Employment and Earnings, Public Sector, Australia

Although earnings estimates relate to the full twelve months of the financial year, employment estimates relate to the last pay period ending in June of the given year. Given these differences, the ABS does not recommend using these components together to calculate average earnings. This is particularly true during the 2019-20, 2020-21 and 2021-22 financial years, given the extent of short-term changes in the labour market during the pandemic.

Estimates in this release may differ from other available information due to differences in coverage and/or the classification of organisations. For further details, refer to the Comparability of results section of the Methodology.

Previously published estimates for 2020-21 have been revised as a result of new information received from organisations and improved imputation. The extent of revisions may differ between data items, states/territories, level of government and industries. These revisions are reflected in the Employees, States and territories and Industry tables, and the associated time series spreadsheet.

For inquiries about these and related statistics, contact the Customer Assistance Service via the ABS website Contact Us page. The ABS Privacy Policy outlines how the ABS will handle any personal information that you provide to us.

This release previously used catalogue number 6248.0.55.002.
 

The last public sector estimates from the Survey of Employment and Earnings (SEE), for 2021-22, were released on 10 November 2022. Those estimates were based on information collected by the ABS from an annual sample survey of public sector organisations.

From 2023, public sector employment and earnings estimates will be compiled using Single Touch Payroll (STP) data from the Australian Taxation Office. These estimates will be released as Public Sector Employment and Earnings (PSEE) to differentiate it from the previous SEE release, starting with the release of information for 2022-23.

With the development underway for PSEE, the ABS has released some early experimental estimates for 2021-22 to provide a sense of what the STP-based estimates will look like.

The experimental estimates from STP are a first look at this data, which will be refined as processes and methods are further enhanced for the first PSEE release in 2023. The first PSEE release will include estimates for both 2021-22 and 2022-23.

The new estimates in PSEE will be comparable but not exactly the same as SEE estimates, given the different data, processes and methods. Further information on this can be found in the background and methodology sections below.

The earnings estimates in this experimental release are more mature than the corresponding employment estimates. This reflects the fact that the earnings estimates are an accumulated flow measure over the financial year, while the employment estimates are a stock measure at a single point in time, derived from the payroll payments. A greater degree of revisions to employment estimates are expected to be seen when the final estimates for 2021-22 are published, given further enhancements to the method for producing level stock estimates from STP data.

There were 2,420,600 public sector employees at the end of June 2022 comprising:

• 364,300 employees in Commonwealth government;
• 1,847,400 in state government; and
• 208,900 in local government.

Total public sector cash wages and salaries in 2021-22 was $209,553.8 million.

Although earnings estimates relate to the full twelve months of the financial year, employment estimates relate to the last pay period ending in June of the given year. Given these differences, the ABS does not recommend using these components together to calculate average earnings.

Estimates in this release may differ from other available information due to differences in source, coverage and/or the classification of organisations.

The ABS has produced statistics using Single Touch Payroll (STP) data since April 2020. To date, STP-based statistics have only been released in the form of indexes in the Weekly Payroll Jobs and Wages in Australia (WPJW) publication. With access to this rich data source, the ABS has been investigating the development of new indicators from STP and is also actively pursuing opportunities to augment and/or replace direct collection from employers where possible.

The Public Sector Employment and Earnings (PSEE) is an early example of an earnings measure that has been developed using STP data. The PSEE will replace the Survey of Employees and Earnings (SEE) in 2023 for the 2022-23 financial year reference period.

Experimental 2021-22 estimates using STP data are included in this release to provide users with an early indication of the similarities and differences between SEE and PSEE, ahead of final 2021-22 estimates being released next year.

The PSEE will include employment and cash wages and salaries estimates which align with the SEE publication outputs, covering:

• Level of Government
• State x Level of Government
• Industry Division (Note small industries – Agriculture, forestry and fishing, Mining, Manufacturing, Wholesale trade, Retail trade, Accommodation and food services, Administrative and support services and Other services are collapsed into Other Industries)

One difference to note between PSEE and SEE estimates is scope changes related to the Australian permanent defence force, which will be included in STP data. Historically, SEE has excluded the defence force personnel from its estimates and only included the defence force Australian Public Sector (APS) employees. Information on the size of the Australian defence force is available in the Defence Annual Report.

The process for transforming STP data into public sector employment estimates follows a similar approach to Weekly Payroll Jobs and Wages in Australia methodology (which contains information on ‘How data are collected' and ‘How data are processed’). More detailed methodology specific to PSEE will be provided when it is released next year.

The concept of Wages and Salaries used for the STP estimates is consistent and conceptually aligned to the Australian Conceptual Framework for Measures of Employee Remuneration and the Australian System of National Accounts 2008. Employing units (profiled and non-profiled populations) are based on the ABS Business Register (ABSBR). This alignment ensures outputs compliment the ABS suite of labour and national accounts data.

The conceptual basis for STP estimates has been aligned with SEE in all aspects where practical. This includes:

• The scope of the STP is the same as that used for SEE and includes all public sector units on the ABS Common Frame;
• Outputs for STP will be based on the Australian System of National Accounts 2008 and Australian Conceptual Framework for Measures of Employee Remuneration;
• The STP count of employees is designed to be as close as conceptually possible to SEE, counting the number of employees paid on the final pay period of the financial year; and
• Estimation methods for both STP and SEE use the same weighted statistical unit approach.

Methods have been developed to:

• Ensure effective alignment with the conceptual basis for SEE; and
• Maximise the value of using STP data, which is administrative data that is close to a complete census of public sector businesses (as well as most private sector businesses).

One of the components of cash wages and salaries includes salary sacrifice. STP phase 1 data has some limitations and is not able to provide a complete measure of salary sacrifice. As such, until STP phase 2 data is mature enough to use, salary sacrifice amounts will be modelled using other available ABS data. The current experimental modelling approach uses the annual ratio of salary sacrifice to cash wages and salaries to estimate the salary sacrifice amounts. The annual ratio is derived by applying the annual movement of the Average Weekly Earnings (AWE)'s ratio to the latest SEE cycle's ratio.

Analysis to date has shown that transitioning from survey-based estimates to administrative based estimates will introduce some differences. Estimates will not align exactly, given the different data sources, processes and methods. These include;

• The data sources are different. SEE collects reported payroll data from government organisations for statistical purposes, while STP data reflects payroll reporting to the ATO for taxation (and other) purposes.
• STP uses the ABS Common Frame population and applies a coverage adjustment (weighting) to the STP data to account for any discrepancies between the two populations. SEE applies Sample Frame Maintenance Procedures to adjust for discrepancies in business operations and the ABS Common Frame.
• STP data processing procedures are quite different from SEE. The STP data needs to be transformed through a few steps including editing, imputation, auto correction and aggregating from monthly to annual information before it can be used to produce estimates. SEE data is transformed from a sample of reporting units and as such is subject to sampling and non-sampling error (Employment and Earnings, Public Sector, Australia methodology, 2021-22 financial year | Australian Bureau of Statistics (abs.gov.au).
• For the ABS Maintained Population STP requires an ABN-TAU Mapping method, to distribute ABN level input data to match the units model enabling it to make use of the ABS Common Frame classifications. SEE receives the reported data at the ABS unit’s model level as requested for a selected business.

The ABS will focus on refining and enhancing the processes and methods used for the PSEE, to ensure the estimates are robust and a suitable replacement for SEE.

The ABS expects to release PSEE with similar timing to when SEE estimates have traditionally been released, in November.

For further information or feedback on this release, please contact us at labour.statistics@abs.gov.au.

04/08/2023 - additional footnote added to tables:

• Public sector employees and cash wages and salaries
• Public sector employees, level of government, states and territories
• Public sector wages and salaries, level of government, states and territories and
• Public sector employees and cash wages and salaries, industry.