There is one jobs number that our pollies can’t spin
As surely as the full moon brings out ghouls and werewolves, each month’s jobs figures bring the politicians out of their lairs.
And so it was on Thursday, when the headline jobless number from the ABS fell from 6.0 per cent to 5.8 per cent, despite hardly any jobs being created in net terms.
Greens employment spokesman Adam Bandt howled that the falling participation rate – from 65.1 per cent to 64.9 per cent, seasonally adjusted – showed that “there aren’t enough new job opportunities for people looking for work, so they’re giving up”.
• Comparison sites drive up private health costs
• Negative gearing proponents’ claims don’t stack up
• Fairfax staff strike over mass job cuts
“This is especially the case for young people, who have continuously faced an unemployment rate twice as high as everyone else.”
Shadow Employment Minister Brendan O’Connor loped into the open to growl about the “crisis of underemployment in this country” adding that “there are in excess of one million people who would like to work more but cannot find the work and monthly hours worked decreased by two million”.
And in the harsh light of the House of Representatives, a slavering Treasurer Scott Morrison accused Labor of “seeking to deride” the fall in headline unemployment, but warned that Australians would not be “intimidated out of their prosperity”.
Fearsome stuff, but how is one to make sense of the competing spin?
Paint by numbers
The first thing to remember is that the ABS data itself has been highly suspect in the past year, due to a bungled shift in the Labour Force survey methodology.
The Bureau changed the way it sampled the jobs market in 2014, but failed to run a parallel survey using the old methodology to see if they reached the same conclusions.
That prompted former ABS head Bill McLennan to predict last year that the new jobs numbers would not be knocked into reliable shape for three to five years.
With that caveat noted, the other reason both sides of politics can spin the numbers in such different directions is that they tend to focus on some of the most rubbery figures in the first place.
The headline rate is most useful if the ‘trend’ figure is used. That figure is smoothed by a strict mathematical formula to give the most likely indication of where things are going.
The seasonally adjusted figure, by contrast, is more volatile and therefore makes for better politicking and better headlines.
Trust the big number
Probably the most reliable number in the survey, because it is harder to spin for political purposes, is the ‘hours worked’ figure.
The survey estimates the total number of hours worked in the economy each month, and in February that number was down two million hours.
That sounds bad, but not as bad as you might think. While in seasonally-adjusted terms, hours worked “decreased two million hours to 1,652.6 million hours”, on the trend measure they “increased 2.5 million hours to 1,653.6 million hours”.
Confusing, eh?
Nonetheless, what the hours worked number is telling us is that the economy is doing better overall under the Turnbull government than it did under the Abbott government.
In the chart below, I have divided the total hours worked figure by the ABS’s ‘estimated resident population’ figures, to give ‘hours worked per resident’.
This is important, because both side of politics use raw numbers to boast about their job creation, when in reality we need about 175,000 jobs to be created each year just to keep up with population growth. Note, too, that as only about half of Australians work, the ‘quarterly’ hours displayed below would average to about 30 hours a week per worker.
A number of historical moments stand out on the above chart.
First, the nation did a bit of overtime during the dot-com boom, but hours per worker fell when that over-hyped bubble burst.
Next, the long expansion of credit in global economies was mirrored in Australia, and accelerated by phase one of the mining boom, resulting in a more fully employed workforce up to 2008.
That crashed back to earth during the GFC, ramped up again during Labor’s debt-funded period of stimulus spending, and then tumbled again as Australia’s terms-of-trade deflated, helped along by the Abbott government’s confidence-trashing 2014 budget.
Clearly many other factors were at work, but that’s the thumbnail version.
Hours worked had recovered a little before Malcolm Turnbull become Prime Minister in September last year, but the good news is that there is nothing in the latest ABS data to say the hours worked figure is declining. It seems to be holding, if not increasing slightly.
That won’t cheer up young Australians out of work, vulnerably employed workers on casual rates, or older workers sidelined in the years before retirement.
But what it does tell us is that when the political rhetoric is pared away, the economy is not in decline. Well not yet, anyway – I’ll save that spooky story for another time.