’s Peter Newman in The Conversation last week, “traffic congestion is not coming for our cities”.

Professor Newman says we can stop worrying about congestion because we’re driving less.

Australian cities peaked in car use per person in 2004, like all developed cities across the globe.

That’s the result he says of the young and wealthy moving into cities, faster public transport relative to cars, the shift to the knowledge economy, and growing use of smart phones, especially by millennials.

Professor Newman is right to say that per capita car use is falling in cities in developed countries, particularly among millennials i.e. population is growing faster than kilometres of travel by car (1).

This is a phenomenon that’s been discussed many times in these pages (e.g. see herehere and here).

The underlying reasons aren’t clear, although the best research suggests it’s probably got little to do with the increasing preference among millennials for higher density living and travel by public transport e.g. see Are millennials driving less?

But he’s wrong about congestion. There’s no evidence it’s going away; indeed, it will almost certainly continue to increase for the foreseeable future under current policy settings.

That’s because although car use has declined in per capita terms, it’s nevertheless increased in absolute terms. In fact it’s increased a lot.

Consider Sydney; it’s by far Australia’s densest city and public transport has a much higher share of travel than anywhere else in the country. It’s the place where cars should be the least useful.

Data collected by the NSW Bureau of Transport Statistics shows average weekday travel by passenger vehicles in Sydney increased from 76.2 million kilometres in 2002/03 to 82.6 in 2012/13 (see exhibit).

That’s an increase of 10.2% over 10 years. Sure, it’s not as big as the 12.9% growth in population over the same period and thus travel per person fell; but it’s still an additional 7.8 million kilometres of vehicle travel on Sydney’s roads each day compared to 10 years previously. (2)

Even if 2004 is taken as the starting point (since Professor Newman says that’s when Australian cities reached peak car) the absolute level of travel in Sydney increased by 7.3 million kilometres up to 2012/13. (3)

Unfortunately, congestion does not dissipate simply because car use is growing slower than population; per capita travel is not the same as absolute travel. Congestion will only abate if vehicle travel reduces in absolute terms.

More motorways won’t reduce congestion, as Professor Newman says. Neither will more public transport infrastructure. In both cases the additional road space that’s created eventually gets consumed in peak periods by latent demand.

Building more infrastructure can still make sense in the right circumstances, though, because it enables more people to get to key destinations, albeit at slower speeds in the peak.

In the case of roads that’s because of congestion; and in the case of public transport it’s largely because of the time required for accessing, waiting and transferring.

But more infrastructure won’t reduce congestion. The only realistic way that’s likely to happen is if use of road space by private vehicles is rationed somehow.

The obvious way to do that is congestion charging i.e. putting a price on use of road space in peak periods. That’s how other scarce resources like power are usually rationed.

That would have many benefits; not least, it would push back the time when additional infrastructure needs to be built.

The slowing of growth in car travel in many developed countries is an important phenomenon that’s rightly attracted lots of attention, but there’s no indication congestion is going away.

It will continue to be one of the key challenges for Australian cities as they get larger. It’s vital that city managers don’t fall for the idea that infrastructure will solve it. It’s also vital they don’t get distracted by wishful thinking.


  1. Although there’s a lot of variation among countries, e.g. driving has fallen sharply among millennials in the US, but only marginally in Japan and Norway.
  2. I’m discussing private vehicle travel here because the topic is congestion; but note that passenger travel by car increased by 10.8 million kilometres over the period, almost three times the increase by public transport.
  3. As the exhibit shows, kilometres of vehicle travel were relatively constant over the period 2002/03 to 2009/10; they didn’t take off until 2010/11.