Weekday journeys to work by mode, Melbourne 2012-13 (source: VISTA)
Weekday journeys to work by duration and mode, Melbourne 2012-13 (data source: VISTA 2012/13)

There’s a common view that Malcolm Turnbull’s ‘30-minute city’ is yet another slight variation on the many terms we already have for a more sustainable urban form e.g. compact city, walkable city, smart growth, new urbanism, urban consolidation.

They all imply smaller dwellings, higher densities and less reliance on cars. But whereas the others leave a lot of wiggle room (e.g. a more compact city), the 30-minute city specifies a precise standard. It insists that travellers should be able to go wherever they want to – irrespective of purpose, place or time – within a maximum travel time of 30 minutes.

Mr Turnbull accepts cars in his conception of the 30-minute city but the version proposed by Shadow Cities Minister, Anthony Albanese, is much more demanding; he says travellers should be able to get to destinations within 30 minutes by public transport, bicycle or on foot. But even that looks conservative compared to what’s in the Victorian Government’s strategic plan for Melbourne, Plan Melbourne; it also rules out cars but sets a target of just 20 minutes!

The 30-minute city implicitly assumes cities like Melbourne or Sydney, even though they have populations well over four million, consist of a large number of self-contained villages with residents necessarily living at much higher densities than is customary in suburbia i.e. smaller dwellings closer together. (1)

How close are our big cities to the Prime Minister’s 30-minute ideal? How realistic is it? Can the politicians actually deliver on this one?

To answer that question, consider that the Victorian Integrated Survey of Travel and Activity (VISTA) says the duration of the trip we most worry about – the journey to work – averages 37 minutes one-way in metropolitan Melbourne.

Mr Turnbull should note that’s an average; breaking it down shows that 63% of all commutes take 30 minutes or longer. And Mr Albanese should consider the fact that 75% of those commutes are made by car; only 19% of work trips are made by public transport, with 5% by active modes i.e. cycling/walking.

But surely journeys to work that originate in the denser inner suburbs are shorter than 30 minutes? No, the average one-way commute originating in Melbourne’s inner suburbs takes 38 minutes. Commutes by car average 33 minutes; those by public transport take much longer i.e. 47 minutes on average (see here for map of what constitutes inner suburbs).

Alright, but no doubt the residents of somewhere like Geelong with a population of just 180,000 all get to work within 30 minutes? No again; the average commute by Geelong residents takes 35 minutes.

If you want to find a real 30-minute city you get closest in a provincial centre like Shepparton, Ballarat, Bendigo or Traralgon. Taken together, the average commute in these places is 20 minutes. Note though that 84% are made by car and the average commute by public transport takes 78 minutes.

So, these numbers illustrate a key reason why the ‘30-minute city’ is more about political fluff than about offering a real policy alternative. It conveniently ignores history or what economists call path dependency i.e. the huge stock of low density dwellings, roads, and reserves that exist in our cities. It ignores the deep cultural expectations of space embedded in the Australian psyche e.g. the ‘save our suburbs’ movement.

The transition from 150 years of spreading urban form to the much higher densities implied by the 30-minute city would be an extraordinary challenge for Australia’s cities, both financially and politically.

We can take it as given that none of the politicians have calculated either the financial or economic costs of this idea. They don’t know if the benefits exceed the costs and they haven’t thought about whether there might be better ways the massive government funding required might be spent.

But the problem isn’t just one of unrealistic and unattainable promises about the future. Another crucial point is many firms don’t want to locate in a village where they’re theoretically restricted to drawing workers from the local population. And what about customers and suppliers?

Many firms want to congregate in a limited number of very large activity centres, especially the CBD and a handful of major suburban activity centres. That’s because, although big centres come with downsides like higher rents, many firms and organisations get an increase in productivity from proximity to other firms i.e. agglomeration economies. Moreover, activity centres aren’t identical; they tend to specialise in one or two industries (see Are all suburban centres the same?).

Another problem is most workers don’t choose the fastest possible commute. They’re also interested in maximising housing quality subject to affordability. They tend to trade-off commute time for housing attributes like a bigger dwelling. In many cases, the optimum location for a household depends on the disparate job and education destinations of multiple household members e.g. mum, dad and children.

Marchetti’s constant suggests that commuters have a travel time budget; given an increase in travel speed, many will elect to commute further for a better dwelling. But note the figure of 30 minutes commonly associated with Marchetti’s constant is an average across all commuters; it does not indicate a maximum acceptable commute time for individuals as Messrs Turnbull and Albanese appear to assume (see Turnbull’s Smart Cities Plan: is that all there is?).

The 30-minute city pretty well already exists for traditionally highly local trips like food shopping and getting to primary school – 90% take less than 30 minutes. But the news isn’t good in terms of Mr Albanese’s ambition that primary students cycle, walk or take public transport; 70% of primary school trips are made by car (see Surely the 30-minute city makes sense for primary school trips?).

The objective of lowering travel times is a good one but we need to understand what the financial and social cost of each increment in improvement involves and whether or not it’s worth it. At present, Australian cities are a long way from the 30-minute ideal – even by car – when it comes to key travel purposes like getting to work, high school and tertiary education, as well as visits to family and friends.

Australians should be very suspicious of the bona fides of politicians who set very long-term quantitative targets. They’re attractive to voters because they sound tangible and therefore seem authentic. They’re irresistable to politicians because they’re so long term they don’t have to worry about the financial or economic costs, or whether the supposed benefits are even real i.e. they won’t be called to account. (2)


  1. Mr Turnbull doesn’t say so, but I’ll assume he sees singular facilities like airports  – and amenities like beaches – as exceptions to the 30 minute rule. Start to get too many exceptions to the rule, though, and the idea becomes pointless for policy, although not necessarily for politicking.
  2. For example, see Victoria’s Brack’s Government’s target set in 2002 for public transport to achieve 20% share of motorised trips in metropolitan Melbourne by 2020; it now accounts for about 11% of all motorised trips (9% if non-motorised travel is counted) and the share is static.