What Australian unemployment statistics measure

Australian unemployment statistics come from the Australian Bureau of Statistics (ABS), which surveys around 0.3% of the working-age population each month. The headline figure you see in news reports — currently around 3.5% to 4.5%, though this shifts — counts people who are out of work, actively looking for a job, and available to start within two weeks. It does not count people who have stopped looking, people in part-time work who want full-time hours, or people in training programs.

This matters because the unemployment rate can look stable while the actual experience of work changes. Someone who gave up looking last month disappears from the statistic. Someone who found three days of casual work a week is counted as employed. The ABS publishes these variations separately, but the news usually reports only the headline number.

The data comes from the Labour Force Survey, released on the second Thursday of each month. It covers the week before the 12th of the previous month, so there is always a lag between what happened and what gets reported. State-level breakdowns come out at the same time, though they are based on smaller samples and move around more month to month.

Key Takeaways

  • The ABS unemployment rate counts people without work who are actively looking and available to start within two weeks, not all people without paid work.
  • The monthly Labour Force Survey is the official source, released on the second Thursday of each month, and covers the previous month's data with a two-week lag.
  • Underemployment (people working fewer hours than they want) and labour force participation (the proportion of working-age people in the job market) are published alongside unemployment but often ignored in headlines.
  • State and territory figures come from the same survey but use smaller samples, so they fluctuate more and are less reliable month to month than the national figure.
  • Long-term unemployment (people out of work for 12 months or more) is tracked separately and tends to rise during recessions, affecting older workers and people with less education more severely.

Where to find the official data

The ABS website (abs.gov.au) publishes the Labour Force Survey results as a free read on release day. The main release includes the national unemployment rate, state breakdowns, age and gender splits, and duration of unemployment. You can also access historical data going back decades, which is useful if you want to see how the current rate compares to previous recessions or booms.

The ABS also publishes a more detailed publication called "Labour Force, Australia, Detailed" which includes underemployment figures, reasons for part-time work, and industry breakdowns. This comes out a few weeks after the headline release and is less widely reported but more useful if you are trying to understand what is happening in a specific sector or demographic group.

If you want to track the data without downloading spreadsheets, the ABS website has interactive tables and graphs. Trading Economics and the Reserve Bank of Australia (RBA) also publish the same figures in their own formats, which can be easier to navigate if you are comparing Australian unemployment to other countries or to inflation and interest rate data.

How the unemployment rate differs from JobSeeker numbers

The unemployment rate and the number of people on JobSeeker (the main unemployment payment) are not the same thing. JobSeeker numbers come from the Department of Employment and Workplace Relations and are published monthly, but they count everyone receiving the payment, not everyone without work. Someone can be unemployed by the ABS definition but not on JobSeeker if they have a partner earning above the income threshold, or if they have savings above the asset limit. Someone can be on JobSeeker but not counted as unemployed if they are in a part-time job that pays below the threshold.

JobSeeker numbers tend to lag behind unemployment rate changes because people take time to explore, and because the payment has stricter asset and income tests than the ABS definition. During the COVID-19 pandemic, JobSeeker numbers spiked much higher than the unemployment rate because the asset test was temporarily removed, so people with savings could access the payment. When the asset test returned, the numbers fell even though unemployment had not changed much.

If you are trying to understand the health of the job market, the unemployment rate is the better measure. If you are trying to understand how many people are receiving income support, JobSeeker numbers are what you need. They are measuring different things, and conflating them leads to confusion about both.

Underemployment and labour force participation

The ABS publishes two figures alongside the unemployment rate that often tell a more complete story. Underemployment counts people who are working part-time but want full-time work, or who want more hours than they currently have. This number is usually higher than the unemployment rate — currently around 6% to 7% — and it tends to rise during downturns even when unemployment falls, because employers cut hours before cutting jobs.

Labour force participation is the proportion of working-age people (15 and over) who are either working or actively looking for work. This has been rising in Australia over the past decade, mainly because more women are in the workforce and people are working longer before retirement. But participation can also fall during recessions when people give up looking, or during booms when people leave the workforce to study or care for family. A falling participation rate combined with a stable unemployment rate can mean the job market is actually weakening — fewer people are looking because they have lost hope.

The ABS publishes these figures in the same Labour Force release, but news reports often focus only on the unemployment rate. If you want to understand what is actually happening in the job market, checking underemployment and participation alongside unemployment gives you a much clearer picture.

How unemployment varies by age, gender, and location

The Labour Force Survey breaks down unemployment by age group, gender, and state. Young people (15–24) have unemployment rates roughly double the national average, currently around 7% to 8%, because they are still entering the workforce and changing jobs frequently. People aged 25–64 have lower rates, usually close to the national average. People aged 65 and over are not included in the labour force survey because most are not in the workforce, though the ABS does track employment for this group separately.

Gender differences in unemployment are usually small — men and women have similar headline rates — but the composition differs. Men are more likely to be in industries that cut jobs during downturns (construction, manufacturing), while women are more likely to be in part-time work and underemployment. Long-term unemployment (12 months or more) affects men more severely than women, partly because men are concentrated in cyclical industries.

State and territory unemployment varies significantly. New South Wales and Victoria usually have rates close to the national average because they are large and diverse. South Australia and Tasmania often have higher rates because they have smaller, more specialised economies. Western Australia and Queensland can swing sharply because they are heavily dependent on mining and construction. These state figures come from smaller samples than the national figure, so they move around more from month to month and should not be read as precise.

Long-term unemployment and what it reveals

The ABS tracks how long people have been unemployed — less than 4 weeks, 4 to 12 weeks, 3 to 6 months, 6 to 12 months, and 12 months or more. Long-term unemployment (12 months or longer) is a key indicator of labour market health because it shows whether people are cycling through jobs or getting stuck. During recessions, long-term unemployment rises sharply and takes years to fall back, even after the headline unemployment rate has recovered.

Long-term unemployment is not evenly distributed. It affects older workers (55+) much more severely than younger workers — people over 55 who lose a job often take two to three times longer to find another one. It also affects people with lower education levels and people in regions dependent on a single industry. During the 2020 COVID recession, long-term unemployment rose quickly but fell faster than in previous recessions, partly because the government extended JobSeeker payments and partly because the recovery was faster.

If you are looking at unemployment statistics and want to know whether the job market is genuinely improving or just cycling people through short-term work, the long-term unemployment figure is where to look. A falling headline rate combined with rising long-term unemployment means people are finding work but not keeping it, which is a sign of a weak market.

How to interpret month-to-month changes

The Labour Force Survey is based on a sample, not a census, so there is always random variation from month to month. The ABS publishes a measure called the "standard error" which tells you the range within which the true figure probably lies. For the national unemployment rate, the standard error is usually around 0.1 to 0.2 percentage points, which means a change of less than 0.2 points could easily be noise rather than a real shift.

This is why economists and the Reserve Bank usually look at three-month or six-month trends rather than single monthly changes. A single month where unemployment rises 0.3 points might be real, or it might be sampling variation. Three months of rises in a row is almost certainly real. The ABS publishes trend data (a smoothed version that removes monthly noise) alongside the raw figures, and this is often more useful for spotting genuine shifts.

State-level figures have larger standard errors because they are based on smaller samples, so they bounce around more. A state unemployment rate that jumps 0.5 points in one month might fall back the next month without anything real having changed. This is why state figures are best used to spot long-term patterns rather than to react to single-month movements.

Frequently Asked Questions

Where can I read the full Labour Force Survey data?

The ABS publishes the Labour Force Survey as a free read on abs.gov.au, usually on the second Thursday of each month. The main release includes national, state, and demographic breakdowns. The detailed tables (called "Labour Force, Australia, Detailed") come out a few weeks later and include underemployment and industry data. Both are free to read as Excel files or view as interactive tables on the ABS website.

Why is the unemployment rate different from the JobSeeker number?

The unemployment rate counts people without work who are actively looking and available to start within two weeks. JobSeeker counts people receiving the payment, which has different income and asset tests. Someone can be unemployed but not on JobSeeker (if they have a partner earning above the threshold), or on JobSeeker but not unemployed (if they are working part-time). They measure different things and are not directly comparable.

How far back does the Labour Force Survey data go?

The ABS publishes Labour Force data back to 1978 on its website. Data before 1978 exists but is less detailed and uses different definitions. If you are comparing current unemployment to historical recessions, the ABS website has historical tables that show how the current rate compares to the 1990s recession, the 2008 financial crisis, and the 2020 COVID recession.

What does underemployment mean, and why is it important?

Underemployment counts people working part-time who want full-time work, or people who want more hours than they currently have. It is usually higher than the unemployment rate and rises during downturns because employers cut hours before cutting jobs. If unemployment is falling but underemployment is rising, it can mean people are finding work but not enough of it, which suggests a weak job market.

How reliable are state-level unemployment figures?

State figures come from the same Labour Force Survey but use smaller samples, so they have larger standard errors and bounce around more month to month. A single-month change in a state figure could easily be sampling noise. State figures are most useful for spotting long-term trends (three months or more) rather than reacting to single monthly movements. The national figure is much more reliable because it is based on a larger sample.