This is not the latest release View the latest release

Labour Account Australia

The Australian Labour Account provides quarterly and annual time series for four quadrants: Jobs, People, Hours and Payments

Reference period
September 2021
Released
8/12/2021

Key statistics

In seasonally adjusted terms for the September quarter 2021:

  • Total jobs decreased 405,500 (2.7%).
  • Filled jobs decreased 377,900 (2.6%) to 14.0 million.
  • Multiple job holders decreased 7.8%.
  • Secondary jobs decreased 69,600 (7.5%).
  • Hours worked decreased 4.7%.
  • Employed people decreased 2.3% to 13.2 million.
 Seasonally Adjusted

Jun qtr 2021 to Sep qtr 2021
% change

Sep qtr 2020 to Sep qtr 2021
% change

Total Jobs -2.73.3
Filled Jobs-2.62.4
Main Jobs-2.32.2
Secondary Jobs-7.55.8
Job Vacancies-7.462.4
Hours Actually Worked-4.71.2
Average Hours Actually Worked Per Job-2.1-1.1
Average Income Per Employed Person3.52.1

Data impacts and changes

Revisions this quarter

Data in the four quadrants of the Labour Account have been revised from previously published estimates. This reflects revisions resulting from the usual review of seasonal factors, and revisions to data sources, but also one-off revisions from the historical revisions to the Australian National Accounts, the outcomes from a recent review of Labour Account hours worked series and methods, and a range of other method and data source enhancements and changes. These changes have resulted in revisions across the entire time series of Labour Account data.

See the 'Historical revisions to the Labour Account' section in this release for more information. Revisions to published estimates can be found in Table 22 of this release.

Improving the estimation of short-term non-residents in the Labour Account

As noted in previous Labour Account releases, the ABS has reviewed the modelling approach for estimating the number of short-term/temporary non-residents who are employed and made revisions to the estimates for this group. The new approach has been implemented along with the range of other improvements as part of the Labour Accounts historical revisions. For further information, see the 'Historical revisions to the Labour Account' section in this release.

Reinstatement of public and private sector estimates

Public and private sector estimates were temporarily suspended in the March quarter 2021 release, given the extent of large quarterly movements in late 2020 and early 2021. The ABS has reviewed these estimates, as additional data have become available, and has reinstated them in this release.

Upcoming changes to thematic grouping and navigation for labour statistics on the ABS website

Labour statistics have traditionally been organised into two thematic groupings on the ABS website: 'Employment and unemployment'; and 'Earnings and work hours'. The ABS is planning to update these themes to better reflect the current range of available labour statistics, better align key labour market concepts with website themes and navigation, and improve discoverability.

The planned new themes are:

  • ‘Employment and unemployment’
  • ‘Jobs’
  • ‘Earnings and working conditions’
  • ‘Labour accounts’

These new thematic groupings and navigation (and the statistical releases which will appear under each theme) can be viewed on our Beta website, at beta.abs.gov.au/statistics.html#labour.

The key changes include:

  • addition of a 'Jobs' theme - to reflect the increasing range of jobs-related data ABS is releasing – including Weekly payroll jobs and wages and Jobs in Australia (currently in 'Earnings and work hours'), in addition to the longstanding statistics on Job vacancies and Job mobility (currently in 'Employment and unemployment').
  • addition of a 'Labour accounts' theme - for cross-cutting quarterly and annual labour account data on jobs, people, hours and payments (currently in 'Employment and unemployment').
  • renaming the 'Earnings and work hours' theme to 'Earnings and working conditions' - to provide a clearer pathway the large range of information available on working conditions beyond wages (work arrangements, flexibility, workplace relations, etc) and reflect that key hours data are available from the 'Employment and unemployment' theme (from the monthly Labour Force Survey) and the Labour account.

The changes are expected to be implemented in the first half of 2022. While implementation of these changes will result in a change to the placement of some statistical releases, and their respective URLs, there will be automatic redirects put in place. These redirects will ensure that existing URLs and bookmarks will continue to work.

If you have any feedback or questions on this new approach, please email us at labour.statistics@abs.gov.au.

Jobs

In seasonally adjusted terms for the September quarter 2021:

  • Filled jobs decreased by 2.6%, following a 1.4% rise in the June quarter 2021. Filled jobs grew 2.4% through the year.
  • The number of main jobs decreased by 308,300 (or 2.3%).
  • The number of multiple job holders decreased by 7.8%.
  • The proportion of vacant jobs decreased to 2.4%, from the 2.5% recorded in the June quarter 2021.
  • The number of public sector jobs decreased by 0.5%, while the number of private sector jobs decreased by 2.9%.

Total jobs

In seasonally adjusted terms for the September quarter 2021, the total number of jobs decreased by 405,500 (or 2.7%), made up of a decrease of 27,600 job vacancies and a decrease of 377,900 filled jobs.

Filled jobs

In seasonally adjusted terms for the September quarter 2021, the number of filled jobs decreased by 377,900 to 14.0 million. 

Filled jobs, by industry, Sep qtr 2021, seasonally adjusted
Filled Jobs ('000)Proportion of all Industries (%)
Agriculture, forestry and fishing (A)438.93.1
Mining (B)180.21.3
Manufacturing (C)837.86.0
Electricity, gas, water and waste services (D)119.50.9
Construction (E)1,156.78.2
Wholesale trade (F)558.84.0
Retail trade (G)1,369.59.8
Accommodation and food services (H)1,023.87.3
Transport, postal and warehousing (I)647.14.6
Information media and telecommunications (J)180.91.3
Financial and insurance services (K)478.63.4
Rental, hiring and real estate services (L)286.52.0
Professional, scientific and technical services (M)1,178.48.4
Administrative and support services (N)997.77.1
Public administration and safety (O)751.75.4
Education and training (P)1,017.07.2
Health care and social assistance (Q)2,067.914.7
Arts and recreation services (R)238.01.7
Other services (S)517.13.7
Total all industries14,046.2100.0
Change in filled jobs, by industry, Sep qtr 2021, seasonally adjusted
Quarterly change (%)Annual change (%)
Agriculture, forestry and fishing (A)2.80.8
Mining (B)-2.9-1.5
Manufacturing (C)-2.80.9
Electricity, gas, water and waste services (D)-5.10.4
Construction (E)-0.54.5
Wholesale trade (F)-4.3-2.6
Retail trade (G)-4.1-1.5
Accommodation and food services (H)-12.2-3.6
Transport, postal and warehousing (I)-4.16.3
Information media and telecommunications (J)-1.3-1.7
Financial and insurance services (K)-2.54.8
Rental, hiring and real estate services (L)-0.54.9
Professional, scientific and technical services (M)-0.82.6
Administrative and support services (N)1.68.6
Public administration and safety (O)-0.20.2
Education and training (P)-3.9-2.4
Health care and social assistance (Q)0.46.8
Arts and recreation services (R)-12.92.6
Other services (S)-2.910.7
Total all industries-2.62.4

Secondary jobs

Secondary jobs are where a person is working more than one job at the same time, and may consist of one or more additional jobs. These jobs can be held by people who have their main job in the same or a different industry. 

In seasonally adjusted terms for the September quarter 2021:

  • Secondary jobs decreased by 69,600 (or 7.5%). 
  • The proportion of secondary jobs to filled jobs was 6.1% compared to 6.4% in the previous quarter.

The three industries with the highest number of secondary jobs were Administrative and support services, Health care and social assistance and Education and training.
 

Statistical discrepancy - Filled jobs

The Labour Account compiles independent estimates of the number of filled jobs from both a household and business perspective. The difference between these two estimates is referred to as the "statistical discrepancy". This discrepancy is reduced to zero through the balancing processes of the Labour Account, through producing a single harmonised or "balanced" number of filled jobs for each industry and the total economy.

In original terms the discrepancy between household sources and business sources was 1,459,300 jobs, or 10.4% of the household estimate, in the September quarter 2021.

While the business sources have been showing stronger jobs growth over the COVID period, this hasn't impacted on overall Labour Accounts aggregates as the Labour Account jobs estimates are constrained to the household side. This is because the Household side source, the monthly Labour Force Survey, provides the best measure of labour market activity. 

Balancing decisions for Rental, hiring and real estate services and Other services were mostly based on household survey sources. All other industries were mostly based on business survey sources.

People

In seasonally adjusted terms for the September quarter 2021:

  • The total number of employed people decreased by 2.3% to 13.2 million. 
  • There were 621,600 unemployed people, a decrease of 71,300 people from June quarter 2021.
  • There were 1,225,300 underemployed people, an increase of 190,200 people from June quarter 2021.

The Labour Account produces the number of people employed from an industry perspective. As a result, the sum of employed people for each industry division does not equal the total number of people employed in the whole economy, given some people are employed in multiple industries.

The three industries with the highest number of employed people in the September quarter 2021 were Health care and social assistance, Retail trade and Professional, scientific and technical services.

Hours

In seasonally adjusted terms for the September quarter 2021, the total number of hours actually worked decreased by 253.7 million hours (or 4.7%) to 5.1 billion hours.

The three industries with the highest number of hours actually worked in the September quarter 2021 were Health care and social assistance, Professional, scientific and technical services and Construction.

Average hours worked per job is the hours actually worked divided by all filled jobs.

In seasonally adjusted terms for the September quarter 2021, average hours worked per job decreased by 2.1% to 366 hours. 

Payments

The Labour Account Payments quadrant presents the costs incurred by enterprises in employing labour, and the income received by people from its provision. Total income consists of compensation of employees and labour income from self-employment. The addition of other related costs to employers to total income will derive total labour costs.

In seasonally adjusted terms for the September quarter 2021:

  • Total labour income increased by $2,955 million (or 1.1%) to $280,026 million. 
  • The average labour income per employed person increased by 3.5% to $21,289.
  • Total compensation of employees increased by 0.4% to $252,511 million.
  • Labour income from self-employment increased by 7.6% to $27,515 million.
  • Total labour costs increased by $692 million (0.2%) to $299,036 million. 

The three industries with the highest total labour income in the September quarter 2021 were Health care and social assistance, Professional, scientific and technical services and Construction.

Historical revisions to the Labour Account

A large number of revisions have been made to the Labour Account in the September quarter 2021, as part of a major historical revisions process.

In addition to reflecting revisions from the recent National Accounts historical revisions process, the Labour Account historical revisions process also reflects the outcomes from a recent review into the approach used to estimate hours worked in the Labour Account and a range of other targeted enhancements to some other methods and data sources. This was the first ever historical revisions process for the Labour Account, which was introduced as one of the first in the world in 2017.

These changes have resulted in revisions across the entire time series of Labour Account data, from September 1994 onwards.

The changes implemented into the Labour Account this quarter include:

  • Australian System of National Accounts historical revisions
  • revisions to hours worked identified through a recent methods review
  • improving the estimation of short-term non-residents in the labour market
  • improving the method for balancing main and secondary jobs
  • improving the method used to derive multiple job holder estimates
  • improving the method for estimating the number of child workers over the COVID period
  • standardising the seasonal adjustment process across all industries

More detail on each of these changes is provided below. Given the number of changes and how inter-connected they are, it is not possible to quantify the individual impact of all of these revisions and changes. An indication of the magnitude of changes is provided where possible.

The overall impact on the Labour Accounts estimate of filled jobs is shown in Graph 1. It shows the revised estimates being slightly lower across most of the historical series.

Australian System of National Accounts historical revisions

The Australian System of National Accounts recently underwent a periodic historical revision process, which was implemented ahead of the September quarter 2021. These revisions focused on improving data quality, adopting new classifications and improving international comparability. As the revisions were constrained to a small number of targeted changes, this historical revision resulted in smaller changes to the National Accounts than previous historical revisions. These revisions to the National Accounts also flowed through to the Labour Account.

For further information, refer to Improved estimates of the annual national accounts: Results of the 2021 historical revisions

Revisions to hours worked

A range of improvements were identified through a recent review into the methods used to produce hours worked estimates in the Labour Account.

The key change relates to the use of discretionary adjustments to hours worked at the industry level. Since the Labour Account was introduced in 2017, a large and increasing number of discretionary adjustments have been applied to hours worked estimates at the industry sub-division and division level, to maximise coherence between the four quadrants at the industry level and limit volatility over time.

However, while intended to increase coherence and minimise the inherent volatility in the series, the review identified that the combined effect of these discretionary adjustments was resulting in the 'oversmoothing' of changes in the hours worked series. This oversmoothing resulted in growth in aggregate Labour Account hours worked which was understated and less coherent with changes in the growth in aggregate Labour Force hours worked data. The ABS has now removed most of these discretionary adjustments.

The review also identified an improvement to the method used to derive hours worked in secondary jobs, which were slightly understated. The new method apportions the hours worked in secondary jobs from the Labour Force annual hours benchmark using a ratio of hours worked in main job and all jobs. In addition to producing a more robust estimate of hours worked in secondary jobs, the new method also produces estimates that are more comparable with other Labour Account estimates.

Graph 2 shows the revisions to the hours worked series. As noted earlier, these revisions reflect the above method enhancements but also the full range of other changes that have been implemented through the historical revisions.

Improving the estimation of short-term non-residents in the labour market

Over the COVID period there have been large reductions in short-term non-resident arrivals in Australia. These reductions are not reflected in Labour Force employment estimates but are accounted for in Labour Account estimates of employment and jobs. Differences between the Labour Force Survey and Labour Account were outlined in the June 2021 Labour Force release.

As foreshadowed in previous Labour Force and Labour Account releases, a new model to estimate the number of short-term/temporary non-residents who are employed has been implemented into the Labour Account with the September quarter 2021. This new method, which incorporates improved data sources and method enhancements, provides a more reliable indicator of the size of this group, and produces aggregates that are more consistent with changes in other labour market aggregates, such as Labour Force estimates.

Previously, estimates for this group needed to be heavily modelled using the flows of non-residents arriving into Australia (on certain visa classes). The new method uses newly derived stock estimates of short-term visitor arrivals who have a visa with working rights to work in Australia, of which a proportion are estimated to be employed, and better accounts for the net change in the number of non-residents in Australia (arrivals and departures).

This change in method has resulted in downward revisions to Labour Account series across the People, Jobs and Hours quadrants. The revised Labour Account data now reflects that prior to the start of the pandemic there were generally between 100,000 and 150,000 employed short-term non-residents in Australia each quarter. Over the COVID period, this fell to around 5,000 as a result of border restrictions.

In addition to this change in the method used to estimate the total number of employed short-term non-residents, there has also been a change to the way main and secondary jobs are balanced in the Labour Account. This has also resulted in further revisions to short-term non-residents estimates (along with other people outside of the scope of the Labour Force Survey, including child workers and defence force personnel). This change is discussed below and has also increased the coherence between the Labour Account and Labour Force employment estimates.

Improving the method for balancing main and secondary jobs

The method of allocating main and secondary jobs as part of the process of balancing household and business side sources in the Labour Account has been improved. This provides greater consistency with other balancing processes in the Labour Account, and produces a more robust allocation of main and secondary jobs for the groups out of scope of the Labour Force Survey (i.e. employed short-term non-residents, child workers and defence force personnel).

The new method constrains the aggregate level of main jobs and employed people to the total employed in the Labour Force Survey after accounting for scope differences. This is achieved by changing the way that main and secondary jobs are apportioned, at the industry level, in the balancing process, using the approach that has been used to balance hours worked.

As a result, there has been a downward revision to the number of secondary jobs and an upward revision to the number of main jobs in the Labour Account.

Graph 3 shows these revisions have resulted in the reallocation of secondary jobs to main jobs (e.g. around 150,000 jobs in the June quarter 2021). The total number of filled jobs is unchanged from this change in method, although is subject to revisions from other changes that have been implemented as part of the historical revisions.

Improving the method used to derive multiple job holder estimates

Data on the number of people who had more than one job, and the number of jobs they held, are sourced from the monthly Labour Force Survey. The Linked Employer Employee Database (LEED), which is the data source for the Jobs in Australia and Personal Income in Australia releases, is used to provide an industry breakdown for multiple job holders.

Historically, the Labour Account has used Labour Force multiple job-holder data from the mid-month of each quarter (i.e. February, May, August and November) to produce quarterly stock estimates of multiple job-holders. However, while this was consistent with the reference period for the Labour Force industry data, which is also in respect of the mid-month of the quarter, it was not consistent with the other Labour Account stock estimates, which are all based on the end of quarter month (March, June, September and December). To improve the coherence between stock estimates in the Labour Account, the Labour Force end-of-quarter month estimates are now used to derive multiple job holder stocks estimates and the multiple job holding rates and ratios.

Graph 4 shows the impact of this change in reference period. While the overall trajectory of the revised series is consistent with the historical series over time, the level estimates for the number of multiple job-holders (and hence the multiple job-holder rates) are generally slightly lower.

Improving the method for estimating the number of child workers over the COVID period

As the Labour Account is designed to measure all employment activity, it includes an estimate of the number of child workers (i.e. employed people under the age of 15), who are out of scope of the monthly Labour Force Survey.

To better account for the impact of the pandemic on child workers, across all quadrants of the Labour Account, an adjustment has been applied from June quarter 2020 onwards, to directly deflate estimates for this population using changes in the employment of 15 year old people.

This change resulted in a downward revision of around 60,000 employed people in June quarter 2020, with subsequent increases reflecting the recovery in employment, prior to the Delta outbreak.

The ABS will monitor the ongoing need for this adjustment over time, including comparing changes in data for people under the age of 15 in Jobs in Australia for 2019-20, when it is available in 2022.

Standardising the seasonal adjustment process across all industries

Previously, there were two different approaches used to seasonally adjust series at the industry level, depending on the size of the industry.

The six largest employing industries were indirectly seasonally adjusted, meaning the aggregate estimates at the division level were the sum of the seasonally adjusted sub-divisions. The other thirteen industries were seasonally adjusted directly (i.e. at the division level).

The ABS is now using a direct seasonal adjustment approach for all industries, regardless of their size.

Data downloads

Changes to Excel file format on the ABS website

In line with updating to more recent technology formats, the ABS will progressively transition to releasing Excel files (e.g. timeseries spreadsheets) in .xlsx format. While this change will improve usability, it may also require changes to automated macros or similar programs that call on the current .xls format.

For Labour Account spreadsheets, this change will take effect from the December quarter 2021, due for release on 9 March 2022. Previously released data will not change.

Changes will be reflected in other labour statistics from the following dates:

  • Characteristics of Employment, to be released on 14 December 2021
  • Employee Earnings, to be released on 14 December 2021
  • Working arrangements, to be released on 14 December 2021
  • Labour Force, Australia, to be released on 16 December 2021
  • Job Vacancies, to be released on 12 January 2022
  • Employee Earnings and Hours, to be released on 19 January 2022
  • Average Weekly Earnings, to be released on 24 February 2022
  • Industrial Disputes, to be released on 10 March 2022

Some labour statistics, such as Weekly Payroll Jobs and Wages in Australia, already publish data in .xlsx format.

Time series spreadsheets

Quarterly estimates for Jobs, People, Hours and Payments by Industry Division and Total All Industries.

Data files

Data Explorer datasets

Annual estimates for Jobs, People, Hours and Payments by Industry Subdivision, Division and Total All Industries.

Caution: Data in Data Explorer is currently released after the 11.30am release on the ABS Website. Please check the reference period when using Data Explorer.

Labour Account balanced - Balanced annual estimates for Jobs, People, Hours and Payments by Industry Subdivision, Division and Total All Industries.

Labour Account unbalanced - Unbalanced annual estimates for Jobs, People, Hours and Payments by Total All Industries.

For information on Data Explorer and how it works, see the Data explorer user guide.

Previous catalogue number

This release previously used catalogue number 6150.0.55.003.