This is not the latest release View the latest release

# Australian Industry methodology

Reference period
2017-18 financial year
Released
31/05/2019

## Explanatory notes

### Introduction

1 This publication presents estimates of the economic and financial performance of Australian Industry in 2017-18. The estimates are produced annually using a combination of directly collected data from the annual Economic Activity Survey (EAS), conducted by the Australian Bureau of Statistics (ABS), and Business Activity Statement (BAS) data provided by businesses to the Australian Taxation Office (ATO).

### Reference period

The period covered by the collection was, in general, the 12 months ended 30 June of the relevant year. Where businesses were unable to supply information on this basis, an accounting period for which data can be provided was used for data other than those relating to employment. Such businesses have made a substantial contribution to some of the estimates presented in this publication. As a result, some estimates reflect trading conditions that prevailed in periods outside the twelve months ended June in the relevant year. For more information refer to the Technical Note: Off-June Year Adjusted Estimates in this issue.

Although financial estimates related to the full 12 months, employment estimates related to the last pay period ending in June. As a result, estimates of wages and salaries per employee may have been affected by any fluctuations in employment during the reference period.

Financial data incorporated all business units in scope of the EAS that were in operation at any time during the year. They also included any temporarily inactive units, i.e. those units which were in the development stage or were not in operation, but still existed and held or acquired assets and liabilities and/or incurred some non-operating expenses (e.g. depreciation, administration costs).

### Classifications

The businesses that contributed to the statistics in this release were classified:

• by state and territory

### Scope

The scope of the EAS consisted of all business entities operating in the Australian economy during 2017-18, except for:

• in most industries, entities classified to SISCA Sector 3 General government. This exclusion particularly affected data presented for Public administration and safety, Education and training and Health care and social assistance (ANZSIC Divisions O, P and Q respectively), in that the estimates related only to private sector businesses. Note, however, that SISCA Sector 3 General government businesses classified to Water supply, sewerage and drainage services (ANZSIC Subdivision 28, within Division D) were included - that is, data for relevant local government organisations (for example) were included in the estimates.
• entities classified to ANZSIC Subdivisions 62 Finance and 63 Insurance and superannuation funds. Note that estimates included in this release for Total selected industries exclude ANZSIC Subdivision 64 Auxiliary finance and insurance services. Estimates for this subdivision are presented as a separate data cube in this issue.
• entities classified to ANZSIC Subdivisions 75 Public administration, 76 Defence and 96 Private households employing staff and undifferentiated goods- and service-producing activities of households for own use.

Government owned or controlled Public Non-Financial Corporations were included.

### Coverage

This section discusses frame, statistical units, coverage issues and improvements to coverage.

### Frame

Businesses contributing to the estimates in this release were sourced from the ABS Business Register (ABSBR), which has two components as described below.

### Statistical units

10 The ABS uses an economic statistics units model on the ABSBR to describe the characteristics of businesses and the structural relationships between related businesses. Within large and diverse business groups, the units model is used to define reporting units that can provide data to the ABS at suitable levels of detail.

11 In mid 2002, the ABS commenced sourcing its register information from the Australian Business Register (ABR) and at that time changed its business register to a two population model. The two populations comprise what is called the Profiled Population and the Non-Profiled Population. The main distinction between businesses in the two populations relates to the complexity of the business structure and the degree of intervention required to reflect the business structure for statistical purposes.

### Non-profiled population

12 The majority of businesses included on the ABSBR are in the Non-Profiled Population. Most of these businesses are understood to have simple structures. For these businesses, the ABS is able to use the Australian Business Number (ABN) as the basis for a statistical unit. One ABN equates to one statistical unit.

### Profiled population

13 For a small number of businesses, the ABN unit is not suitable for ABS economic statistics purposes and the ABS maintains its own units structure through direct contact with businesses. These businesses constitute the Profiled Population. This population consists typically of large or complex groups of businesses. The statistical units model below caters for such businesses:

• Enterprise group: This is a unit covering all the operations in Australia of one or more legal entities under common ownership and/or control. It covers all the operations in Australia of legal entities which are related in terms of the current Corporations Law (as amended by the Corporations Legislation Amendment Act 1991), including legal entities such as companies, trusts and partnerships. Majority ownership is not required for control to be exercised.
• Enterprise: The enterprise is an institutional unit comprising:
• a single legal entity or business entity, or
• more than one legal entity or business entity within the same enterprise group and in the same institutional subsector (i.e. they are all classified to a single SISCA subsector).
• Type of activity unit (TAU): The TAU is comprised of one or more business entities, sub-entities or branches of a business entity within an enterprise group that can report production and employment data for similar economic activities. When a minimum set of data items is available, a TAU is created which covers all the operations within an industry subdivision (and the TAU is classified to the relevant subdivision of the ANZSIC). Where a business cannot supply adequate data for each industry, a TAU is formed which contains activity in more than one industry subdivision.

### Coverage issues

14 The ANZSIC based industry statistics presented in this publication were compiled differently from activity statistics. Each ABN unit or TAU on the ABSBR has been classified (by the ATO and the ABS respectively) to its single predominant industry class, irrespective of any diversity of activities undertaken.

15 Some businesses engage, to a significant extent, in activities which are normally carried out by different industries. For example, a predominantly mining business may also undertake significant amounts of manufacturing. Similarly, a mining business may produce significant volumes of goods which are normally produced in different mining industries. Where a business makes a significant economic contribution to industries classified to different ANZSIC subdivisions, the ABS includes the business in the Profiled Population, and 'splits' the TAU's reported data between the industries involved. Significance is determined using total income.

16 A TAU's reported data are split if the inclusion of data relating to the secondary activity, in the statistics of the industry of the primary activity, distorts (by overstating or understating) either the primary or secondary industry statistics at the ANZSIC subdivision level by:

• 3% or more, where the industries of the primary and secondary activities are in the same ANZSIC Division
• 2% or more, where the industries of the primary and secondary activities are in different ANZSIC Divisions.

17 The ABS attempts to maintain a current understanding of the structure of the large, complex and diverse business groups that form the Profiled Population on the ABSBR, through direct contact with those businesses. Resultant changes in their structures on the ABSBR can affect:

• the availability of such businesses (or units within them) for inclusion in the annual economic collections
• the delineation of the units, within those groups, for which data are to be reported.

18 The ABS attempts to obtain data for those businesses selected for direct collection and which ceased operation during the year, but it is not possible to obtain data for all such businesses.

### Improvements to coverage

19 Data in this release were adjusted to allow for lags in processing new businesses to the ABSBR, and the omission of some businesses from the register. The majority of businesses affected, and to which the adjustments applied, were small in size. As an example, the effect of these adjustments was generally 4% or less for most ANZSIC Divisions and for most states and territories.

20 Adjustments were made to include new businesses in the estimates for the period in which they commenced operation, rather than when they were processed to the ABSBR.

### Definition of key terms

22 Selected key terms are described below. Definitions for the data presented can also be found in the Glossary.

### Industry performance measures

23 This release presents a wide range of data that can be used to analyse business and industry performance.

24 Businesses report in the EAS according to Australian accounting standards applying at the time of reporting, leading to differences in reporting over time as standards are updated. In addition, differences in accounting policy and practices across businesses and industries can lead to some inconsistencies in the data input to the Australian Industry statistics. Although much of the accounting process is subject to standards, there is still a great deal of flexibility left to individual managers and accountants through the accounting policies and practices they adopt. For example, the way profit is measured is affected by management policy about such issues as depreciation rates, bad debt provisions and write off, and goodwill write off. The varying degree to which businesses consolidate their accounts may also affect any industry performance measures calculated.

25 A range of performance measures, usually referred to as ratios, can be produced from the data available from businesses' financial statements. The performance measures presented in this publication comprise:

• profitability ratios, which measure rates of profit on sales
• debt ratios, which indicate the ability of businesses to meet the cost of debt financing
• investment ratios, which indicate the capacity of business to invest in capital assets
• labour measures, which relate to output, labour costs and employment.

26 The above limitations are not meant to imply that analysis based on these data should be avoided, only that they should be borne in mind when interpreting the data presented in this publication.

27 Industry value added (IVA) is the measure of the contribution by businesses in each industry to gross domestic product. The IVA table presents estimates of the components of IVA for all industries that are within the scope of the collection.

28 There are two types of businesses: 'market' and 'non-market' producers. Market producers sell their output to achieve a profit, whereas non-market producers sell their output at economically insignificant prices. IVA is derived differently for market and non-market producers. The industries in which non-market producers make the most significant contribution to IVA are Division Q Health care and social assistance (private) and Division S Other services. See the Glossary definition of IVA for further detail.

### Survey design

29 In order to minimise the load placed on providers, the strategy for this survey was to use, as much as possible, information sourced from the ATO, thus reducing the size of the direct collect sample needed to maintain the range and quality of information available to users of statistical data. The frame (from which the direct collect sample was selected) was stratified using information held on the ABSBR. Businesses eligible for selection in the direct collect sample were then selected from the frame using stratified random sampling techniques.

30 Businesses were only eligible for selection in the survey (the direct collect sample) if their turnover exceeded a threshold level, or the business was identified as being an employing business (based on ATO information), as at the end of the reference period. Turnover thresholds were set for each ANZSIC class so that the contribution of surveyed businesses accounted for approximately 97.5% of total industry class turnover as determined by BAS data. A sample of 18,129 businesses was selected for the directly collected part of the 2017-18 EAS. Each business was asked to provide data sourced primarily from financial statements. Businesses were also asked to supply key details of their operations by state and territory, enabling production of the state/territory estimates. For the first time in 2012-13, the ABS introduced online questionnaires for business surveys.

31 Businesses which met neither of these criteria are referred to as 'micro non-employing businesses'. These businesses were not eligible for selection in the sample. For these units, BAS data were obtained and annualised, then used to model employment, income and expenses which were added to the directly collected estimates to produce the statistics in this release. For more information please refer to the Technical Note: Estimation Methodology in this issue.

### Effects of rounding

32 Where figures have been rounded, discrepancies may occur between totals and the sums of the component items. Proportions, ratios and other calculated figures shown in this publication have been calculated using unrounded estimates and may be different from, but are more accurate than, calculations based on the rounded estimates.

### Comparison with other ABS statistics

33 In some cases estimates in this release may differ slightly from those from other sources. These differences may be the result of sampling or non-sampling error, or may result from differences in scope, coverage, definitions or methodology.

### Revisions

34 Estimates for the 2015-16 and 2016-17 reference years have been revised since the previous issue of this publication. The revisions result from the review of new information received from the businesses in the direct collect sample. The revisions are incorporated in this release and in associated data cubes available free online. Note that the extent of revisions may differ for individual industries and/or between data items.

35 For more information on improvements of the methodology used for producing the estimates for micro non-employing businesses, please refer to the Technical Note: Estimation Methodology in this issue.

### Further information

36 A range of further information is available, as described below.

### Related releases

37 The following ABS releases present economy-wide and industry specific data:

• (cat. no. 8679.0)
• (cat. no. 8681.0)

38 The national accounts estimates in Australian System of National Accounts (cat. no. 5204.0) include businesses classified to industries not in scope of the EAS. This includes current price estimates on Division K Financial and insurance services, as well as Ownership of dwellings and the General government sector. For more information on the scope of the collection for Australian Industry please refer to paragraphs 6-21 above.

### Other information available

39 Most years the ABS conducts detailed industry surveys targeting specific industries of interest. For 2018-19, the ABS will be conducting a survey of the Division Q Health care and social assistance industry with results expected to be released in mid 2020. Previous issues in this release included detailed articles and data cubes relating to the following industries:

•

40 The ABS issues a daily Release Advice on its web site which details products to be released in the week ahead.

### Acknowledgement

42 ABS publications draw extensively on information provided freely by individuals, businesses, governments and other organisations. Their continued cooperation is very much appreciated; without it, the wide range of statistics published by the ABS would not be available. Information received by the ABS is treated in strict confidence as required by the Census and Statistics Act 1905

### Use of Australian Taxation Office (ATO) data in this publication

43 The results of these statistics are based, in part, on tax data supplied by the ATO to the ABS under the Income Tax Assessment Act 1936 which requires that such data are only used for statistical purposes. No individual information collected under the Census and Statistics Act 1905 is provided back to the ATO for administrative or regulatory purposes. Any discussion 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 ATO's core operational requirements.

44 Legislative requirements to ensure privacy and secrecy of these data have been followed. Only people authorised under the Australian Bureau of Statistics Act 1975 have been permitted to view data about any particular business and/or person in conducting these analyses. No information about individual taxpayers (persons) has been released to the ABS. Aggregated personal income tax data are confidentialised by the ATO before release to the ABS. 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.

## Technical note - data quality

### Reliability

1 The estimates in this release are based on information obtained from a sample survey, the Economic Activity Survey (EAS), and from administrative data collected by the Australian Taxation Office (ATO). Any collection of data may encounter factors that impact the reliability of the resulting statistics, regardless of the methodology used. These factors result in non-sampling error. In addition to non-sampling error, sample surveys are also subject to inaccuracies that arise from selecting a sample rather than conducting a census. This type of error is called sampling error.

### Sampling error

2 The majority of data contained in this release have been obtained from a sample of businesses. As such, these data are subject to sampling variability; that is, they may differ from the figures that would have been produced if the data had been obtained from all businesses in the population. One measure of the likely difference is given by the standard error, which indicates the extent to which an estimate might have varied by chance because the data were obtained from only a sample of units. There are about two chances in three that a sample estimate will differ by less than one standard error from the figure that would have been obtained if all units had been included in the collection, and about nineteen chances in twenty that the difference will be less than two standard errors.

3 Sampling variability can also be measured by the relative standard error (RSE), which is obtained by expressing the standard error as a percentage of the estimate to which it refers. The RSE is a useful measure in that it provides an immediate indication of the percentage errors likely to have occurred due to the effects of random sampling, and this avoids the need to refer also to the size of the estimate. Selected data item RSEs at the industry division level for Australia are shown in the table below. Detailed relative standard errors are available on request.

4 To illustrate, the estimate of total income for Mining in 2017-18 was $254,554m. The RSE of this estimate is shown as 0.3%, giving a standard error of approximately$764m. Therefore, there are two chances in three that, if all units had been included in the survey, a figure in the range of $253,790m to$255,318m would have been obtained, and nineteen chances in twenty (i.e. a confidence interval of 95%) that the figure would have been within the range of $253,026m to$256,082m.

5 The size of the RSE may be a misleading indicator of the reliability of the estimates for (a) operating profit before tax, (b) earnings before interest, tax, depreciation and amortisation and (c) industry value added. It is possible for an estimate to legitimately include positive and negative values, reflecting the financial performance of individual businesses. In this case, the aggregated estimate can be small relative to the contribution of individual businesses, resulting in a standard error which is large relative to the estimate.

### Relative standard errors

EmploymentWages & SalariesSales & service incomeTotal incomeTotal expensesOPBTEBITDAIVA
2017-18%%%%%%%%
Agriculture, forestry and fishing2.53.12.92.83.18.16.44.4
Mining0.50.40.30.30.40.80.40.3
Manufacturing1.20.80.60.60.63.62.71.1
Electricity, gas, water and waste services2.01.30.90.90.91.91.51.3
Construction2.02.12.92.83.16.26.42.7
Accommodation and food services4.52.72.32.32.510.28.02.9
Transport, postal and warehousing1.61.01.21.21.44.72.91.4
Information media and telecommunications1.51.00.70.70.86.22.01.1
Rental, hiring and real estate services2.22.72.64.42.97.53.62.8
Professional, scientific and technical services2.61.52.22.12.66.48.72.5
Education and training (private)2.31.93.21.91.79.97.52.1
Health care and social assistance (private)1.71.52.31.61.96.66.51.9
Arts and recreation services4.21.52.22.11.411.211.44.5
Other services2.72.33.63.13.110.256.13.1
Total selected industries(a)0.80.50.70.70.82.21.50.6

a. Excludes Division K Financial and insurance services.

### Non-sampling error

6 Error other than that due to sampling may occur in any type of collection, whether a full census or a sample, and is referred to as non-sampling error. All data presented in this publication are subject to non-sampling error. Non-sampling error can arise from inadequacies in available sources from which the population frame was compiled, imperfections in reporting by providers, errors made in collection, such as in recording and coding data, and errors made in processing data. It also occurs when information cannot be obtained from all businesses selected. The imprecision due to non-sampling variability cannot be quantified and should not be confused with sampling variability, which is measured by the standard error.

7 Although it is not possible to quantify non-sampling error, every effort is made to minimise it. Collection forms are designed to be easy to complete and assist businesses to report accurately. Efficient and effective operating procedures and systems are used to compile the statistics. The ABS compares data from different ABS (and non-ABS) sources relating to the one industry, to ensure consistency and coherence.

8 Differences in accounting policy and practices across businesses and industries can also lead to some inconsistencies in the data provided to compile the estimates. Although much of the accounting process is subject to standards, there remains a great deal of flexibility available to individual businesses in the accounting policies and practices they adopt.

9 The above limitations are not meant to imply that analysis based on these data should be avoided, only that the limitations should be considered when interpreting the data. This publication presents a wide range of data that can be used to analyse business and industry performance. It is important that any analysis be based upon the range of data presented rather than focusing on one variable.

### Reference period

10 Where businesses were unable to supply data for the 12 months ended 30 June 2018, an accounting period for which data can be provided is used for data other than those relating to employment. All businesses were asked to report employment for the last pay period ending in June 2018.

11 Estimates of financial data in some industries, such as Mining and Manufacturing, are heavily impacted by fluctuating commodity prices. In these industries, reporting for an accounting period other than the year ended 30 June can result in estimates different from what they would have been, had the business reported for an accounting period ended 30 June.

12 The impact that businesses reporting for accounting periods other than those ended 30 June 2018 had on the estimates is presented in the 'Off-June year adjusted estimates by subdivision' data cube in this release. See the Technical Note on Off-June Year adjusted estimates for more detail.

### Quality indicators

13 In the 2017-18 EAS, there was an 87.1% response rate from all businesses that were surveyed and found to be operating during the reference period. Data were imputed for the remaining 12.9% of operating businesses. This imputation contributed 9.7% to the estimate of total income for Total selected industries.

## Technical note - estimation methodology

### Introduction

1 The availability of Business Activity Statement (BAS) data collected by the Australian Taxation Office (ATO) has provided the Australian Bureau of Statistics (ABS) with opportunities to improve the efficiency of collection designs and estimation for its business surveys, while at the same time reducing the reporting burden placed on businesses.

2 Under taxation law, data may be passed by the Commissioner for Taxation to the ABS for specified statistical purposes. Accordingly, turnover and wages information sourced from ATO BAS data were used to improve the accuracy of the 2017-18 industry estimates which were produced using data items collected directly by the ABS from businesses.

#### Estimation methodology

The 2017-18 survey continued to use generalised regression estimation, first introduced in the 2006-07 survey. This estimation method enabled maximum use of observed linear relationships between data directly collected from businesses in the survey and auxiliary information.

4 When the auxiliary information is strongly correlated with data items collected in a survey, the generalised regression estimation methodology will improve the accuracy of the estimates. The auxiliary variables used in this survey were turnover and wages sourced from the BAS data of 2,126,439 businesses (including the direct collect sample).

#### Producing estimates

For the purpose of compiling the estimates in this publication, data for businesses as recorded on the ABS Business Register (ABSBR) contributed via one of three categories (or 'streams') in accordance with significance and collection-related characteristics.

The following table illustrates the ways in which Australian businesses contributed to the estimates in this publication.

#### Summary of data sources, total selected 2017-18

Type of businessCompletely Enumerated (CE) StreamGeneralised Regression Estimation StreamBusiness Activity Statement (BAS) Stream
Sources of data*ABS surveyABS Survey and ATO Business Activity StatementsATO Business Activity Statements
The number of businesses that are selected to provide data4,55113,291708,268
Contribution to total income for Total selected industries50.7%48.2%1.1%

#### Completely enumerated (CE) stream

The CE stream consisted of directly collected survey data for those units recorded on the ABSBR as being economically significant units and units significant to small state estimates.

#### Generalised regression estimation stream

The generalised regression estimation stream comprised directly collected data for those sampled units which were not in the CE stream and had turnover, in aggregate, above the bottom 2.5 percentile of BAS sales for that subdivision, or were identified as employing businesses (based on ATO information).

#### Business Activity Statement (BAS) stream

The BAS stream comprised data for those non-employing businesses whose turnover, in aggregate, was below the bottom 2.5 percentile of BAS sales for that ANZSIC subdivision.

10 Data for the BAS steam was produced using a technique that used BAS turnover to model income from sales of goods and services and BAS non-capitalised purchases to model purchases. The modelling parameters were based on the relationship between BAS data and reported data for small businesses in the direct collect sample over 3 years and were defined at the industry level. Wages and salaries were modelled as 0 since these businesses were non-employers. Employment was based on the business type of (legal) structure.

11 Estimates for each of the selected industries were produced by aggregating the contributing data streams.

#### State and territory estimates

12 For further information on the method used to compile state and territory estimates refer to the Technical Note on State and Territory Estimates in this issue.

## Technical note - state and territory estimates

### Introduction

1 The 'Australian industry by division' data cube contains annual state and territory estimates for Australian industry for the 2015-16 to 2017-18 reference periods. For earlier time periods see previous issues of this release. The estimates use a combination of data directly collected in the Economic Activity Survey (EAS) and Business Activity Statement (BAS) data sourced from the Australian Taxation Office (ATO).

#### Estimation methodology

EAS state and territory estimates are produced for employment, wages and salaries and sales and service Income. For the purpose of compiling the estimates in this publication, data for businesses as recorded on the ABS Business Register (ABSBR) contribute via one of three categories (or 'streams') in accordance with significance and collection-related characteristics. The data are produced using a combination of estimation methods. The estimation method depends on the stream:

• Stream 1: Completely enumerated and medium size multistate businesses
• Stream 2: Sampled sector units
• Stream 3: Micro Non-Employing Units (MNEUs).

#### Stream 1 - completely enumerated and medium size multistate businesses

This stream consists of units recorded on the ABSBR as being economically significant units, plus medium size businesses which according to information recorded on the ABSBR operate in more than one state or territory and have 20 or more employees.

Stream 1 units contribute to the state and territory estimates via state data reported in the EAS.

#### Stream 2 - sampled sector units

This stream comprises of units selected in EAS which are not in Stream 1 and have turnover, in aggregate, above the bottom 2.5 percentile of BAS turnover for that industry, or are identified as employing businesses (based on ATO information).

Stream 2 units contribute to the state and territory estimates via proration of their data across the states and territories.

State/territory stratification is not used for selecting the EAS sample. As a result, state and territory estimates can be inaccurate due to under or over representation of some states in the selected sample. To minimise this risk, ABS uses a synthetic estimation method. The data reported by Stream 2 units is prorated across the states and territories using state/territory proration factors. State/territory proration factors are produced based on ABSBR survey frame information. For sales and service income, BAS turnover is used; for wages and salaries and employment, BAS wages is used. The proration factors are calculated at the stratum level.

#### Stream 3 - Micro Non-Employing Units (MNEUs)

This stream comprises non-employing businesses whose BAS turnover, in aggregate, is below the bottom 2.5 percentile of BAS turnover for that ANZSIC subdivision. Stream 3 units contribute to state and territory estimates for the main state of operation.

Stream 3 units are not sent a survey form, but enumerated based on data sourced from the ATO. Data for Stream 3 are produced using BAS turnover to model sales and service income. The modelling parameters are based on the relationship between BAS data and reported data for small businesses in the direct collect sample over 3 years and have been defined at the industry level. Wages and salaries have been modelled as 0 since these businesses are non-employers. Employment is based on the business type of (legal) structure.

10 For Stream 3 units, it is assumed that the business only operates in one state or territory. This is a reasonable assumption, given these units have no staff and very low turnover. As such, all of their reported data for the above variables contributes to the estimates of their main state of operation.

## Technical note - finer level manufacturing industry estimates

### Introduction

1 The 'Manufacturing industry' data cube contains finer level estimates for the Australian Manufacturing industry for the 2015-16, 2016-17 and 2017-18 reference periods. The estimates used a combination of data directly collected in the ABS Economic Activity Survey (EAS) and Business Activity Statement (BAS) data sourced from the Australian Taxation Office (ATO).

#### Estimation methodology

The estimation method used to create the Manufacturing estimates made use of observed linear relationships between data collected from businesses in the EAS and auxiliary information available from BAS data. Where the auxiliary information was strongly correlated with data items collected in the EAS, this information was used to create predicted values for non-profiled businesses and small profiled businesses that were not selected in the survey. The auxiliary variables used to create predicted values were:

• BAS total sales (to model sales and service income)
• BAS wages and salaries (to model wages and salaries, industry value added (IVA) and employment).

#### Producing estimates

3 For the purpose of compiling the estimates in this publication, data for businesses as recorded on the ABS Business Register (ABSBR) contributed via one of three categories (or 'streams') in accordance with significance and collection-related characteristics.

4 The following table illustrates the ways in which division C Manufacturing businesses contributed to the estimates to the finer level estimates for the Manufacturing industry.

#### Summary of data sources 2017-18

Sources of data*ABS survey (Direct Collect)ABS Survey and ATO Business Activity Statements (Modelled)ATO Business Activity Statements (Partially Modelled)
Contribution to sales and service income for Manufacturing industry66.7%32.6%0.7%

#### The survey stream

The survey stream consisted of businesses with directly collected EAS data.

#### The modelled stream

The modelled stream included all businesses not selected in the EAS (the survey stream) whose turnover, in aggregate, was above the bottom 2.5 percentile of BAS sales for that industry, or were identified as employing businesses (based on ATO information).

7 Modelled data were created through the use of robust, trimmed regression estimators, which used survey data regressed against BAS data. The BAS data were found to have a high correlation with corresponding data from the EAS. The regression factors were obtained by utilising units from the survey stream and comparing their reported survey data with their matching BAS data. These regression factors were created at the ANZSIC subdivision level.

8 Sales and service income was modelled using BAS total sales as the auxiliary variable; wages and salaries, employment and IVA were modelled using BAS wages and salaries. Modelling of employment also took into account the business type (i.e. type of legal organisation) using a factor created at the ANZSIC division level. Modelled data for units in the modelled stream were created by multiplying their BAS data by the calculated regression factors.

#### Business Activity Statement (BAS) stream

9 The BAS stream comprised data for those non-employing businesses whose turnover, in aggregate, was below the bottom 2.5 percentile of BAS sales for that ANZSIC subdivision.

10 Data for the BAS steam was produced using a technique that used BAS turnover to model income from sales of goods and services and BAS non-capitalised purchases to model purchases. The modelling parameters were based on the relationship between BAS data and reported data for small businesses in the direct collect sample over 3 years and were defined at the industry level. Wages and salaries were modelled as 0 since these businesses were non-employers. Employment was based on the business type of (legal) structure.

11 Initial national ANZSIC class and state/territory ANZSIC subdivision estimates for the Manufacturing industry were produced by aggregating the contributing data streams.

#### State and territory ANZSIC subdivision estimates

12 Additional rules were applied to produce state/territory ANZSIC subdivision estimates:

• for businesses (from any stream) operating in only a single state or territory, their initial estimates contributed to the relevant state or territory and ANZSIC subdivision estimates.
• for businesses (from the survey stream) operating in more than one state or territory, their initial estimates (i.e. directly collected EAS data) contributed to the states and territories in alignment with the EAS methodology.
• for businesses (from the modelled stream) operating in more than one state or territory, their initial estimates were prorated across the states and territories in which they operated, based on a factor calculated at the ANZSIC division level from surveyed multi-state units of similar size. These modelled multi-state businesses accounted for only a small proportion of the estimates.

13 The ANZSIC class Manufacturing estimates for 2017-18 were created subject to the constraint of being additive to national ANZSIC subdivision estimates produced from the EAS. This was also true for state/territory estimates: the state/territory estimates within an ANZSIC subdivision were constrained to sum to the EAS estimate. This meant that the aggregate across all state/territory estimates for a given subdivision aligned with the EAS national subdivision estimate.

14 However, the aggregate across all ANZSIC subdivision estimates for a given state/territory were not constrained to add to the state/territory by ANZSIC division level EAS estimates. Consequently, for each state and territory, there are minor differences between the division level estimates contained in this data cube and EAS estimates presented in the other data cubes in this release.

#### Assumptions in the model

15 The quality of estimates depends on the validity of the following assumptions underpinning the modelling:

• the national ANZSIC subdivision estimates and state/territory division estimates produced from the EAS were of sufficient quality to warrant disaggregation, respectively, at ANZSIC class level and state/territory level
• it was valid to distribute the difference between EAS national subdivision estimates and the initial subdivision estimates, based on the size of the modelled stream
• the relationship between the EAS data items and the BAS data items was meaningful and consistent. Analysis supports this assumption, with the correlation being of consistent quality to produce reliable estimates
• the auxiliary (BAS) data was of high quality
• the industry coding was accurate on the ABS maintained Business Register.

Users should consider the suitability of these assumptions when interpreting the estimates.

## Technical note - Off-June year adjusted estimates

### Introduction

1 The data collected in the Economic Activity Survey (EAS) generally represent the 12 month period ended 30 June of the relevant year. However, where businesses were unable to provide data on this basis, an alternate, or Off-June year accounting period was used. As a result, in some instances estimates may reflect trading conditions occurring outside of the published reference year. The 'Off-June year adjusted estimates by subdivision' data cube provides a view of the EAS data adjusted to a June year end for all businesses.

2 The methodology used data collected through the Quarterly Business Indicators Survey (QBIS) to model the impact of Off-June year reporting on selected ANZSIC industry subdivisions. Using QBIS data, 'Off-June year' factors were generated that, when applied to EAS data, resulted in a set of estimates adjusted for Off-June year reporting. These estimates have been presented by ANZSIC division and subdivision. The data items were wages and salaries, total income, total expenses and industry value added (IVA), presented for the reference years 2015-16 to 2017-18. This information was previously published as an experimental series in Information Paper: Experimental Estimates for Australian Industry Adjusted for Off-June Year Reporting (cat. no. 8169.0).

#### The Off-June year reporting problem

3 The 2017-18 EAS had a reference period ending 30 June 2018; that is, the aim of the EAS was to measure economic activity over the 12 months from 1 July 2017 to 30 June 2018. Analysis of data from EAS showed that the majority of businesses reported for this reference period, but for some industries a substantial proportion reported for some other reference period. Off-June year reporting was more prevalent in industries with a high degree of foreign ownership, such as Mining, Manufacturing and Wholesale trade, as many countries use different accounting periods to Australia. Considering the 2017-18 EAS, the types of 'Off-June year' reporting periods typically observed were:

• reporting period ending 31 December 2017;
• reporting period ending 31 March 2018; and
• reporting period ending 30 September 2018.

4 Table 4.1 gives an estimate of the percentage of the population likely to report on a financial and Off-June year basis, and their contribution to total IVA, based on the 2017-18 EAS. The prevalence of Off-June year reporting does not vary much from year to year. However, although businesses reporting for an Off-June year financial period may be in the minority, their contribution to overall estimates of IVA can be substantial. For example, 4% of businesses in the Mining industry reported for a non-standard financial year and contributed 56% of total IVA in 2017-18.

#### Table 4.1 Prevalence of Off-June reporting in EAS 2017-18, and the contribution of these businesses to IVA

Estimate of percentage of population by reporting period(a)Estimate of percentage contribution to IVA by reporting period(a)
Financial year reportersOff-June year reportersFinancial year reportersOff-June year reporters
%%%%
2017-18
B Mining9644456
C Manufacturing9916436
D Electricity, gas, water and waste services9918119
E Construction100-9010
H Accommodation and food services9918317
I Transport, postal and warehousing100-8614
J Information Media and Telecommunications9917327
L Rental, hiring and real estate services100-928
M Professional, scientific and technical services100-8020
P Education and training (private) (b)9553169
Q Health care and social assistance (private) (b)100-955
R Arts and recreation services9918317
S Other services100-8812
Total selected industries100-7723

- nil or rounded to zero (including null cells)
a. Includes all businesses in scope of the Off-June year adjusted estimates, except for non-employing entities below a certain turnover threshold. Other businesses with less than 20 employees are included in the financial year reporter category.
b. Adjustments are applied to labour costs only.

#### Scope of the Off-June year adjustment

5 The scope of the 'Off-June year adjusted estimates by subdivision' data cube was based on the EAS, with further constraints imposed to match the scope of the QBIS. For a detailed explanation of the scope and methodology of these surveys, see the Explanatory Notes in this release and in Business Indicators (cat. no. 5676.0). In brief, the scope of the Off-June year adjusted estimates consisted of all businesses on the Australian Bureau of Statistics Business Register (ABSBR) operating in the Australian economy during the reference period, except for:

• in most industries, organisations classified to Standard Institutional Sector Classification of Australia (SISCA) Sector 3 General government. The one industry for which general government units were included was Water supply, sewerage and drainage services (ANZSIC Subdivision 28, within Division D Electricity, gas, water and waste services);
• businesses classified to ANZSIC Division A Agriculture, forestry and fishing;
• businesses classified to ANZSIC Division K Financial and insurance services;
• businesses classified to ANZSIC Subdivision 77 Public order, safety and regulatory services;
• businesses classified to ANZSIC Subdivision 96 Private households employing staff.

6 These divisions and subdivisions are not included in the QBIS and as such no adjustments could be applied to the businesses classified to these industries.

7 No adjustments were applied to data reported by Off-June year reporters with employment of 20 or less.

8 While the private sector components of Education and training and Health care and social assistance (ANZSIC Divisions P and Q, respectively) were conceptually in scope of these analyses, QBIS does not collect information on sales and service income or other expenses for these ANZSIC divisions. Thus no adjustment was applied to these data items for these industries.

#### Methodology

9 The estimates published in the 'Off-June year adjusted estimates by subdivision' data cube were derived by the following process:

• For each EAS ANZSIC subdivision in scope, subdivision Off-June year factors were determined for each data item and each of the Off-June year reporting types;
• QBIS units with incorrectly reported or unrepresentative data in the ANZSIC subdivision were removed from the subdivision factors;
• Significant EAS units that were also selected in the QBIS collection were assessed for appropriateness to receive an individualised (unit) Off-June year factor (instead of receiving a subdivision factor);
• Off-June year reporting EAS businesses were then assigned either a unit factor (if deemed appropriate) or its respective ANZSIC subdivision Off-June year type factor. New values were calculated for these businesses, representing an estimate of how the business would have reported for the standard financial year (that is, 1 July to 30 June). Final aggregated data was then produced on a standard financial year basis.

#### Creating subdivision Off-June year factors

10 It was necessary to create twenty-seven separate factors for each in scope ANZSIC subdivision, as demonstrated in Table 4.2 using the 2013-14 financial year as an example.

#### Table 4.2 The Off-June year modelling factors required for each ANZSIC subdivision

Sales and service incomeWages and salariesOther expensesClosing inventories of raw materialsOpening inventories of raw materialsClosing work-in-progress inventoriesOpening work-in-progress inventoriesClosing inventories of finished goodsOpening inventories of finished goods
ANZSIC SubdivisionDec 13Dec 13Dec 13Dec 13Dec 13Dec 13Dec 13Dec 13Dec 13
ANZSIC SubdivisionMar 14Mar 14Mar 14Mar 14Mar 14Mar 14Mar 14Mar 14Mar 14
ANZSIC SubdivisionSep 14Sep 14Sep 14Sep 14Sep 14Sep 14Sep 14Sep 14Sep 14

11 The factors were formulated from a subset of businesses sampled in the QBIS which met the following criteria:

• For sales and service income, wages and salaries and other expenses factors: Reported a non-zero value for these data items for the seven relevant quarters which covered all possible types of reporting periods. For example, for 2013-14 the relevant quarters are March 2013 through September 2014. This condition eliminated businesses which either started up or closed down during the period;
• For inventory factors: Reported a non-zero value for sales and service income and inventories for eight relevant quarters (December 2012 through September 2014) to ensure an opening inventory value;
• Did not report a value for the above items in one quarter greater than 10 times that of an adjacent quarter. This condition eliminated businesses with extreme values; and
• Had an employment size of 20 or more. This removed small businesses, whose data were not expected to be impacted by Off-June year reporting in the EAS.

12 Where there were five or less contributing QBIS businesses in an ANZSIC subdivision, it was considered that the number of observations was insufficient for producing the Off-June year factors. In those cases the Off-June year factors were produced at ANZSIC division level.

13 Sales and service income, other expenses and inventories factors were not generated for Education and training and Health care and social assistance (ANZSIC Divisions P and Q respectively), as the information is not collected by the QBIS (see Scope and Population above). For the same reason, inventory factors could only be generated for Mining, Manufacturing, Electricity, gas, water and waste service (Subdivisions 26 Electricity supply and Subdivision 27 Gas supply only), Wholesale trade, Retail trade and Accommodation and food services (ANZSIC Divisions B, C, D, F, G and H respectively).

14 Subdivision level factors were not applied (by design) for Off-June year reporters in Electricity, Gas, Water and Waste Services (ANZSIC Division D).

15 For each data item, quarterly weighted QBIS data reported by the subset of businesses established above were summed to give an aggregate value for each in scope ANZSIC subdivision. These aggregate quarterly values were then used to create factors that model the impact of Off-June year reporting for each of the four data items, by each in scope subdivision. To calculate each factor, a ratio of the summed data from the four quarters of the standard financial year is divided by the summed annualised data from the four quarters of the relevant Off-June year reporting period, as described by Equation 4.1 using the 2013-14 financial year as an example.

#### Equation 4.1. Calculating Off-June year factors

$$\large{\begin{array}{l} Off - June \ Factor_{DEC} = \frac{\left( \underline{O}_{SEP13} + \underline{O}_{DEC13} + \underline{O}_{MAR14} + \underline{O}_{JUN14}\right)} {\left( \underline{O}_{MAR13} + \underline{O}_{JUN13} + \underline{O}_{SEP13} + \underline{O}_{DEC13}\right)} \\\\ Off - June \ Factor_{MAR} = \frac{\left( \underline{O}_{SEP13} + \underline{O}_{DEC13} + \underline{O}_{MAR14} + \underline{O}_{JUN14}\right)} {\left( \underline{O}_{JUN13} + \underline{O}_{SEP13} + \underline{O}_{DEC13} + \underline{O}_{MAR14}\right)} \\\\ Off - June \ Factor_{SEP} = \frac{\left( \underline{O}_{SEP13} + \underline{O}_{DEC13} + \underline{O}_{MAR14} + \underline{O}_{JUN14}\right)} {\left( \underline{O}_{DEC13} + \underline{O}_{MAR14} + \underline{O}_{JUN14} + \underline{O}_{SEP14}\right)} \end{array}}$$

where Q is quarterly QBIS data aggregated by industry subdivision for the subset of businesses identified above.

16 Since inventories are stock variables (that is, represent a quantity existing at a particular point in time) the formulae for deriving inventories factors differed slightly, as described by Equation 4.2.

#### Equation 4.2. Calculating Off-June year inventories factors

$$\large{\begin{array}{l} Inventories \ Factor_{DEC} = \frac{\left ( \underline{O}Inv_{JUN14} \right)} {\left ( \underline{O}Inv_{DEC13} \right)} \\\\ Inventories \ Factor_{MAR} = \frac{\left ( \underline{O}Inv_{JUN14} \right)} {\left ( \underline{O}Inv_{MAR14} \right)} \\\\ Inventories \ Factor_{SEP} = \frac{\left ( \underline{O}Inv_{JUN14} \right)} {\left ( \underline{O}Inv_{SEP14} \right)} \\\\ \end{array}}$$

Factors were produced for total opening and total closing inventories.

17 The factors generated in these equations give an indication of the variability in trading conditions between Off-June year reporting periods and the standard Australian financial year. A factor of 1 indicates no variability, implying there is no effect of Off-June year reporting on estimates published in Australian Industry. Conversely, the further a factor lies from 1, the greater the impact of Off-June year reporting on industry estimates.

18 An example of the calculation of factors for Subdivision 14, Wood product manufacturing is provided below for the 2011-12 EAS. Quarterly sales and service income estimates derived from in-scope QBIS data (see Table 4.3) were used to produce Off-June year factors (see Example 4.1) which were applied to EAS estimates of sales and service income.

#### Table 4.3 Calculating factors example

Sales of goods and services, Subdivision 14

Wood product manufacturing

#### Table 4.3 Calculating factors example sales of goods and services subdivision 14 wood product manufacturing

Sales and service income estimates derived from in scope QBIS data(a)