What does it cost to travel in Sydney? Spatial and equity contrasts across the metropolitan region

Abstract

There is a strong belief, often perceptual, that residents in the outer suburbs of Sydney are at a transport disadvantage in terms of the generalised cost of daily travel in absolute terms, and in relation to the percentage of income, personal and household, spent each day on travel. This paper investigates this claim using the Sydney Household Travel Survey, an annual survey of randomly selected individuals, from June 1997 to June 2008, a total of 92,413 respondents. We pool the entire data set, adjusting costs for different years, and undertake a spatial interrogation of the data, initially for 13 sub-regions, and then drill down to postcodes to identify sources of systematic variation in the daily generalised cost of travel for individuals and households. In assessing the evidence, we compare public transport outlays with car outlays, where the latter is defined in terms of marginal outlays (i.e., fuel and parking) and all costs (i.e., marginal outlay plus car ownership costs). Given the cost of using public transport (i.e., fares) we speculate that the provision of improved public transport services (and switching from car to some extent) is likely to result in a lower monetary cost of travel, but only if individuals and/or households dispose of vehicles. This seems to apply even where public transport offers a lower travel time, which is not sufficient to compensate for retention of the car. If they retain their cars, then given the lower marginal cost of car use compared to public transport, the contribution of improve public transport translated into a switch of usage from car to public transport will have little impact on accessibility and equity. Hence the entire argument hinges on what response will be made to car ownership in the presence of a non-marginal injection of investment in public transport. The paper also cautions about making statements on mobility equity at a highly spatially aggregate level (i.e., a sub-region), in contrast to establishing the causal links at a more spatially disaggregate level (i.e., the postcode).