Mode split Sydney 2012/13 measured in two ways: by number of trips and number of kilometres (source data: NSW BTS)

The exhibit compares travel in Sydney in 2012-13 by mode, measured in two ways: (a) by the number of person trips, and (b) by the number of person kilometres of travel. It’s based on data from the NSW Bureau of Transport Statistic’s household travel survey.

The stand out difference between the two measures is trips made solely by walking. This mode accounts for a very large share of all travel in Sydney when measured in terms of the number of trips, winning 17.5% mode share. That’s considerably bigger than public transport’s 11.4% share.

However walking’s share of all travel is decidedly modest when measured in terms of kilometres covered; only 1.7%. That’s because trips on foot are very short on average compared to those made on mechanised modes.

These aren’t the only measures (e.g. travel is also measured by duration) but they’re the most common and arguably the most useful. Neither measure is “better” than the other; they each have their merits.

For example, the number of trips is important for calculating the peak level of travel demand and hence the maximum capacity of infrastructure like roads and rail lines. The capacity of a motorway depends much more on how many cars are seeking to use it at the same time, than how long the trips are.

The number of kilometres is important for showing the environmental impact of travel. It’s the customary basis for calculating the carbon emissions, local pollution and energy consumption of different modes. It also provides a finer measure of exposure than trips for examining phenomena like road casualties.

However there are pitfalls in placing too much reliance on a single measure. For example, walking’s 1.7% share of kilometres of travel might lead to the conclusion that walking isn’t especially important in Sydney.

But as noted, walking accounts for a large proportion of all trips. That they’re very short is mainly the result of a land use pattern that enables some Sydneysiders to make some of their journeys on foot, whether to visit friends, go to the shops, or take a stroll in a pleasant neighbourhood.

Imagine if the land use pattern meant those 17.5% of trips were all taken by car instead! Thinking only in terms of kilometres is biased against the reduction in trip length that comes from higher density and mixed uses (see also Is “per passenger km” the right metric for comparing modes?).

I’ve zeroed in on walking because the difference in share between the two measures for this mode is literally an order of magnitude; walking’s share is 10.5 times larger when measured by trips as compared to kilometres.

The differences are also important in relation to other modes, however they’re vastly less dramatic. If walking is disregarded so that only motorised travel is considered, the differences between modes decreases e.g. public transport’s mode share in Sydney is 14% when calculated by trips but 16% by kilometres; the figures for private vehicles are 84% and 82% respectively. (1)

With a population of almost 5 million, Sydney is Australia’s largest city and by far its densest (see Population density: is Sydney an outlier?). Even though the mode share of private vehicles has declined over the last 10 years it nevertheless remains high whichever way it’s measured.

In their efforts to address the transport demands of Australia’s cities, it’s important that policy-makers consider a range of measures and are aware of their limitations and inherent biases.


  1. I’ve included ‘Other’ in the motorised comparison. This category includes bicycles but also taxis and ferries.