ABSTRACT: Like other large metropolitan areas, Sydney has experienced rapid decentralisation of homes and workplaces since 1981 with increasing use of the car for the journey-to-work. Previous research (Parolin and Kamara, 2003) has shown that at least half of the employment growth is concentrated in employment clusters or sub-centres that display a spatial patterning and hierarchy that is likely to affect commute distances and the spatial separation of homes and workplaces. The aims of this paper are twofold; first, to identify employment centres (ECs) in 1981 and 2001 across the Sydney metropolitan area and, secondly, to examine shifts in commuting to employment centres by car and public transport. In this process one is able to test the co-location hypothesis which maintains that employment decentralisation over time leads to a jobs-housing balance (co-location) and fairly constant average commuting distances and durations. Only distances are examined in this paper. The analysis proceeds by using journey-to-work data at the travel zone level and the calculation of statistics on shifts in average distances, and VKT for car and public transport between 1981 and 2001. Patterns in shifts in commute distances and VKT by car and public transport for identified centres are examined.
Analysis of journey-to-work data for 1981 and 2001, and identification of ECs in these years, has demonstrated the significance of centred employment and polycentric urban growth in the Sydney metropolitan area. A total of 28 centres were identified in 1981 and 39 in 2001. Moreover, it has been shown that the co-location hypothesis does not hold for the Sydney area; this was true for 1981 and 2001. The shifts in mode split over the 20 year period indicate the increased use of the car for the journey-to-work in identified ECs and the relatively low use of public transport (bus and train) for travel to jobs in these centres.
For planning and policy relevance, these findings suggest that prescriptions of higher levels of self-sufficiency in new growth areas (better co –location of jobs and people) are not likely to generate lower average distances for the journey –to-work. More polycentric urban growth, even that focussed on public transport corridors, is not likely to reduce average commute durations; the evidence suggests that they will increase – on average. All this strongly suggests that factors other than proximity to workplace have influenced and will continue to influence where workers reside. It is of paramount importance for metropolitan and transport planners to better understand what these other factors are likely to be if we are to move towards achieving sustainability.