FORTHCOMING CHANGES
RECENT DEVELOPMENTS
As recently announced on the ABS website (refer to 'Changes to ABS survey program for 2008-09'), the ABS is facing a tight budget situation in 2008-09, which has led to a range of reductions in the ABS work program. One change will be an approximate two-thirds reduction in the sample size for the Retail Business Survey from the July 2008 reference month (when compared with the sample used to compile the monthly estimates of turnover contained in this issue). The ABS is implementing this sample reduction in such a way that the sample can be increased again in the future should the ABS funding position change.
OBJECTIVE OF SAMPLE REDESIGN
The principle objective of the current sample design is to show month to month movement of retail turnover and provide corresponding quarterly movements used in the compilation of household final consumption expenditure in the quarterly national accounts. This objective will change with the survey redesign. The emphasis will move to producing a quarterly series with similar detail to that released in the current monthly publication whilst maintaining a broad level monthly series. The Hotels and licensed clubs and Selected service industries will be excluded from the redesigned survey to maximise sample available for higher priority industries. Work is being undertaken to model the impact of the redesigned sample on survey estimates. The results of this work will be used to determine the breakdown of data by industry and state to be released monthly and quarterly. Further information on the redesign will be included in the May 2008 issue of this publication.
SAMPLE METHODOLOGY
It is expected that the redesigned survey will include 2900 businesses each quarter composed of approximately 500 large businesses surveyed every month (currently about 730 businesses) and about 2400 sampled businesses (currently about 3200 businesses) over the quarter. In order to optimise quality given the available resources, a 'one in two out' strategy will be adopted for collecting data from sampled units. Businesses in the sample sector will be allocated evenly across the three months of a quarter with approximately 800 sample sector businesses included each month. These businesses will be required to provide a monthly estimate of turnover for the month of the quarter to which they have been allocated. They will then not be required to report data for the next two months; i.e. a business allocated to the first month of a quarter will be required to report a monthly estimate for the July, October, January and April reference months.
QUALITY OF ESTIMATES
The new sample, while smaller, will still be representative, with businesses selected from all states and territories and all in-scope industries. However, there will be increased volatility in the estimates, particularly the original and seasonally adjusted estimates, resulting from both the smaller sample size and the rotation effect of having a different third of the sample in each month. This is in addition to the rotation effect (also observed in the current design) that occurs in the first month of each quarter when some businesses are replaced in the sample.
Given the increased volatility in the original and seasonally adjusted estimates, the ABS will encourage users to focus on the trend estimates as the increased volatility will be dampened through the 'trending' process, although increased revisions will be observed at the end of the series, compared to current trend estimates.
The most common way to quantify the sampling volatility is to examine the relative standard errors (RSEs) on the estimates. The tables below show the RSEs resulting from the current survey design and the RSEs based on modelling the expected impact on estimates from the survey redesign.
Retail relative standard errors, State |
|  |
| Current design RSEs | Modelled RSEs from new design | |
| Monthly | Monthly | Quarterly | |
| State | % | % | % | |
| |
| New South Wales | 1.5 | 2.5 | 1.9 | |
| Victoria | 1.5 | 2.7 | 1.9 | |
| Queensland | 1.6 | 2.9 | 2.0 | |
| South Australia | 1.7 | 2.7 | 2.0 | |
| Western Australia | 1.7 | 3.0 | 2.1 | |
| Tasmania | 2.6 | 3.6 | 2.4 | |
| Northern Territory | 2.0 | 3.7 | 2.7 | |
| Australian Capital Territory | 1.7 | 3.3 | 2.4 | |
| Australia | 0.7 | 1.3 | 0.9 | |
| |
Retail relative standard errors, Publication industry |
| |
| Current design RSEs | Modelled RSEs from new design | |
| Monthly | Monthly | Quarterly | |
| Publication industry | % | % | % | |
| |
| Food retailing | 0.9 | 1.6 | 1.1 | |
| Department stores | 0.0 | 0.0 | 0.0 | |
| Clothing and soft good retailing | 2.6 | 4.3 | 2.9 | |
| Household good retailing | 1.8 | 2.8 | 2.0 | |
| Recreational good retailing | 4.6 | 7.5 | 5.2 | |
| Other retailing | 3.6 | 5.4 | 3.9 | |
| Hospitality and services | 2.4 | . . | . . | |
| Australia | 0.7 | 1.3 | 0.9 | |
| |
| . . not applicable |
COMPARABILITY OF ESTIMATES
The introduction of the new design may result in a shift in the level of the series. If the level shift is sufficient to distort the seasonally adjusted time series, the historical series will be revised to make the time series of estimates as continuous as possible.
Measuring month to month movements will not be the main priority of the new design although broad details will still be released. With the 'one in two out' strategy, the month to month movement is expected to exhibit increased volatility from the different group of units included in each month's sample. As a result, it is likely that the monthly ABS commentary analysis will compare data with three monthly intervals to highlight the underlying behaviour of each sample cohort.
INFORMATION SESSIONS
Information sessions will be held on Wednesday 18 June at the ABS offices in Sydney (morning) and Melbourne (afternoon) to describe the new survey design and the implication for retail time series. To obtain further information and register for the information sessions please contact Rod Cowie on
<rod.cowie@abs.gov.au> or (02) 6252 6720.
Print Page
Print All