One of the perennial issues in urban transport policy is the duration of the journey to work. Time spent commuting by car is frequently said to be implicated in a range of social problems, from diabetes to family breakdown (e.g. see here, here, here, here, here and here)

So it’s worth taking a fresh look at just how much time we spend on commuting and other trip purposes. The exhibit shows the average duration of one-way commutes and other journeys in Sydney using data from the NSW Bureau of Transport Statistics (BTS) household survey.

Average journey times (in minutes) are presented separately for residents of inner-, middle- and outer-ring suburbs. The first panel presents the data for car drivers, and the second for train passengers.

Since the data is collected via a household survey, note that according to transport researcher David Levinson, Vierordt’s Law says “people are more likely to over-estimate short times and under-estimate long times”.

Sydney’s a good choice for comparing car travel with public transport and thinking about the implications for the future. It’s Australia’s most populous city; it’s got by far the highest population density; easily the highest mode share for public transport; and it has a stronger system of suburban centres on train lines than the other capitals.

There are some very interesting things going on in the exhibit, but the key ones are:

  • Average journey times are much, much shorter for drivers than for train passengers. The BTS data shows that’s true for all purposes including commuting. Indeed, the average commute by car for outer suburban Sydneysiders is quicker than the average commute for inner-ring train users; and
  • Average journey times don’t vary much for car drivers irrespective of how far they live from the CBD, i.e. what ring they’re in. However, in the case of train travellers, trip times are much longer for outer-ring residents. That doesn’t just apply to commutes; it also applies to social and recreational trips.

Train journeys take significantly longer on average than car journeys for a number of reasons:

  • Trains require walking to and from stations, waiting time and in some cases transfers;
  • Trains travel in their own right-of-way, so they’re fast over long distances; this gives residents the option of increasing their housing options by living further from destinations; and
  • Sydney’s “hub and spoke” train system is focused on the huge concentration of activities in the very small area covered by the CBD. Trains enable Sydneysiders to access the jobs and other attractions of the CBD from all over the metropolitan area.

The land use and infrastructure patterns underlying train trips and car trips are very different. Car journey times are shorter in part because there’s no waiting time, no stopping time and they mostly travel door-to-door. A more important factor though is that the great bulk of activities people want to get to have “decentralised” from the centre to the suburbs, along with population.

Car-oriented cities deal with traffic congestion in part by building more roads but mostly by keeping densities relatively low and accommodating growth at the periphery (i.e. growth centres). Local destinations are no longer within a manageable walk like they once were, but they’re now within a relatively quick drive (e.g. the average shopping trip by car in Sydney’s outer suburbs is 13 minutes).

The BTS data indicates that, other than for a small range of trips (e.g. commutes to the CBD), trains are a long way from being good substitutes for cars in Sydney. It also suggests a number of challenging questions, including:

  • Have cars had a more instrumental role in limiting average trips times across the metropolitan area than trains?
  • Are average trip times likely to get longer as trains increase their mode share at the expense of cars?
  • Would accelerating the growth of suburban job centres and improving the level of train service reduce average trip times by train? Or would it lead to travellers living even further away?
  • In view of the differences in average travel times, are trains generally a plausible substitute for cars for local trips?
  • Would a more connected “grid” of frequent train, tram and bus services have a big impact on the travel time gap to cars?

I’ve focused on trains here to keep the discussion manageable, but note that the average duration of trips by bus is also significantly longer than it is by car. That holds for all purposes in all rings, according to the BTS data.

Although the differences are not as marked, the data shows a similar pattern with average trip distance (kilometres). The average length of trips by car in Sydney is appreciably shorter than it is by train in all rings and for all purposes (I’ll look at distance in detail another time).

The numbers in the exhibit are all averages for large areas, so there’ll be plenty of variation at the local level (e.g. some trips much faster, some much slower). The “big picture” takeaway though is that cars still offer a compelling proposition relative to trains, even in congested Sydney.

It’s deluded to think cars are on their way out. Policymakers need to stop ignoring the obvious and start thinking seriously about how to ration demand in congested locations; how to make cars cleaner, quieter and more fuel-efficient; how to make them safer for other street users; and how to make them more respectful of urban amenity.