Australian Bureau of Statistics
8501.0.55.006 - Information Paper: ANZSIC 2006 Implementation in Retail Trade Statistics, July 2009, July 2009
Latest ISSUE Released at 11:30 AM (CANBERRA TIME) 03/08/2009
|Page tools: Print Page Print All RSS Search this Product|
Available from the Downloads tab of this issue are mock-ups of the publication and time series spreadsheets as they will appear from the July 2009 issue. Also available is a concordance between the current and new time series identifiers.
The Australian and New Zealand Standard Industrial Classification (ANZSIC) provides a basis for the standardised collection, analysis and dissemination on an industry basis of economic data including Retail trade statistics. The Retail trade statistics are currently based on the 1993 edition of ANZSIC.
In 2002, the ABS and Statistics New Zealand commenced a review to maintain the currency and relevance of ANZSIC. The review found that new materials, technologies and production techniques had been adopted which affected the way industry and businesses operate. Most importantly, whole new industries had emerged and needed to be included in the classification.
Australian and New Zealand Standard Industrial Classification 2006 (cat. no. 1292.0) was released in February 2006. Along with the new classification, various support tools to assist users in the transition from ANZSIC 1993 to ANZSIC 2006 were released via the ABS website. These tools include correspondences between the old and new classifications, class change tables (which show how each ANZSIC 1993 class was affected and record where the ANZSIC 1993 primary activities are classified in ANZSIC 2006) and indexes of primary activities. The ANZSIC 2006 publication and these support tools can be downloaded from the ABS website free of charge.
For information about plans for implementing ANZSIC 2006 in other ABS statistics see the Information Paper: Update on ANZSIC 2006 Implementation (cat. no. 1295.0.55.001).
SCOPE OF RETAIL TRADE STATISTICS UNDER ANZSIC 2006
The scope of the Retail trade series will align with the Retail trade division of ANZSIC 2006 with the following exceptions.
Retail trade statistics will continue to include selected food services (Cafes, restaurants and takeaway food services) which under ANZSIC 2006 are included in the Accommodation and food services division.
The ANZSIC 2006 Retail trade division includes the following industries which will continue to be excluded from Retail trade statistics:
Household equipment repair services are included in the ANZSIC 1993 Retail trade division but are out-of-scope of Retail trade statistics. Under ANZSIC 2006 these activities are not included in the Retail trade division and will remain out-of-scope of the statistics.
CLASSIFICATION CHANGES IMPACTING ON RETAIL TRADE STATISTICS
Changes to Retail trade statistics as a result of the introduction of ANZSIC 2006 can be due to:
ANZSIC 2006 Class Change Tables (cat. no. 1292.0.55.003), available from the ABS website, is useful for understanding the detailed changes that have occurred with the release of ANZSIC 2006. The tables provide information about ANZSIC 1993 activities that have moved to another class, classes that have been split and actions taken in classifying these activities in the ANZSIC 2006 classification. Refer to Retail trade and Accommodation and food services commencing on electronic page 40 noting the industries mentioned above that are in-scope of Retail trade statistics. The main changes are detailed below.
The major change for Retail trade statistics as a result of the ANZSIC 2006 definition of Retail trade is the move to the Manufacturing division of Bread and cake retailing, where the goods are baked on the premises. This will become out-of-scope. Other changes include:
Businesses in-scope under both ANZSIC 1993 and ANZSIC 2006 but changing publication subgroup
To align with the structure of ANZSIC 2006, parts of some publication industry subgroups will move to a different industry subgroup under ANZSIC 2006. The main changes are summarised in Table 1.
ANZSIC 2006 includes separate classes for non-store based retailing and retail commission-based buying and/or selling which are included in the publication industry subgroup Other retailing n.e.c..
The program for implementing ANZSIC 2006 was developed to facilitate the release of the Australian National Accounts on the new basis from late 2009. Retail trade data has been collected on an ANZSIC 2006 basis for the April, May and June 2009 reference months in parallel with the current survey industry classification. The first results on an ANZSIC 2006 basis will be monthly current price estimates released in the July 2009 issue of Retail Trade, Australia (cat.no. 8501.0) which will be released on 9 September 2009. Quarterly chain volume measures up to June quarter 2009 will be first released on an ANZSIC 2006 basis in the August 2009 issue of Retail Trade, Australia.
INDUSTRY GROUPINGS TO BE USED IN RETAIL TRADE STATISTICS
The ANZSIC 2006-based industry groupings to be used for Retail trade statistics will be similar to those in the current publication although the order in which the industries are presented has changed to better align with the ordering used in ANZSIC 2006. Retail trade statistics are presented at two levels of industry grouping - six broad industry groups and 15 detailed industry subgroups.
The six broad industry groups to be included in Retail Trade, Australia are:
The detailed industry level of output released for Retail trade is the industry subgroup level. The 15 retail industry subgroups to be available on the ABS website are shown in Table 2. This table shows the industry group to which each industry subgroup contributes and the relevant ANZSIC 2006 class.
CONTENT OF THE MONTHLY PUBLICATION AND SPREADSHEETS
The revised monthly publication will reflect the content of the current publication except that two additional tables showing quarterly data will be included.
Table 3 shows the tables to be included in the publication. Monthly tables are included in each issue and the quarterly tables are only included in the March, June, September and December issues. The additional quarterly tables are tables 5 and 6.
The above publication tables will also be available in time series spreadsheet format from the Downloads tab of the ABS website. In addition, the time series spreadsheets listed in table 4 will be available from the Downloads tab.
Publication tables 1 to 4 and spreadsheet tables 11 to 16 will be included from the July 2009 issue of Retail Trade, Australia. Tables 5 to 10 will be included in the publication from the September 2009 issue but will be available as spreadsheet tables from the August 2009 issue.
A mock-up of the publication and shells for the time series spreadsheets can be accessed via the Downloads tab of this paper on the ABS website.
A concordance between the current and proposed time series identifiers is also available from the Downloads tab.
IMPACT ON TIME SERIES
Introducing a revised industry classification will impact on the Retail trade time series. The Retail trade series will change as businesses in an industry class or part of a class move into-scope or out-of-scope of Retail trade or change the industry subgroup to which they contribute. The classification changes will impact on:
In addition, the benchmarks used to group businesses on the Retail trade survey frame have been updated to reflect more up-to-date information, primarily sales reported to the Australian Taxation Office on Business Activity Statements.
The number of businesses selected in the Retail trade sample will not change significantly. The current sample includes 500 'larger businesses' and 2,700 'smaller' businesses and the ANZSIC 2006 sample will include 500 'larger businesses' and 2,750 'smaller' businesses. However, due to the changes mentioned above, it was not possible to include the same businesses in both the current and ANZSIC 2006 samples. Of the 2,700 'smaller' businesses selected in the current sample, 1,500 were also selected in the ANZSIC 2006 sample. All the most significant 'larger businesses' were selected in the ANZSIC 2006 sample.
The changes being made to the Retail trade series will result in a shift in the level of the series from the ANZSIC 1993 to the ANZSIC 2006 estimates. The ABS has conducted the Retail trade survey on both the current and ANZSIC 2006 bases for three months (April to June 2009).
The difference in the level of the two series will be measured. The difference in the level will be backcast in the historical series to make a time series of estimates on an ANZSIC 2006 basis. The aim is to make the time series as continuous and comparable as possible and to maintain, as far as possible, the integrity of the period to period seasonally adjusted movements. It is likely that this will be achieved at the total Retail and broader industry and state levels but it can be more difficult to achieve at the state by industry level.
The revised time series on an ANZSIC 2006 basis will be released in the July 2009 issue of Retail Trade, Australia along with an analysis of the impact of the level shift.
The seasonal adjustment of monthly Retail trade statistics involves direct seasonal adjustment at the Australian total level, State total level and the state by industry subgroup level. Broad level industry group totals are derived from the detailed subgroup series. In addition to this a methodology is used on the seasonally adjusted time series to force the sum of fine-level (state by industry subgroup) estimates to be equal to the Australian total.
As there is only backcast historical data on an ANZSIC 2006 basis, the seasonal pattern will initially be unknown for the ANZSIC 2006 estimates and the ANZSIC 1993 seasonal pattern will instead be used as a proxy. This uncertainty in the seasonal pattern may lead to increased volatility in the seasonally adjusted series in the short term. Uncertainty during the period affected by the Government's stimulus package and other influences associated with global economic conditions has also led to uncertainty in the seasonal factor estimates and, due to this, forward seasonal factors calculated on data up to November 2008 are being used from the December 2008 reference month onwards. Concurrent seasonal adjustment will be reinstated when estimates of the Retail trade seasonal pattern become more stable. A detailed review of the seasonal adjustment factor estimates is planned prior to the release of the December 2009 issue.
The Retail trade trend series have been suspended as at November 2008 as it is not possible to determine the trend in retail turnover through the period affected by the Government's stimulus packages and other influences associated with global economic conditions. For further details refer to the December 2008 issue of Retail Trade, Australia (cat. no. 8501.0).
If you have any queries about the changes described in this paper please contact Rod Cowie on email@example.com or (02) 6252 6720.
These documents will be presented in a new window.
This page last updated 2 June 2011