Guide to labour statistics

Labour statistics are about people. They measure people's participation in work – their earnings, type of work, working hours and conditions – as well as their success in, or barriers to, finding work.

Labour statistics provide insight into the economy and the effect of policy settings on the population through measures related to labour supply and demand, as well as its price.

Labour statistics sit at the intersection of economic and social statistics. They provide information about the size and structure of Australia's workforce required for policy formulation, evaluation and macro-economic modelling, and measure relationships between employment, income and other social and economic characteristics.

Four key components related to work and the broader labour market are measured:

• People
• Jobs 
• Hours
• Payments 

The various data sources and available information related to people, jobs, hours and payments are designed for different purposes, and can have different definitions, timings or features. We encourage you to use a range of measures together to better understand the state of the labour market. For more information, see Scope of Australian labour statistics and Uses and users of labour statistics in Labour Statistics: Concepts, Sources and Methods.

ABS labour statistics are drawn from four key types of data sources, or "pillars" of data, which provide complementary insights into the labour market across a range of topics. 

For more information, see Methods: Four pillars of labour statistics in Labour Statistics: Concepts, Sources and Methods.

Our labour-related statistical releases inform on issues related to work and the labour market. These releases are drawn from a variety of different data sources and are available on the ABS website.

Each release generally includes a high-level overview of findings, spreadsheets containing more detailed data, and a methodology page providing explanatory information on the methods used to produce the release. Some releases also have accompanying TableBuilder and/or microdata products to enable you to produce customised tables or undertake more detailed analysis.

Most releases include a media release covering high level statistical information drawing out the main stories of interest from the latest statistics.

Some labour statistics releases are accompanied by articles providing in depth analysis of a statistical topic of interest. They explain the data, changes over time and underlying factors.

The tables below show our releases grouped according to the labour market theme(s) they provide data for.

There are a number of ways to access more detailed data through self-service or working with the ABS, in addition to what is available on the ABS website including:

• TableBuilder – Create, save and download your own tables using many topics
• Microdata – Basic and detailed unit record information released through the DataLab to approved users for statistical analysis and research
• Integrated data – The ABS is an accredited Integrating Authority, and works with government, research institutions and businesses on data integration projects that have a clear public benefit
• Customised data requests – Consultancy service providing aggregate data tailored to meet your needs

Some of these services may incur a fee, see Data Services for more information.

• Labour Statistics: Concepts, Sources and Methods provides a comprehensive description of the concepts and definitions underpinning labour statistics and the data sources and methods used to compile them. It should be read together with release specific methodology pages.
• Guide to labour statistics is a summary resource on key labour market topics, designed to help people to understand labour statistics and identify the best source for their analysis. Additional topics will be added over time.

We produce a wide range of earnings statistics from a mix of data sources, for many different purposes. It can be challenging to choose the right earnings data to use. This guide will help you to understand the different features of our various earnings measures and sources and choose the correct measure to suit your needs.

Earnings statistics generally refer to gross (pre-tax) cash wages and salaries paid to employees at regular intervals for work done as well as paid leave. They exclude irregular payments, employers' social contributions and severance and termination pay, as well as the value of 'non-cash' benefits provided to employees as part of a salary package.

Earnings statistics are one of four main areas of interest in measuring employee remuneration. The Earnings chapter of Labour Statistics: Concepts, Sources and Methods has more information on earnings and employee remuneration related concepts and how we produce the data.

Each measure is designed for different purposes and has its own strengths and weaknesses - they are not simply different ways of measuring the same thing. 

Use the guides linked below to learn more about the ways we measure earnings and when to use each measure.

• Earnings levels guide
• Average earnings guide
• Wages change and growth guide
• Weekly and hourly earnings guide
• Aggregate earnings guide

This table summarises the most relevant data sources for each earnings measure. We produce additional data sources which also include earnings statistics.

Use this section to find earnings data you're interested in by topic. Clicking on the topic heading will provide more details.

This table summarises the most relevant earnings data sources by topic. Preferred sources change depending on who is providing the information - household sources are generally preferred for person characteristics with business sources preferred for job and employer characteristics. However, business sources also generally provide more accurately reported earnings as data are obtained from employers' payrolls. The quality of earnings data has been prioritised when assigning ratings in the table below. For more information on the strengths and limitations of different sources, please see the Earnings chapter of Labour Statistics: Concepts, Sources and Methods.

Some of these data sources have extra topics available through their TableBuilder or microdata products. We produce additional data sources which also include earnings statistics on these topics.

This section summarises the earnings data available according to their key features. It also lists other information which may help you to understand earnings data.

We produce many data sources measuring earnings and employee remuneration related concepts. The most relevant data sources are included below.

• Earnings and working conditions statistics
• Labour Statistics: Concepts, Sources and Methods - Earnings
• Understanding earnings in Australia using ABS statistics
• ABS data sources for earnings, employment income and total personal income

Earnings level estimates measure the value of earnings in dollar terms. The Earnings guide includes information about other earnings measures. 

Average Weekly Earnings, Australia (AWE) is our headline measure of average earnings. 

We have produced average weekly earnings estimates since 1941. The current survey commenced in 1983 and data has been released every six months since 2012, and quarterly before that. You can learn more about AWE over time on the Average Weekly Earnings methodology page.

AWE provides estimates by sex, state/territory, and public/private sector for three earnings series:

• Full-time adult ordinary time earnings
• Full-time adult total earnings
• Total earnings.

We also produce a cash earnings series, which is the most comprehensive measure of average earnings in Australia.

You can use AWE to analyse earnings levels over time.

Other sources of earnings level estimates provide extra detail beyond what is available from AWE.

Employee Earnings and Hours

• Composition and distribution of weekly and hourly earnings, hours paid for and pay-setting methods.
• Includes data by full-time/part-time status, industry, sex, state/territory, public/private sector, age, casual status and occupation.
• Employee Earnings and Hours is sourced from the two-yearly Survey of Employee Earnings and Hours.

Employee earnings

• Includes earnings data by sex, state/territory, occupation, industry and education qualifications.
• Employee earnings is sourced from the annual Characteristics of Employment Labour Force supplementary survey.

Personal Income in Australia

• Personal income estimates (including employee earnings) by age, sex, state/territory, and over 2,200 regions.
• Personal Income in Australia is sourced from the Linked Employer-Employee Dataset (LEED), which is based on tax data.

See the Earnings page of the Labour Statistics Concepts, Sources and Methods for more information on earnings concepts and measures.

Average earnings can be represented as either a median or mean value.

While the mean is the most commonly understood measure of average, median measures are most representative of an "average" employee’s earnings as earnings data has a positively skewed distribution. The mean is higher than the median value because of a small number of people with very high earnings.

The Earnings guide includes information about other earnings measures. 

• Most representative measure of the "average" level of earnings.
• Provides the "middle" earnings figure, where half of people earn more than the median earnings value and half earn less than the median earnings value.
• Employee Earnings and Hours, Australia includes median earnings by age, method of setting pay, employment status, occupation and industry.
• Employee earnings includes median earnings by education qualification, main and secondary jobs, and working arrangements.

• Personal Income in Australia (and Jobs in Australia) includes median income estimates for small geographic areas.

• The arithmetic average, calculated by dividing total earnings by the total number of people (or employees).
• Often just referred to as average earnings, however they do not represent the earnings of the "average" person.
• Average Weekly Earnings, Australia includes mean weekly earnings by industry, state/territory and public/private sector.
• Employee Earnings and Hours, Australia includes average hourly total earnings for non-managerial employees in addition to a range of weekly estimates.
  • Managerial employees (more likely to have high earnings) are excluded so these EEH estimates are closer to median measures than estimates from AWE.
• Employee earnings includes mean earnings by education qualification, main and secondary jobs, and working arrangements.
• Personal Income in Australia includes average income estimates for small geographic areas.

• Give more detail than average measures by providing information on the range of earnings of the population.
• Refer to the level of earnings at which a certain percentage of people earn more or less than that value.
• For example, 50% of people earn more and 50% earn less than the median (midpoint) value, 75% earn more than the 1st (of 4) quartile and 25% earn less, 10% earn more than the 9th (of ten) decile and 90% earn less.
• Employee Earnings and Hours, Australia provides robust estimates by full-time/part-time status, method of setting pay, occupation and industry. 
• Employee earnings provides earnings distribution data by education qualification, working arrangements and main job.
• Personal Income in Australia provides income distributions for small geographic areas.

Source: Microdata and TableBuilder: Employee Earnings and Hours, Australia, May 2021

Wages change (or growth) over time can be measured in a few different ways:

• wage inflation - changes in the price of wages and salaries unaffected by compositional factors
• change in the earning levels of employees 
• change in total wages paid in the economy.

We have earnings estimates available to measure wages change in each of these ways. The right data source for you will depend on the purpose of your analysis.

The Earnings guide includes information about other earnings measures. 

The quarterly Wage Price Index (WPI) measures change in the price of wages and salaries (similar to earnings) in the Australian labour market over time. In a similar way to the Consumer Price Index (CPI), it follows changes in the hourly rate paid to a fixed group (or “basket”) of jobs.

The WPI measures pure price change by removing the effect of compositional factors including the quality or quantity of work performed or the composition of the workforce. This separates it from other ABS earnings measures.

WPI movements do not reflect changes in:

• tasks or responsibilities
• hours worked
• job holder (for example junior rate, completion of trade certificate)
• location where work is performed.

Use the WPI if your focus is wage inflation or inflationary pressures associated with wages and salaries. Estimates are available by industry, state/territory and public/private sector.

Average Weekly Earnings (AWE) measures average gross (pre-tax) weekly earnings paid to employees and is collected every six months. Movements in average weekly earnings levels provide an alternative view of wages growth to the WPI by reflecting real-world changes in both the level of earnings per employee and in the composition of employment.

This can include changes in the:

• proportion of full-time, part-time, casual and junior employees
• distribution of occupations within and across industries
• distribution of employment between industries.

The Average Weekly Earnings, Australia methodology page includes information you will need to draw the correct conclusions about wages growth from this data.

Use AWE if your focus is on understanding changes in the level of average earnings over time, which reflects the influence of real-world changes in the composition of the labour market.

The total wages index measures changes in the volume of wages and salaries paid each week. It is part of the Weekly Payroll Jobs and Wages in Australia estimates, which are released monthly. These estimates allow us to see week-to-week changes in the labour market for the first time.

Total wages index movements are more affected by compositional changes than other earnings measures, as they reflect changes in aggregate wages and salaries paid rather than average earnings per job or employee. Changes in the number of jobs being worked affect the total wages index. It is also affected by:

• changes in hours worked
• cyclical payments including bonuses, commissions or lump sum payment of leave loading
• payment of penalty rates for public holidays
• irregular payments including overtime, ad hoc or one off payments.

Use the total wages index if your focus is short term changes and understanding the effects on total (rather than job or employee) earnings.

Other total earnings measures

The Labour Account Payment quadrant measures costs to employers in employing labour and the income received by people for providing their labour. The Aggregate earnings guide includes more information about total earnings measures.

Earnings measures can be derived for many time periods. Weekly and hourly measures are common across our various earnings sources. Measures with shorter time periods provide a common period for comparison of earnings levels, whereas measures with longer time periods are more affected by compositional or structural factors.

The Earnings guide includes information about other earnings measures. 

• Provide a common period for comparing earnings, removing the effect of differences in total hours worked.
• For example, we can compare how much full-time and part-time workers receive for the same amount of time worked (that is, for each hour they worked).
• Employee Earnings and Hours, Australia (EEH) is the best source for hourly earnings data.
• EEH includes data for non-managerial employees by sex, occupation, industry, state/territory, public/private sector, and employer size.
• Employee earnings includes hourly earnings data by education qualification, main and secondary jobs, and working arrangements every year.

• Influenced by changes in the overall composition of the workforce over time.
• Affected by changes in hours worked and work patterns, as well as changes in the level of earnings of employees.
• Average Weekly Earnings, Australia includes weekly earnings estimates by industry, state/territory, and public/private sector every six months
• Employee Earnings and Hours, Australia provides estimates by age, sex, occupation and employer size every two years.
• Employee earnings includes estimates by education qualification, main and secondary jobs, and working arrangements every year.

Aggregate (total) earnings and labour income measures provide an economy wide view of the cost of labour. They measure total costs paid by employers and incomes received by people in return for working.

The Earnings guide includes information about other earnings measures.

• Measures total wages and salaries and employers' social contributions (such as superannuation) paid by employers to employees.
• Australian National Accounts, National Income, Expenditure and Product provides state/territory, public/private sector and industry estimates every quarter.
• Australian National Accounts: State Account provides estimates for each state/territory by industry each year.
• Labour Account, Australia provides quarterly industry estimates and annual industry sub-division estimates.

• Measures the income received by people for working.
• Total labour income includes compensation of employees (above) and labour income from self-employment.
• Labour Account, Australia provides quarterly industry estimates and annual industry sub-division estimates.

• Business Indicators, Australia provides quarterly estimates of employee wages and salaries for the private sector by industry.
• Employment and Earnings, Public Sector, Australia provides annual estimates of cash wages and salaries for the public sector by level of government, state/territory, and industry.
• Personal Income in Australia (and Jobs in Australia) provides total personal income estimates for more than 2,200 geographic areas.
• Labour Costs, Australia provides estimates of main labour costs paid by employers, such as employee earnings and payroll tax.

The gender pay gap is the difference in the earnings of men and women, expressed as a proportion of men's earnings.

There are many approaches to measuring the gender pay gap, and many factors that influence it, so no single measure can provide a complete picture. Instead, a range of measures should be considered together to understand the comparative earnings of men and women.

This guide will help you to learn more about the gender pay gap and the key ABS indicators, as well as the factors influencing the gap. Our Earnings guide also includes information about our various earnings measures and how to use them.

The gender pay gap describes the difference between the "average earnings" of men and women. It is not a measure of gender pay equality or equal pay - these are concepts that reflect the extent to which men and women are paid the same for performing the same or comparable work. Unequal pay is only one factor which may influence the gender pay gap.

Gender pay gap measures reflect the various social and economic factors affecting earnings and earning capacity of men and women (e.g. paid hours worked, occupation, industry, pay-setting methods, educational attainment, working arrangements, discrimination, and many more factors). There are other labour market measures where a gender gap exists including participation in paid work and hours worked.

There are four key ABS indicators derived from the two-yearly Employee Earnings and Hours (EEH) survey and the six-monthly Average Weekly Earnings (AWE) survey we use as a starting point for analysis of the gender pay gap:

1.  Median hourly cash earnings (EEH)
2.  Mean hourly cash earnings (EEH)
3.  Median weekly cash earnings (EEH)
4.  Mean weekly cash earnings (AWE)

These indicators, available on our Gender indicators page, provide a high-level snapshot of the gender pay gap. This is consistent with the approach used by the International Labour Organization (ILO) in their Global Wage Report 2018/19: What lies behind the gender pay gaps. Finer population (e.g. full-time employees) can be used for more in-depth analysis.

In addition, the following two ABS indicators, based on the ordinary time earnings of full-time adult employees are also presented on Gender indicators:

5.  Mean weekly ordinary time earnings of full-time adult employees (AWE) - the most commonly cited measure of the gender pay gap
6.  Mean weekly ordinary time cash earnings of full-time adult employees (AWE) - the 'cash earnings' equivalent of the commonly cited measure (which includes amounts salary sacrificed)

Mean weekly ordinary time earnings of full-time adult employees has historically been the most cited measure and is available on a long-term comparable basis. However, it is important to note that unlike the first four indicators above, this measure excludes amounts salary sacrificed, which was first collected in 2006.

We also include an equivalent of this commonly cited measure that includes salary sacrifice ('cash earnings'). The cash earnings series is the most comprehensive measure of earnings and is consistent with our latest underlying earnings concepts. Prior to 2006, salary sacrifice was excluded from our earnings concepts. Following a review, we implemented changes to our earnings conceptual framework to include amounts salary sacrificed. Estimates that exclude salary sacrifice are still produced in AWE to provide an uninterrupted historical time series (see Information Paper: Changes to ABS measures of employee remuneration), alongside estimates that include salary sacrifice.

Cash earnings series generally produce smaller gender pay gaps due to the prevalence of salary sacrifice arrangements in female dominated industries, such as Health care and social assistance. Women, on average, have higher salary sacrifice amounts than men.

There are many approaches to measuring the gender pay gap, and many factors that influence it, so no single measure can provide a complete picture. Instead, a range of measures should be considered together to understand the comparative earnings of men and women. Each measure will show a different sized pay gap, reflecting the impact of differences in the distribution of earnings amongst men and women (median v mean earnings) and compositional factors related to hours worked (hourly v weekly earnings).

Our data sources

We recommend a combination of indicators from EEH and AWE - to leverage the greater detail available from EEH data with the greater frequency and timeliness of AWE data.

EEH provides detailed compositional earnings data for men and women every two years, allowing for comparison of weekly and hourly, and mean and median earnings. AWE provides a long time series of mean weekly earnings for men and women. AWE measures are published every six months (three months after the survey reference period) so provide more frequent and timely, but less detailed, indicators of the gender pay gap. For more information on comparing EEH and AWE statistics, see A guide to understanding employee earnings and hours statistics.

The gender pay gap differs between indicators because of differences in the composition and distribution of male and female earnings. This is why we recommend using a range of indicators together, particularly in the early stages of any gender pay gap analysis. Use this section to understand how the choice of data measure influences the size of the gap.

​​​​​​Average earnings can be derived as either a median or mean value.

The median is the most representative measure of average earnings, as it is the midpoint of earnings distributions, where half of people earn more than the median earnings value and half earn less than the median earnings value.

Mean measures are calculated by dividing total earnings by the total number of employees. They are affected by the distribution of earnings of the population. A relatively small number of highly paid employees can skew the mean higher. The Average earnings guide includes more information on these measures and how to use them.

Impact on the gap

Earnings for both men and women have a positively skewed distribution, with approximately three in five employees earning less than the mean. However, a larger proportion of men than women are highly paid.

As a result, the difference between mean and median earnings for men is larger than the difference between mean and median earnings for women. Gender pay gap measures derived using mean earnings data will usually produce a larger gap than measures derived using median earnings data.

Source: Microdata and TableBuilder: Employee Earnings and Hours, May 2023

Earnings measures are generally presented on either a weekly or hourly basis. Weekly measures are more affected by differences in the overall composition of the workforce, hours worked and work patterns. Hourly measures remove the effect of differences in hours worked each week. The Weekly and hourly earnings guide includes more information on these measures and how to use them.

Impact on the gap

Gender pay gap measures derived using weekly (or annual) earnings for men and women reflect that women do less paid work on average than men. As a result, these measures show a larger gap than measures derived using hourly earnings data which provides a common basis for comparison.

Ordinary time earnings and total earnings measures are available.

• Ordinary time earnings include payments for award, standard or agreed hours of work, including allowances; penalty payments; payments by measured result; and regular bonuses or commissions. 
• Total earnings include ordinary time earnings and overtime earnings. Overtime earnings are payments for hours worked in excess of award, standard or agreed hours of work.

The Earnings chapter in the Labour Statistics: Concepts, Sources and Methods has more information on earnings and employee remuneration related concepts and how we produce the data.

Impact on the gap

Gender pay gaps derived from total earnings rather than ordinary time earnings provide a measure that reflects all the earnings received by men and women. Men, on average, are more likely to work overtime and have higher overtime earnings.

In addition to all employee measures, full-time and part-time status is widely used to categorise people or jobs in terms of the number of hours worked. In our business surveys, we classify employee jobs as full-time or part-time based on whether the person has been engaged by the employer on a full-time or part-time basis. In AWE, data is produced for full-time adult employees and other employees (i.e. employees who are part-time or paid at junior rates). In EEH, data can be analysed for all full-time employees or for full-time adult employees.

Our household surveys (including the monthly Labour Force Survey) define people as employed part-time if they usually work less than 35 hours per week and actually worked less than 35 hours in the reference week. People are classified as full-time if they usually work 35 hours or more per week, or actually worked 35 hours or more in the reference week (even if they usually work less than 35 hours per week). For more information, please see the Employment arrangements chapter in Labour Statistics: Concepts, Sources and Methods.

Impact on the gap

Many women work part-time so the choice of full-time, part-time or all employee measure will affect the derived gender pay gap when weekly or annual data are used.

Earnings differentials of full-time employees have traditionally been used to provide a more common basis for comparison, however this results in a gender comparison excluding almost half of all working women. Measures that include all employees (regardless of working hours) will show a larger gender pay gap as women work less hours than men, on average. Full-time women also work less hours, on average, than full-time men.

Measures of weekly earnings that are limited to part-time workers show varied results given they include a broad range of hours worked, so measures of hourly earnings are preferred.

Use this table to find earnings data sources which can be used to measure the gender pay gap by topic (for example, additional characteristics).

Household sources are generally preferred for person characteristics with business sources preferred for job and employer characteristics. However, business sources provide more accurately reported earnings as data are obtained from employers' payrolls rather than the recall of employees or their partners. The quality of earnings data has been prioritised when assigning ratings in the table below. For more information on the strengths and limitations of different sources, please see the Earnings chapter of Labour Statistics: Concepts, Sources and Methods.

This section summarises available data which can be used to measure the gender pay gap. It also includes other information which may help you to understand gender pay gap measures and factors influencing the gap.

We produce many earnings data sources which can be used to measure the gender pay gap. The most relevant data sources are included below.

We produce many data sources which provide information on labour market outcomes of men and women beyond earnings measures. The most relevant data sources are included below.

Our Gender indicators page includes a range of economic and social indicators for men and women.

ABS resources

• Gender indicators
• Labour Statistics: Concepts, Sources and Methods
• Earnings guide - Guide for ABS labour statistics

WGEA resources

These resources provide additional insight into factors influencing the gender pay gap.

• What is the Gender Pay Gap - Workplace Gender Equality Agency
• The Gender Pay Gap Calculator
• WGEA Scorecard 2022: The state of gender equality in Australia
• She's Price(d)less - The economics of the gender pay gap
• Wages and Ages: Mapping the Gender Pay Gap by Age - Workplace Gender Equality Agency

ILO resources

• Global Wage Report 2018/19: What lies behind the gender pay gaps - International Labour Organization

We produce a range of industry employment statistics to provide different insights. It can be challenging to choose which data to use. This page will help you understand what industry employment data are available, where to find them and how they are best used.

The Employment and Jobs chapters of Labour Statistics: Concepts Sources and Methods have more information on employment and jobs related concepts and how we produce the data.

The Labour Account measures jobs, employed people, hours, and income in Australia. It uses the best available labour market data to create our most comprehensive source of information. A full list of data sources used is available in the Labour Account chapter of the Labour Statistics: Concepts, Sources and Methods. 

The key data source for Labour Account employment is the Labour Force Survey (LFS), which provides robust estimates of employment and unemployment. It is then combined with unpublished Quarterly Business Indicator Survey (QBIS) data which provides higher quality industry of employment distribution estimates than LFS. This is explored more in 'Assigning Industry' below.

Use the Labour Account for analysis of employment (or employed people) and jobs by industry division and subdivision over time.

Labour Force Survey (LFS) industry data is released earlier than Labour Account industry data (as the Labour Account relies on a range of data sources). While the LFS data may provide some useful early insights, the Labour Account data is our definitive read on industry employment and jobs levels and changes over time. The LFS industry data provides important context for other LFS data, however collecting industry information from a household survey has limitations and these estimates may not provide the best indication of real-world levels or changes.

There are also some key differences in how the Labour Account and LFS industry employment estimates are produced. The Labour Account includes estimates of the number of people and jobs in each industry, whereas LFS only includes estimates of the number of employed people in each industry, based on their main job.

Assigning Industry

The LFS assigns industry according to a person's description of their main job. For example, a sales assistant works at a small store selling stationary at the front of a warehouse. Their customers are members of the public, however their employer makes most of its profits through wholesale contracts supplying stationary to a department store chain. Based on the sales assistant’s description of their job, they would be assigned to the Retail trade industry in the LFS.

The Labour Account uses the economic activity of the employer to determine industry, so the sales assistant’s job will be counted in the Wholesale trade industry. This is also where their contribution to production would be reflected in the National Accounts.

Multiple job holders

An employed person can have more than one job, so there will always be more filled jobs than employed people in Australia. We refer to employed people with more than one job as multiple job holders.

The LFS only collects an employed person’s industry of main job, even if that person is a multiple job holder.  As a result, the LFS industry estimates only reflect the number of people employed in each industry in their main job. For example, Alex works 5 days a week as a barista in a café and has a second job stacking shelves at a supermarket on Tuesday nights. In the LFS industry employment estimates, Alex will be counted as one employed person in the Accommodation and food services industry as their main job is as a barista.

The Labour Account industry employment estimates count everyone working a job in each industry regardless of whether it is their main job or a secondary job. This provides estimates of the total number of people employed in each industry. Alex will be counted as an employed person in both the Accommodation and food services and Retail trade industries, although they will only be counted as one employed person at the Australia level.

For more information, see Labour Account: Person Quadrant in Labour Statistics: Concepts Sources and Methods.

Scope

The Labour Account industry estimates include three groups of workers who are outside of the scope of the LFS. These are child workers (employed children aged 5-14), employed short-term non-residents and permanent defence force personnel.

Reference period

The reference periods of the LFS and Labour Account can also affect comparisons. LFS industry employment data is collected in the mid-month of every quarter (February, May, August and November). The Labour Account people and jobs estimates, including for industry of employment, are as at the end of the last month in the quarter (March, June, September and December). Differences in the reference periods can cause some variation between the published estimates.

Industries with the largest differences in employed people between LFS and Labour Account are shown in the graph below.

The large difference in Administrative and support services is mostly due to the different ways industry is assigned for labour hire workers. In the LFS, people who work for a labour hire firm are likely to incorrectly identify their employer and the activity undertaken at their workplace as the business they are providing labour to (rather than who they are actually employed and paid by - the labour hire firm). In the Labour Account, labour hire workers are assigned to the Administrative and support services industry, as reported by their employer.

Weekly Payroll Jobs and Wages in Australia was first released in April 2020. The estimates use Single Touch Payroll (STP) data from the Australian Tax Office (ATO) to measure changes in jobs paid and reported through STP (payroll jobs) each week. This is the first time we've been able to see week-to-week changes across all industries in the labour market.

Use Weekly Payroll Jobs and Wages in Australia to analyse short term changes in industry employment, and as an early indicator to other labour releases.

Example: Measuring industry employment at year-end

The period before Christmas sees an increase in labour market activity, followed by lower business activities around public and school holidays, and employees taking annual leave over the year-end period. The size of these changes differ between industries.

Retail trade has an increase in payroll jobs through December then drops at the end of the month after pre-Christmas trading. Education and training starts to decrease from the start of school holidays, then increases steadily until school returns in February.

The labour market was also influenced by the Omicron outbreaks and extreme weather events over this period in late 2021 and early 2022. Seasonality spotlight: 2021 year-end includes more information.

Jobs in Australia provides industry of employment data by age, sex, employment size and more than 2,200 regions. The statistics are derived from the Linked Employer Employee Dataset (LEED), which brings together personal income tax and business data. 

Use Jobs in Australia for detailed analysis of employment and jobs, particularly by region. You can create your own customised tables using Microdata and TableBuilder: Jobs in Australia.

Example: Comparing industry employment by region

In Western Australia (and at a national level), the industries with the most jobs are Healthcare and social assistance, Administrative and support services and Retail trade. By comparison, in Karratha the industries with the most jobs are Administrative and support services, Construction and Mining. This reflects the strong influence of mining-related activities in the surrounding Pilbara region on the local labour market.

This table summarises the most relevant industry employment data sources by topic. The quality of industry data has been prioritised when assigning ratings. 

Some of these data sources have extra topics available through their TableBuilder and microdata products. We produce additional data sources which also include industry employment statistics on these topics not included below (for example, estimates from our Labour Force supplementary surveys).

This section summarises the industry employment data available according to their key features. It also lists other information available to help you understand industry employment data. 

We produce many data sources measuring industry employment. The most relevant data sources are included below.

• Employment and unemployment statistics
• Jobs statistics
• Labour Statistics: Concepts, Sources and Methods
• Australian and New Zealand Standard Industrial Classification (ANZSIC)

We produce a range of different jobs and employment statistics at a regional level to provide insights into local labour markets. It can sometimes be challenging to choose which data to use. This page will help you understand what regional labour market data are available, where to find them and how they are best used.

We use the Australian Statistical Geography Standard (ASGS) to divide data across Australia up into estimates for each region.

Our regional labour market data is generally based on a persons place of usual residence, although the Census does have Place of work data, in addition to the standard usual residence-based geographic data.

You can explore some of the data available for different ASGS regions – including on other topics – on the Data by region page.

These are three main ABS sources of regional labour market data:

• Labour Force Survey (LFS) - including both traditional direct survey estimates and new higher quality modelled estimates
• Linked Employer-Employee Database (LEED) – which is based on tax data 
• Census of Population and Housing

The LFS has traditionally been the most used source of contemporary regional labour market data, although the direct survey estimates have some limitations. While this LFS information has been more timely and frequent than the more detailed annual LEED data or five-yearly Census data, the usefulness of regional estimates is limited by the smaller sample counts that contribute to each region (that is the share of the total large national sample in each of the roughly 90 Statistical Area Level 4 regions).

It is for this reason that the Australian Government invested in improved modelled estimates, which provide the best measure of timely and frequent regional employment and unemployment, and short-term changes in local labour markets. 

These new modelled estimates are now being released on a monthly basis. The ABS recommends using these over the direct survey estimates whenever possible. See the 'Modelled v direct estimates' section for more information.

The LEED data, which is published in Jobs in Australia and Personal Income in Australia, while less timely than the LFS estimates, provides more detailed geographic and other information (e.g. detailed Industry information), and useful insights into longer-term structural change in the labour market.

See the Labour Statistics: Concepts, Sources and Methods for a more detailed comparison of the various sources of regional labour market data. 

The LFS is a monthly collection which provides information about the labour market activity of Australia's resident civilian population aged 15 years and over. The LFS primarily provides estimates of employment and unemployment for the whole of Australia, and in each state and territory. Estimates at the lower geographic levels are a secondary output from the survey, made possible – with limitations – by the large and representative sample. 

The LFS provides detailed information on regional labour market estimates by demographic characteristics. We have traditionally published estimates of regional labour market data at the Capital City and Balance of State and Statistical Area Level 4 (SA4), including by sex and age, on a monthly basis.

In addition to these longstanding ‘direct’ survey estimates (that only use the LFS data, without any additional specific methods), we now produce higher quality modelled Labour Force estimates (at the SA4 level) on a monthly basis (in Labour Force, Australia, Detailed). These new modelled estimates are our best source of timely information on employment and unemployment for regions.

Prior to April 2024, we only produced regional LFS estimates directly from the survey responses. As these regional estimates are based on smaller sample sizes in each region, they are of a lower level of statistical quality compared to those produced at the national and state and territory levels. Smaller sample size means sampling variability has a greater impact on the results at a regional level. Over time, fluctuations in the data occur across most of the regional labour force data, particularly in regions with smaller populations. These fluctuations are generally caused by sampling variability rather than changes in underlying labour market conditions, though sometimes they are caused by actual local events.

The modelled regional labour force estimates enhance the direct survey estimates of regional labour market data using the combined power of administrative data and statistical modelling.

We use de-identified administrative data sources to model these regional labour market estimates. The two key data sources used are:

• Single Touch Payroll (STP) data from the Australian Tax Office, and
• JobSeeker and Youth Allowance recipients data from the Department of Social Services (DSS)

These administrative data sources contain granular geographic information and are regularly updated. The model produces more stable and reliable estimates, through leveraging the strong relationship between the administrative data and the survey data. More detail on the methodology has been included below.

The model produces estimates of:

1. employed people 
2. unemployed people
3. people not in the labour force
4. unemployment rate
5. employment to population ratio 
6. participation rate

We currently publish these new modelled estimates two weeks after the LFS headline figures, in Labour Force, Australia, Detailed, but later in 2024 these will be published one week after (at the same time as the direct survey estimates).

We use a Rao-Yu model to produce these improved regional labour force estimates. It uses the relationship between the administrative data and the survey estimates that is observed across SA4s. The model also makes use of the correlation of labour force status within SA4s and over time. A paper describing the Rao-Yu model can be found in the 'Modelled v direct estimates' section below.

Final adjustments are applied so that the modelled SA4 level estimates are additive to the state level survey estimates.

We use the data sourced from the DSS to model time series changes in unemployment, while employment is modelled using STP data, with the DSS payment recipients data also applied where relevant.

We have published two papers on the development of the model now being used – Improving SA4 level estimates using administrative data models and Further refinements to modelled SA4 level Labour force estimates using administrative data.

The modelled estimates are also the best source of data for analysis at the Greater Capital City Statistical Areas level. While estimates for capital cities and the balance of states have not been directly modelled, the aggregated modelled SA4 data are better than the direct survey estimates, particularly for the balance of states which generally have smaller populations and smaller sample sizes.

The modelled estimates have some limitations in the amount of data items and the length of the time series that are currently available. While the ABS recommends using the modelled estimates whenever possible, for analysis focusing on age and sex (or other variables that are also not currently available), or for analysis over a longer period of time (pre-2016 for unemployment analysis or pre-2020 for employment), users should continue to use the direct survey estimates. When using the direct survey estimates, the ABS recommends using (moving) annual averages. 

We will provide further advice on how to use the modelled in conjunction with the direct survey estimates, e.g for regional data by age and sex, or for comparisons over time where there are currently no modelled estimates, in the coming months.

Future improvements

We are currently exploring modelling regional labour force estimates at lower levels, including SA3 and SA2 regions.

We will also produce SA4 modelled estimates by age and sex, with initial estimates expected to be released around June or July 2024.

Jobs in Australia presents information on jobs and employed people sourced from the Linked Employer-Employee Database (LEED). Regional information is available for three levels of statistical areas (SA4, SA3, and SA2) and the Local Government Area (LGA) level.

The LEED uses tax data to create a detailed labour market dataset, including characteristics of employees, jobs and employers. Outputs from the LEED are also available in Personal Income in Australia, which publishes the number of income earners, amounts received, and the distribution of income by three statistical areas (SA4, SA3, and SA2) and Local Government Area (LGA).

You can also create custom tables from the data using TableBuilder: Jobs and Income in Australia.

We recommend using LEED data for detailed or complex analysis of employment in regional labour markets, beyond what is available from the more timely and frequent Labour Force estimates, but more frequently than every five years.

The five-yearly Census provides a rich snapshot of all people living in Australia on Census Night. It is the leading source of information for small population groups and areas. As well as labour force status, the Census also collects information on characteristics of people and households, enabling analyses across a broad range of socioeconomic dimensions.

Census data is available at all levels of geography, and can be accessed a number of ways, including microdata products and Community Profiles.

You can find the full range of available Census products from Find Census data.

We recommend using Census data for detailed regional labour market analysis, where a focus on detailed person or household characteristics is important.

As well as the sources listed above, labour force status at a regional level is included as a variable within a range of some of our other data sources.

• The Labour Force supplementary surveys and labour Multi-Purpose Household Survey topics (Characteristics of Employment; Participation, Job Search and Mobility; Barriers and Incentives to Labour Force Participation; Retirement and Retirement Intentions), provide regional data in relation these topics. These data are available in Microdata and TableBuilder: Characteristics of Employment, Microdata and TableBuilder: Participation, Job Search and Mobility, Microdata and TableBuilder: Barriers and Incentives to Labour Force Participation and Microdata and TableBuilder: Retirement and Retirement Intentions.
• The Education and Work and Qualifications and Work surveys provide data on engagement in work and/or study, current and recent study, qualifications, and transitions to work. Regional data is available at the SA 4 level in Microdata and TableBuilder: Education and Work and Microdata and TableBuilder: Qualifications and work. 
• If you are interested in information on health conditions and risk factors, the Microdata and TableBuilder: National Health Survey includes labour force information.
• If you are interested in information on disability, ageing and carers, the Microdata and TableBuilder: Disability, Ageing and Carers has employment and income data at sub-state geographic levels.
• information on employment outcomes of migrants is available in Microdata and TableBuilder: Permanent migrants in Australia and Microdata and TableBuilder: Characteristics of recent migrants. You can also use the migrant information available from the LEED in TableBuilder: Jobs and Income of Employed Persons.

This table summarises the most relevant regional labour market data sources by topic. The quality of regional data has been prioritised when assigning ratings. 

Some of these data sources have topics available through their TableBuilder and microdata products. 

This section summarises the regional labour market data available according to their key features. It also lists other information which may help you to understand regional labour market data.

We produce many data sources measuring the regional labour market and related concepts. The most relevant data sources are included below, including which of the data source ‘pillars’ of labour statistics they are based on (household survey data, business survey data and/or administrative data).

There are a number of other sources of key regional labour market data produced by other organisations:

• Internet Vacancy Index
  The Internet Vacancy Index (IVI) is a monthly count of online job advertisements compiled by Jobs and Skills Australia. Data are available by occupational groups, skill level groups, state or territory and by regional areas.
• Small Area Labour Markets
  Jobs and Skills Australia produces quarterly Small Area Labour Markets (SALM) estimates of unemployment and the unemployment rate at the Statistical Area Level 2 (SA2) and Local Government Area (LGA) level.
• Nowcast of Employment by Region and Occupation
  Jobs and Skills Australia produce monthly estimates of employment by occupation at the Statistical Area 4 (SA4) level. 
• JobSeeker Payment and Youth Allowance Recipients
  The Department of Social Services produce monthly counts of Job Seeker and Youth Allowance recipients at the Statistical Area 2 (SA2) level.