Officially, employment grew not at all in 2011 after surging 363,500 in 2010. Yesterday in an invitation-only seminar attended by The Age assistant statistician Paul Mahoney said employment probably climbed 30,000 to 35,000 more than officially acknowledged in the first nine months of 2011 and climbed 60,000 to 70,000 less than acknowledged in 2010.
This means official figures overstated the weakness in the labour market that led the Reserve Bank to cut rates at the end of 2011 and overstated the strength that led it to push up rates at the end of 2010.
“We acknowledge we have problems with the way we are benchmarking the labour force at the moment,” Mr Mahoney said. “We are not hiding behind this, we are being very open about it. This is a public seminar, this is going to be repeated a few times.”
The problem arises because in order to convert the results of its door to door employment survey into figures for whole nation the ABS has to estimate the size of the Australian population. Usually it gets the estimate right. But at times when the rate of population growth is changing rapidly it can get it wrong. Instead of revising the official employment figures when more correct population information comes to hand it instead revises its estimate of future population growth. This means incorrect employment figures remain on the public record and future employment growth figures are adjusted in the opposite direction to compensate.
The ABS will attempt to save money by moving its monthly employment survey on line, posting passwords and login codes to the 29,000 households that take part rather than visiting them and following up with phone calls. It will also abolish or make less frequent a number of less-important labour force surveys.
The unemployment rate - presently 5.1 per cent - is unaffected by the Bureau’s problems with its measure of employment.
In today's Canberra Times, Sydney Morning Herald and Age
From the ABS:
UPDATE August 10:
The RBA on the ABS methods in its Quarterly Statement
"When a discrepancy between the official population
estimate and the labour force survey population benchmark
arises, the normal approach for constructing the
population benchmarks removes the difference over
time by forcing the projected population benchmark
to gradually converge toward the expected
official estimate of the population.
[Our] adjusted estimate suggests that employment
growth was around ½ percentage point lower
in 2010 than the published estimate and around
½ percentage point higher over the past year. By
coincidence, this bias has exaggerated the recent
cycle in employment growth."
RBA - How the ABS Labour Force Figures Misrepresent Jobs Growth
. May: Tim Colebatch has cracked it. Why the jobs data is flawed
. January: "For job seekers it was the worst in 20 years"
. November: Standstill. We are no longer making enough jobs