OVERVIEW OF ABS ANALYSIS


In response to the requirement for international shipping container statistics, the ABS investigated the feasibility of producing and releasing container statistics by:

• applying ABS confidentiality requirements to a container statistics dataset provided by Customs and Border Protection, and
• producing experimental shipping containers data.

Under the Census and Statistics Act, 1905 the ABS must not publish information in a manner that is likely to enable the identification of a person or organisation, unless it meets the strict rules set out in the Statistics Determination 1983 (for further information see Proposed outputs).To understand the implications of this requirement for container statistics, standard ABS confidentiality practices were applied to the file.

Utilising all the variables available for shipping container movement statistics would provide a large amount of cross-classified information that could play an important part in research and analysis of Australia's current and future infrastructure. However, this need for information has to be balanced against the ABS's legislative requirements and the awareness that, at a detailed level, the information could potentially identify commercially sensitive information about an individual business.

The Customs and Border Protection dataset used by the ABS, covering the financial year 2009-10, represents a census of all inward international shipping container movements, i.e. containers used to import goods. The dataset includes Self Assessed Clearances (SACs). SACs are designed to clear goods that have a customs value of AUD$1,000 or less, whilst ensuring the correct reporting of revenue assessable goods is met. As delivery postcodes are not available for these clearances they are flagged in the dataset as unknown. The number of these unknown delivery postcodes on postcode analysis was found to be negligible.

Empty containers are not included in the data, and informal clearance processes (which do not require postcode information to be provided in the Import Declaration) are also excluded.

Analysis tables were produced using statistical software packages and present the number of cells that contain information that could potentially enable the identification of an organisation. The number of blank cells was also produced.

The production of experimental results included the use of mapping software to produce transport corridor and depot maps for Australia's five major ports (Sydney, Melbourne, Brisbane, Adelaide and Fremantle). These maps show delivery postcodes for the top 10 TEU totals (as derived from the Customs and Border Protection file), as well as selected road and rail infrastructure and suburbs.

The following items from the Customs and Border Protection file were included in the analysis tables:

• Tariff Classification Number
• Container Count
• Twenty Foot Equivalent Units
• Delivery Postcode
• Port Aligned
• Year, Quarter and Month.

These fields are defined in Data requirements and policy drivers.

For TCN, data were produced at 2, 4, 6 and 8- digit levels.

For a geographic dimension, analysis focussed on the cross classification of the Delivery Postcode and Port Aligned variables for TEU sum. This included ranking the top 10 postcodes for TEU totals for selected ports, plus aggregating postcode data to State and Territory level.