One month ago, when Austrlia released job numbers that too ridiculous to believe - they were a 6 sigma beat - the local mainstream media - very much unlike that of the US - threw up all over the numbers [5].
This is how Australia's primary media outlet, ABC, slammed the number with the following excoriating reaction of the economic propaganda, [6]which would certainly have been censored in the US.
A fall in the unemployment rate from 6.2 per cent to 5.9 per cent in the space of one month? Be sceptical. Treat with extreme caution.
Nearly 60,000 jobs created in Australia last month? A fall in the jobless rate in Victoria of 0.7 percentage points? Predictably, the numbers moved markets. Yet the seasonally adjusted labour force estimates from the Australian Bureau of Statistics for October sound incredible and they should be treated as just that: not credible.
There were 44,640 minutes in October and, if you accept the estimate, 58,600 jobs were created. As the Age's economics editor Peter Martin points out, count only normal business hours, and that equates to 5.5 workers being hired each minute of the working day.
Even the economist known as "Mr Sunshine", Craig James of Commsec, is not buying it. "It's hard to believe that almost 60,000 jobs were created in one month with 40,000 of these jobs in full-time positions," was his response.
The ABS is itself cautions against placing too much credence on the monthly figures, which are based on a changing sample, particularly the seasonally adjusted data. The statistician encourages people to focus on the trend estimate (which had the unemployment rate unchanged).
And, after a series of stuff ups, revisions and methodological changes over the past year, there is even more room for caution.
Last year, the ABS was forced to abandon seasonally adjusted labour force numbers for a period after conceding they were unreliable. The former chief statistician recently said the data was not worth the paper it was written on.
And then, two days ago, when everyone was expecting Australian "jobs" to come back to earth, Australia comedian statisticians decided to shock and awe everyone by doubling down on grotesque economic "data" when the reported that in November, when consensus expected a drop of 10,000 jobs, the Australian economy "added" 71,400 jobs, the most in 15 years (this coming at a time when Australia's biggest customer is going through a hard landing), the US equivalent of 1 million jobs in one month.
The number was so ludicrous, it was not just a 6 sigma outlier: it was an 8 sigma!
[7]
While in the US this epic bullshit would have been paraded by the propaganda outlets left and right, in Australia once again, the media decided to call out the jobs "data" for what it is was: utter garbage.
And while last month it was ABC who dared to point out the goalseeked emperor is naked, this time it was the turn of the SMH [8]. Here is what it said.
No one can accuse Employment Minister Michaelia Cash of failing to take advantage of an opportunity.
Presented with official employment figures that were literally incredible (impossible to believe), she took them at face value and trumpeted them in tweets and in a press conference as a triumph for her government.
There were 71,375 more Australians in jobs last month and 301,725 more in jobs over the year, and since January the unemployment rate has tumbled from 6.4 to 5.8 per cent. "It equates to around 2500 jobs per day," she told the press conference. "Over the calendar year, it's the highest jobs growth since 1989."
She could have added that it equates to one new job every 12 seconds, which it does if the jobs were only offered in working hours. It's even more than the 56,100 new jobs created in October, and way above the long-term average of 15,000 per month.
But it's not the real thing. The Bureau of Statistics makes that pretty clear. It's had a run of misfortune with its sample rotations.
The employment survey is massive - 26,000 houses and flats each month, representing about 50,000 people. But they are not always the same people each month. Every survey one-eighth of the survey is "rotated out" and a new one-eighth is "'rotated in". It stays for eight months.
Occasionally, the one-eighth that is rotated in is very different from the one-eighth that is rotated out, making the number of people in work appear to either jump or fall, even if the employment status of everyone in the survey hasn't changed.
It happened in October, and the bureau was upfront about it. It said then that only half of the reported jump in employment would have been seen if it had examined just the participants present in both surveys.
Now it has happened again, for the second month in a row. The bureau says that if it had restricted itself to comparing like with like, employment would have climbed a mere 5300 in November, a fraction of what was reported.
The dollar soared more than three-quarters of a cent on the news, climbing back up above 73 US cents. The traders weren't waiting to find out the truth, which is probably that employment is growing, but more slowly than we are being told.
The minister was asked whether she had concerns about the reliability of the figures. She said she did not.
So congratulations to Australia: at least you still have a media that thinks critically occasionally, and dares to question the bullshit it is being spoonfed. We can assure you that in the US such critical thinking will get your branded as a "conspiracy theorist" and almost certainly added to every terrorist watch in existence.
