Australia’s September jobs report was all over the place, and that’s putting it mildly.
Employment fell, particularly among full-time workers which recorded its largest one month drop in over five years. That was partially offset by a surge in part-time employment, leaving it up more than 5% than the levels of a year earlier. Unemployment dropped to the lowest level since early 2013. Male labour force participation cratered to the lowest level on record and total hours worked rose by 4 million hours to 1.66 billion.
And unemployment in Tasmania dropped a whopping 0.7 percentage points to 6.5% in just one month.
While most economic notes received following the data release have a common prevailing theme — don’t pay too much attention to the seasonally adjusted figures — some have been brave enough to put a reason behind the wild swings seen during the month.
Yes, that Census.
Another initiative from the ABS that was in the news for all the wrong reasons.
Here’s what Felicity Emmett, head of Australian economics at the ANZ, had to say on the subject:
We think the volatility in the part-time/full-time split may reflect the impact of temporary census workers. Most census workers were already employed, so the extra hours worked for the census could have temporarily pushed people into the full-time category only to reverse now that these jobs have finished.
Paul Dales, chief Australia and New Zealand economist at Capital Economics, agrees that census workers likely played a part, noting that “the surge in part-time employment is probably partly because some people who worked on the Census notched up full-time hours in August but worked fewer than 35 hours in September”.
According to the ABS, part-time workers are those who normally work less than 35 hours per week and who actually worked less than 35 hours in the week the employment survey was conducted.
If a survey respondent worked more than 35 hours in the week the survey was conducted — even if they would normally work less — they are classified as a full-time worker.
On face value that may explain the divergence between full and part-time employment seen in September, but it certainly doesn’t explain all of the volatility in the seasonally adjusted figures.
Indeed, after Australia’s 2011 Census, full-time employment increased by 14,500 in September 2011, with overall employment increasing by a larger 31,400.
It’s all very messy and unreliable, and something that needs to be fixed given its importance to monetary and fiscal policy.