Unemployment - what's in a number?
Updated
Contrary to what many people think, the ABS unemployment figures are based on a random survey conducted each month. (ABC News: File photo)
Unemployment. It is one of the most watched, and perhaps least understood, economic figures.
For a start, comments frequently come up on ABC News stories saying the Bureau of Statistics figures understate unemployment because not all unemployed people claim benefits from Centrelink.
While it is true the headline unemployment statistics do tend to understate serious underemployment, it is not because the figures are based on Centrelink claims.
Contrary to what many people think, the Australian Bureau of Statistics unemployment figures are based on a random survey of around 29,000 private homes and some non-private residences (such as hotels), conducted each month. The ABS estimates its sample covers around 0.33 per cent of Australia's working-age population.
It is like a mini-census, with the difference being that it only surveys a small fraction of the population. This is a big difference and has many implications.
The first is that you do not need to be claiming Centrelink benefits to be counted as unemployed - the ABS asks people how much they worked the week before the survey, whether they were available and willing to work, and how many hours they wanted to work.
It is from the responses to these questions that the unemployment figures are drawn.
However, that does not mean that the headline unemployment figure tells the full story - you only have to have worked an hour a week to be considered employed.
The ABS has responded to the increasing trend towards part-time employment by regularly including more comprehensive figures that cover underemployment in its reports.
That includes figures showing the number of people working full-time and part-time, released each month; data showing the total number of hours worked each month; and quarterly releases showing the underemployment and under-utilisation rates.
The underemployment rate shows the percentage of people working who would like more hours; while the under-utilisation rate estimates the total percentage of people working less than they would like, combining those underemployed and those totally unemployed.
Another major implication from the ABS employment survey is that it has a margin of error - because it does not survey everyone it would be a statistical miracle if it ever got the number of unemployed people exactly right.
Even if it did, the number would be out of date before it was published.
That is why it is important to look at the trend, and anticipate that there will be some unexpected dips and rises in the seasonally adjusted figures that most news organisations use in their headlines, and many solely rely on.
Work hours rising
All these factors are applicable in the February employment data.
In the previews before the data, the January unemployment rate was quoted, correctly, as 5.3 per cent. But the ABS found some statistical errors it could remove, or refinements it could make, during the past month and revised January's jobless rate to 5.2 per cent.
That meant today's rate of 5.3 per cent became a 'rise' in unemployment, when most economists were anticipating the rate would remain 'steady' at 5.3 per cent.
The creation of 400 jobs leads to a rise in unemployment because the population grew at a faster rate than job creation, although participation in the workforce dipped slightly.
However, the rise in unemployment covered by the headlines tends to mask an important trend: there were 11,400 extra full-time jobs recorded in February's survey, offset by a fall in 11,000 part-time jobs.
That is hinting at the early stages of a trend towards employers who cut some of their employees' hours putting their staff back onto full-time hours.
There is further evidence of the emergence of this trend in the sharp 2.4 per cent rise in total hours worked last month.
Meanwhile, the longer-term trend figure for underemployed people has eased slightly from 7.8 per cent in August 2009 to 7.7 per cent in February.
That, combined with the decline in unemployment from a peak of 5.8 per cent to 5.3 per cent, has resulted in the under-utilisation rate falling from 13.6 per cent in August last year to 13 per cent in February.
So, despite the headlines of 'rising unemployment', if anything the February data suggests a positive trend towards people getting back the hours at work they lost during the downturn.
First posted
Comments
Comments for this story are no longer available. ABC policy is to delete comments on stories three months after they are published.
Search ABC News
News Just In
- Carpark fire forces evacuation
-
Rudd 'recovering well' after operation
- Storm front batters homes, cuts power
- Fires kill dozens in Russia heatwave
-
Abbott heads north for defence announcement
- US economic growth slows to sluggish pace
-
Wikileaks urged to stop publishing Afghan leaks
- Pakistan floods kill more than 400
- Charges laid over fatal World Cup bombings
-
Black Saturday report to tackle policy changes
-
Coalition hits lead in latest poll
- Police probe shots fired at restaurant, shop
ABC Analysis
- iPhone 4 arrives in Australia
- Agendas and bias on the media trail
- Campaign booms shake the party room
- Getting less than the best treatment
- Labor's game of political Cluedo
- The hangover after Gillard's Labor party
- Pyongyang's pissing match
- Capitalising on the TV guide
- Wanted: election 2010 boredom buster
- Judgement Day: Black Saturday report release looms
- TV the winner in 2010
- Aker can't hack getting the sack
