How to read the monthly jobs figures

Where the numbers come from

Each month the Australian Bureau of Statistics surveys tens of thousands of households about whether people worked, and for how long, in a reference week. From that sample it estimates employment and unemployment for the whole country. The release usually comes out in the middle of the following month at 11:30am Sydney time; the calendar shows the next date.

The four figures that matter

Unemployment rate. The share of people in the labour force (working or looking for work) who do not have a job. It is the figure the Reserve Bank and markets react to first.

Jobs added. The change in the number of people employed. Australia’s equivalent of the US payrolls number, but from a household survey rather than employers’ records.

Full-time and part-time. A month with strong total jobs growth but falling full-time work is weaker than it looks. FinVolio shows full-time jobs separately for that reason.

Participation and the employment ratio. If people give up looking for work, the unemployment rate can fall without anyone finding a job. The employment-to-population ratio avoids that trap.

Why one month can mislead

Because the figures are estimated from a sample, monthly changes carry a margin of error, and changes of tens of thousands of jobs can be within it. Parts of the sample are replaced each month, which adds noise of its own. The ABS publishes a trend series for this reason, and economists look at the average over three months or more before concluding that the job market has turned.

What the Reserve Bank takes from it

A tight job market supports spending and wages, which can keep inflation up; a loosening one does the opposite. The Bank watches unemployment, hours worked and wage growth together. A rise in unemployment of several tenths of a point over a few months has historically been an early sign of a slowing economy.