Is public transport winning the battle for commuters?
New Census 2011 data released yesterday shows public transport is on the rise for commuting and driving is waning. But there’s no guarantee the change will be sustained.
The big news yesterday was the release by the Australian Bureau of Statistics (ABS) of the much anticipated journey to work data from the 2011 Census.
The headline story is public transport and cycling increased their share of work trips in all mainland capital cities over the last five years, while cars lost share. But cars still dominate travel.
The Sydney Morning Herald noted a sizeable increase in the share of commutes in Sydney taken by public transport. A key reason, it reports, was the opening of the Epping to Chatswood rail line.
The Age also noted a big jump in public transport use in Melbourne, mostly on trains. I don’t know how the figure was derived, but the paper reckons cars now have an unbelievably low 65% share of commutes (it initially reported a whopping 12.1 percentage point drop since 2006 in car’s share!).
You can easily look at the data yourself, starting with the ABS’s Community Profiles page – just input a location and download the Basic Community Profile (Table B46). Fortunately though, Charting Transport’s Chris Loader has already done the heavy lifting.
According to his calculations, the share of journeys to work by car increased marginally in Adelaide and Canberra over the last five years, but fell in other cities. It dropped 2.1 percentage points in Perth, 1.5 in Sydney and 0.7 in Brisbane.
The fall in Melbourne was 1.7 percentage points, from 76.5% to 74.8%. That’s a notable improvement by historical standards, but it’s a far cry from The Age’s figure of 65%!
Conversely, public transport’s mode share rose significantly in Melbourne (by 2.2 percentage points), Sydney (2.1) and Perth (2.1). The improvement was modest in Brisbane (1.1) and static in Adelaide (-0.1) and Canberra (-0.1).
In looking at the results, bear in mind that the Census confines itself to collecting data on the journey to work, which only accounts for around a quarter to a third of all trips. That’s an important point, because public transport wins a much bigger share of work trips than most other journey purposes.
While it’s obviously very early days, here are some initial observations.
It might be tempting to think the loss in car’s mode share represents a direct transfer to public transport, but that’s not at all clear. As appears to be the case in the US, much of it might be going to workers based at home – they comprise a significant and generally growing proportion of the workforce (e.g. 4.2% of “trips” in Melbourne in 2011).
The data indicates public transport patronage grew strongly over the last five years in the leading cities. It can’t be assumed though that the trend will automatically continue.
The success of public transport in capturing work trips is in large part due to the significant concentration of jobs and rail infrastructure in the city centre. In some cities this “market” is approaching saturation and there’s a limit to its potential size (e.g. Melbourne’s extended CBD has circa 15% of metro jobs).
Sustaining the improvement will require massive investments in well-targeted public transport infrastructure and improved operational policies. Supportive land use changes will also be required but it’s unlikely they’ll be enough.
Consider that Sydney’s population-weighted density is 52 persons per hectare and Melbourne’s is 33 (I’ve explained these numbers before). As noted, public transport’s share of work trips at the 2011 Census was 22.9% in Sydney and 15.8% in Melbourne.
That implies the density of Melbourne would have to increase by 57% in order to achieve a 7 percentage point increase in mode share. In Brisbane’s case, density would have to double to increase mode share by 8.2 percentage points.
This is of course a simplified illustration. It doesn’t, for example, account for employment-weighted density.
Nevertheless, it’s a sobering reminder of the limits of land use policy. As I’ve said many times before (e.g. see here), achieving significant mode shift will also require severe constraints on car use.
Next time I plan to take an early look at what Census 2011 tells us about cycling and walking in our major cities.