Income and journey to work patterns – investigations for Melbourne 1996, 2001 and 2006

Document Type

Conference Paper

Publication Date


Subject Area

ridership - commuting


There have been a number of recent studies examining journey to work patterns in Melbourne based on ABS Census statistics (VicRoads 2009, Dept of Transport 2008, Mees, Sorupia and Stone 2007). These studies analyse, in particular, mode share and travel volumes by origin and destination and are a useful primary data source that assist transport planning and other similar activities. Journeys to and from work make up about a quarter of the ~11.5m trips made by Melburnians each weekday, and are mostly made in the AM and PM peaks. They are a substantial part of the load placed on a city’s transport system, and a leading contributor to peak-hour congestion and overcrowding (about 45% of non-walking trips in the AM peak are to work1).

This study compares Census journey-to-work data with income data. It provides a new addition to the research in this field by making this comparison at the Census unit record level and then, for reasons of confidentiality, presenting the results by labour force region (LFR). Results were produced for 1996, 2001 and 2006. The analysis is part of a wider joint project between the Victorian Department of Transport (DOT), the Australian Bureau of Statistics (ABS), and VicRoads called the ABS Experimental Journey to Work Income and Expenditure Tables. By providing a direct link between income and journey to work data at the unit record level, the results of this study will assist provide more depth to our understanding of the Melbourne travel market.

This paper examines the relationships between origin, destination, mode of journey to work, and a worker’s income. A more detailed analysis of the Inner LFR is provided in §4: this LFR stands out from the others because of given its relatively high income levels and public transport mode share.

When considering changes over time, we concentrate on the differences between the 1996 and 2006 Censuses, mentioning the 2001 statistics only if they suggest the 1996–2006 trend is much stronger in one than the other.

Abstract extracteded from introduction.