|Page tools: Print Page Print All RSS Search this Product|
5 Also excluded are the following persons who are not regarded as employees for the purposes of this survey:
6 The sample for AWE, like most Australian Bureau of Statistics (ABS) business surveys, is selected from the ABS Business Register (ABSBR) which is primarily based on registrations to the Australian Taxation Office (ATO) Pay As You Go Withholding (PAYGW) scheme. The ABSBR is updated quarterly to take account of:
7 The estimates include an allowance for the time it takes newly registered businesses to be added to the survey population.
8 Businesses which have ceased employing are identified when the ATO cancels their PAYGW registration. In addition, businesses which have not remitted under the PAYGW scheme for the previous five quarters are removed from the population.
9 A sample of approximately 5,400 employer units is selected from the ABS Business Register to ensure adequate state, industry and sector representation. The sample is updated each survey period to reflect the changes described in paragraph 6. These changes arise from the emergence of new businesses, takeovers and mergers, changes to industry classification, changes in the number of employees, and businesses which have ceased operations. Such updating of the business register can contribute to movements in the AWE estimates.
10 Sample redesigns are undertaken periodically for all ABS business surveys to ensure the survey design continues to be optimal. The most recent sample redesign for the Survey of Average Weekly Earnings was implemented for the November 2017 release.
11 Prior to this, a sample redesign of the AWE survey was also implemented in August 2009 incorporating the Australian and New Zealand Standard Industrial Classification (ANZSIC), 2006 (cat. no. 1292.0).
12 The statistical unit for the survey comprises all the activities of an employer in a particular state or territory based on the Type of Activity Unit. For further information on the statistical unit see paragraphs 19 to 35. Each statistical unit is classified to an industry which reflects the predominant activity of the business. The statistical units are stratified by state, sector, industry and employment size, and within each stratum, statistical units are selected with equal probability.
13 The statistics in this release are classified to industry in accordance with the Australian and New Zealand Standard Industrial Classification (ANZSIC), 2006 (cat. no. 1292.0). This replaced the 1993 edition of ANZSIC in the August 2009 issue of this publication, which had been in use since 1994.
14 The 2006 edition of ANZSIC was developed to provide a more contemporary industrial classification system, taking into account issues such as changes in the structure and composition of the economy, changing user demands and compatibility with major international classification standards.
15 Up until May 2012, Average Weekly Earnings was conducted on a quarterly basis. However, the frequency of the AWE survey is now biannual, with the May 2012 edition being the last quarterly issue and the November 2012 edition the first produced on a biannual basis. AWE data is now produced twice a year relating to the May and November reference periods only. Data is collected and released on the same basis as before for the May and November reference periods. For full details on the change in frequency, refer to the Information Paper: Changes to Average Weekly Earnings, Australia, April 2012 (cat. no. 6302.0.55.002).
16 As a result of the change in frequency, new seasonally adjusted and trend estimate series are produced (refer to paragraphs 60-70 below).
IMPACT OF STATISTICAL CHANGES IMPLEMENTED IN AUGUST 2009
17 With effect from the August 2009 edition, this publication presents data on the basis of ANZSIC 2006. At this time, the ABS also implemented a sample redesign. The changes resulted in a shift in the level of the series from ANZSIC 1993 to ANZSIC 2006 estimates. The difference in the level of the two series was measured and backcast into the historical series to make a time series of estimates on an ANZSIC 2006 basis. Differences at the state, sector and Australia levels are generally insignificant and within released standard errors for each series.
18 Published industry series have been backcast and data from August 1994 to May 2009 are available on the basis of both editions of ANZSIC on the ABS website. More information about these changes can be found in the Information Paper: Changes to Average Weekly Earnings, Australia, Aug 2009 (cat. no. 6302.0.55.002).
ABS ECONOMIC UNITS MODEL
19 The Economic Units Model is used by the ABS to determine the structure of Australian businesses and other organisations. The model consists of:
20 The EG and LE are institutional units and the TAU and Location are producing units.
21 The LE and the TAU are the main institutional and producing units used by the ABS to produce statistical outputs.
22 Diagram 1 illustrates the nature of the relationships between the different units within the model.
Diagram 1: ABS Economic Units Model*
* The legal entity (LE) statistical unit is generally equivalent to a single Australian Business Number registration
23 The Legal Entity (LE) is an institutional unit covering all the operations in Australia of an entity which possesses some or all of the rights and obligations of individual persons or corporations, or which behaves as such in respect of those matters of concern for economic statistics. Examples of legal entities include companies, partnerships, trusts, sole (business) proprietorships, government departments and statutory authorities. Legal entities are institutional units. In most cases the LE is equivalent to a single Australian Business Number (ABN) registration.
24 The Enterprise Group (EG) is an institutional unit that covers all the operations within Australia's economic territory of legal entities under common control. Control is defined in Corporations legislation. Majority ownership is not required for control to be exercised.
25 The Type of Activity Unit (TAU) comprises one or more legal entities, sub-entities or branches of a legal entity that can report productive and employment activities. TAUs are created if accounts sufficient to approximate Industry Value Added (IVA) are available at the Australian and New Zealand Standard Industrial Classification (ANZSIC) subdivision level.
26 A Location is a producing unit comprised of a single, unbroken physical area from which an organisation is engaged in productive activity on a relatively permanent basis, or at which the organisation is undertaking capital expenditure with the intention of commencing productive activity on a relatively permanent basis at some time in the future.
CLASSIFICATION OF UNITS
27 Various classifications are applied to the units in the ABS Economic Units Model. The main classifications applied are:
28 ANZSIC is used to classify the industry in which the TAU has productive activity. Further information on this classification can be found in Australian and New Zealand Standard Industrial Classification (ANZSIC), 2006 (cat. no.1292.0).
29 SISCA provides a framework for dividing the Australian economy into institutional sectors. Further information on this classification can be found in Standard Economic Sector Classifications of Australia (SESCA), 2008 (cat. no.1218.0).
ABS BUSINESS REGISTER
30 The ABSBR is a list of businesses and organisations operating in Australia and is based on the Australian Business Register (ABR). Organisations are included on the ABR when they register for an ABN. The Commonwealth Government requires all government departments and agencies to make use of the ABR to reduce government imposed reporting load, and to use the ABN as the primary reference number for all dealings between government and business. The ABSBR is used to create frames for the various business surveys run by the ABS.
31 The results of these statistics are based, in part, on ABR data supplied by the Registrar to the ABS under A New Tax System (Australian Business Number) Act 1999 which requires that such data is only used for the purpose of carrying out functions of the ABS. No individual information collected under the Census and Statistics Act 1905 is provided back to the Registrar for administrative or regulatory purposes. Any discussions of data limitations or weaknesses is in the context of using the data for statistical purposes, and is not related to the ability of the data to support the ABR's core operational requirements. Legislative requirements to ensure privacy and secrecy of the data have been followed. Only people authorised under the Australian Bureau of Statistics Act 1975 have been allowed to view data about any particular firm in conducting this survey. In accordance with the Census and Statistics Act 1905, results have been confidentialised to ensure that they are not likely to enable identification of a particular person or organisation.
32 It is not practicable for the ABS Economic Units Model to be applied to all ABR registrants and is therefore organised into two parts: the profiled population, and the non-profiled population.
33 Profiled Population: Businesses and other organisations which are considered sufficiently complex and significant, are profiled according to the Economic Units Model. These enterprise groups typically have multiple legal entities, multiple TAUs and are among the largest contributors within industries.
34 Non-Profiled population: Businesses and other organisations with less complex structures. They are regarded as an enterprise group with a single legal entity and a single TAU in accordance with the Economic Units Model. Information for units in the non-profiled population is largely sourced from the ABR.
35 The two populations are mutually exclusive and cover all organisations in Australia which have registered for an ABN.
GENERAL NOTES ON ESTIMATES
36 AWE statistics represent average gross (before tax) earnings of employees and do not relate to average award rates or to the earnings of the 'average person'. AWE estimates are derived by dividing estimates of weekly total earnings by estimates of the number of employees. Changes in the averages may be affected not only by changes in the level of earnings of employees but also by changes in the overall composition of the wage and salary earner segment of the labour force.
37 There are several factors which can contribute to compositional changes, including variations over time in the proportions of full-time, part-time, casual and junior employees; variations in the occupational distribution within and across industries; and variations in the distribution of employment between industries. Such effects may apply differently within different states and territories, and over time.
38 AWE statistics closely follow the International Labour Organisation's concept of 'Statistics of average earnings'. The data is collected in respect of a typical week and, therefore, may not reflect events such as Christmas trading. Further, the data excludes irregular and infrequent payments, such as annual bonuses. For these reasons, caution is advised if using AWE to derive annualised average earnings.
39 Prior to May 2014, surprise outliering was used as the sole methodology to identify and reduce the impact on the estimates of a business whose weighted survey response is an outlier i.e significantly different to businesses in a group with similar characteristics (based on employment size, state and industry). Surprise outliering involves treating the identified outlier as if it were the only extreme unit in the group's population. The outlier is given a weight of one and the weights of the other units in the group are adjusted upwards accordingly.
40 In the May 2014 issue, winsorisation methodology was introduced as the primary method to treat outliers in AWE. Winsorisation moderates the impact of an outlier business without the harsh impact of the surprise outliering approach. This improved methodology provides more stable time series estimates. Surprise outliering continues to be used for a small number of extreme values that may not be sufficiently moderated by the winsorisation method.
41 An analysis of the May 2014 estimates was undertaken to identify the impact on the estimates of the change in methodology. At the Australia level the impact of the change was found to be minimal. However, for some data items in some industries and states there is an impact on the estimates. For further information on outliers, refer to Chapter 17 of Labour Statistics: Concepts, Sources and Methods, 2018 (cat. no. 6102.0.55.001).
42 For further information in understanding Average Weekly Earnings statistics, please refer to the feature article A Guide to Understanding Average Weekly Earnings Statistics, published in the November 2014 AWE release.
AVERAGE WEEKLY CASH EARNINGS
43 The definition of earnings currently used in the AWE survey is, broadly, current and regular payments in cash to employees for work done. Thus, earnings series from the AWE survey historically excluded amounts salary sacrificed, as these had been considered conceptually as payments in kind. However, under the revised conceptual framework for measures of employee remuneration, as presented in Information Paper: Changes to ABS Measures of Employee Remuneration, 2006 (cat. no. 6313.0), amounts salary sacrificed are now considered conceptually to be wages and salaries in cash. Accordingly, the AWE questionnaire was redesigned and, from August 2007, the collection of information on amounts salary sacrificed by employees commenced. However, the AWE series has continued to be published on the old conceptual basis (i.e. exclusive of amounts salary sacrificed) to maintain long term comparability of the time series.
44 Although the AWE survey has conceptually excluded amounts salary sacrificed, in practice, there was evidence that earnings series from the AWE survey had inadvertently included some amounts salary sacrificed. The ABS worked closely with data providers to identify any instances of misreporting, and to amend their reporting practices where necessary.
45 As a result of the separate collection of salary sacrificed amounts from August 2007, and other analyses, the ABS was able to quantify the extent of mis-reporting that had occurred, and to estimate the impact of this mis-reporting on the historical series. Consequently, AWE data series for August 1996 through to May 2008 were revised to exclude all amounts salary sacrificed. For further information see Information Paper: Revisions to the Average Weekly Earnings Series, Aug 2008 (cat. no. 6302.0.55.001) released 11 November 2008.
46 Since the May 2011 edition of this publication, Average Weekly Cash Earnings (AWCE) series have also been released as additional (not replacement) AWE series. The difference between the AWCE and the AWE series is the average weekly amount salary sacrificed. Data relating to the AWCE series are available in the data cubes on the Downloads tab at the top of this page. For more information relating to the AWCE series, refer to the Information Paper: Release of Average Weekly Cash Earnings Series, May 2011 (cat. no. 6302.0.55.003) and for broad level analysis and findings refer to the Information Paper: Changes to Average Weekly Earnings, Australia, April 2012 (cat. no. 6302.0.55.002).
COMPARABILITY OF SERIES
47 The current AWE series, based on information obtained from a sample survey of employers, was introduced in August 1981. Prior to September 1981, the AWE series was based principally on information from payroll tax returns. Revised estimates of AWE for the period August 1981 to November 1983 were included in Average Weekly Earnings, States and Australia, March 1984 (cat. no. 6302.0) published on 12 July 1984 and available on the ABS website. Users who need a measure of the movement in earnings for a period which spans both the payroll tax based and employer survey series should refer to Table 3 in that publication which presents both series linked to a common index base (August 1981 = 100.0).
CHANGES TO TIME SERIES SPREADSHEETS
48 From the May 2015 issue, all time series identifiers used in the Average Weekly Earnings spreadsheets (i.e. table 1 to 14H) changed, as highlighted in the Information Paper: Average Weekly Earnings, Australia: Upcoming Changes to Time Series Spreadsheets (cat. no. 6302.0.55.004), issued on 14 May 2015.
ESTIMATES OF MOVEMENT IN AWE
49 AWE is designed to provide estimates of the level of average earnings at a point in time and, while not designed for movements in earnings, the frequency of collection supports a time series of these level estimates. Data on the average level of earnings are useful for providing a level benchmark to compare a specific amount to an average level of earnings (for example, what an individual earns compared to the average).
50 As the primary purpose of AWE is to estimate the level of average earnings in Australia, the standard errors for the period-to-period movements are much higher proportionally than for the level estimates. Estimates of movement should be interpreted with this in mind. An alternative source for estimates of movements in the price of wages in Australia is the Wage Price Index, Australia (cat. no. 6345.0) (see paragraphs 51-56).
COMPARABILITY WITH WAGE PRICE INDEX
51 Period-to-period movements for the AWE series are not comparable with those for the Wage Price Index (WPI). It is important to recognise that the two series have different purposes and concepts and use different sample selection and estimation methodologies.
52 The AWE survey is designed to measure the level of average earnings in Australia at a point in time. It does this by obtaining data from selected businesses on the total earnings paid to their employees and the total number of employees in the business, for a specific pay period. Together, this data is used to derive the mean, or average, earnings. These sample data are then weighted to provide estimates for the whole population of in scope businesses.
53 The WPI is a price index designed to measure the change over time in the price of wages and salaries. It does this by pricing specific jobs, in terms of wage and salary payments to employees occupying the jobs, and collecting information from businesses each quarter on price changes in those jobs. It is unaffected by changes in the quality and quantity of labour services purchased by employers.
54 In addition to changes in the price of labour, AWE estimates are affected by changes in hours worked and by compositional changes in the employee workforce (see paragraphs 36 and 37). The WPI prices a fixed quantum of labour services for each job and, hence, changes to base earnings resulting from increases in hours worked or from changes in the composition of the employee workforce will not be reflected in the index.
55 For further information on comparability between AWE and WPI, please refer to the feature article Average Weekly Earnings and Wage Price Index - What do they measure? published in the May 2014 AWE release.
56 For further information on the WPI, please refer to the Explanatory Notes of Wage Price Index, Australia (cat. no. 6345.0) and Wage Price Index: Concepts, Sources and Methods, 2012 (cat. no. 6351.0.55.001) which are available on the ABS web site.
ALTERNATIVE ABS EARNINGS DATA
57 Information about wages and salaries paid to employees is used for many purposes including economic analysis, social research, policy formation and evaluation, and research by employer and employee associations. In addition to AWE, the ABS publishes a variety of other information on wages and salaries (generally referred to as 'earnings'), from both household and employer surveys. For further information on these other sources, please refer to the feature article Understanding Earnings in Australia Using ABS Statistics published in Employee Earnings, Benefits and Trade Union Membership, Australia, August 2013 (cat. no. 6310.0).
EFFECTS OF ROUNDING
58 Estimates of average weekly earnings are rounded to the nearest 10 cents.
59 Estimates of percentage change have been calculated using unrounded estimates and may be different from, but are more accurate than, movements obtained from calculating percentage changes using the rounded estimates presented in this publication.
60 Seasonal adjustment is a means of removing the estimated effects of normal seasonal variation from the series so that the effects of other influences can be more clearly recognised. Seasonal adjustment does not aim to remove the irregular or non-seasonal influences which may be present in any particular series. Influences that are volatile or unsystematic can still make it difficult to interpret the movement of the series even after adjustment for seasonal variation. If a time series has no identifiable seasonality it is not seasonally adjusted.
61 In 2012, as part of the transition from a quarterly to a biannual frequency, the ABS conducted an assessment of seasonality in the biannual AWE series. Based on the information available at the time, it was determined that moving to a biannual frequency eliminated seasonality for most AWE series and for these series the seasonally adjusted estimate was exactly equal to the original estimate. Subsequent reviews into the seasonality of biannual AWE series have shown there is a seasonal behaviour in some series that previously had no seasonal adjustment. For these series, seasonal factors are now applied to adjust the original estimate. There are other series for which seasonal adjustment is no longer applied because the seasonal behaviour previously assumed has now been assessed as insignificant. For these series the seasonally adjusted estimate will now equal the original estimate.
62 The biannual seasonally adjusted series, commencing November 2012, uses the ABS's existing quarterly seasonal adjustment method. For the purpose of seasonal adjustment, linear interpolation is used to impute "missing" quarterly original observations based on the succeeding and preceding survey estimates. In this way a quarterly original data series is synthesised from the actual biannual data collected. These synthesised estimates are used in the seasonal adjustment process and are not released. The concurrent seasonal adjustment technique is used to estimate seasonal factors from this quarterly synthesised original data.
63 Under concurrent seasonal adjustment, the estimates of seasonal factors are improved as new or revised original estimates become available each period. However, for this collection, the seasonally adjusted estimates up to May 2012, presented in the May 2012 edition, will not be revised as they were based on actual quarterly observations, where as those after that point are based on biannual observations.
64 Seasonally adjusted estimates can be smoothed to reduce the impact of irregular or non-seasonal influences. Smoothed seasonally adjusted series are called trend estimates.
65 The ABS considers that trend estimates provide a more reliable guide to the underlying direction of the original estimates and are more suitable than either the seasonally adjusted or original estimates for most business decisions and policy advice.
66 The trend estimates in this publication, obtained by dampening out the irregular component from the seasonally adjusted series, are calculated using a centred 7-term Henderson moving average of the seasonally adjusted estimates of quarterly synthesised original data. Estimates for the three most recent periods cannot be calculated using this centred average method; instead an asymmetric average is used. The changes to the moving average formulae can lead to revisions in the trend as data for subsequent periods becomes available. Revisions to the original data and re-estimation of seasonal adjustment factors also cause revisions to trend estimates. If a series is highly volatile then the trend estimates will be subject to greater revision for the latest few observations as new data become available. However, it is important to note that this does not make the trend series inferior to the seasonally adjusted or original series.
67 Please note that calculating seasonally adjusted and trend estimates on the synthesised quarterly series resulted in a slight change in the level of the data. When the new series were implemented, the change in the level of data was calculated against historic data. At the Australia level, the maximum differences for full-time adult male average weekly ordinary time earnings between estimates based on the two frequencies were $4.20 in the trend series and $4.60 in the seasonally adjusted series. Over the length of the series the mean differences were $0.48 for the trend series and $0.76 for the seasonally adjusted series.
68 Those users seeking historical seasonally adjusted and trend estimates will be required to access past AWE editions, which are available on the ABS website. It is advised that seasonally adjusted and trend estimates produced before and after the May 2012 edition are not directly comparable and these historical series before the May 2012 edition will not be produced from less frequent biannual observations.
69 The privatisation of Telstra Corporation in November 2006 impacted on the private sector and public sector AWE series. For the purposes of ABS statistics, this change from public sector to private sector was effective from March quarter 2007. The effect of this change was significant for both the private sector and public sector series. As a result, a trend break was applied to both series between November 2006 and February 2007. For more information please see Information Paper: Future Treatment of Telstra in ABS Statistics, 2007 (cat. no. 8102.0), released 26 February 2007.
70 For further information, see A Guide to Interpreting Time Series - Monitoring Trends (cat. no. 1349.0).
71 The following publications contain related information:
These documents will be presented in a new window.