Beware the employment data from the Census?

Yesterday saw the release of the second tranche of Census 2016 data part of which included information relating to employment. However, we would suggest that we need to be careful when reading too much into the Census data as it relates to employment.

Employment is one of the few indicators that is measured (or at least estimated on the basis of a labour force survey) on a monthly basis by the ABS at a national, state and regional level. In addition the ABS also produce employment by industry data on a quarterly basis, also at a national, state and regional level. As regular readers will know, we have some concerns around the quality of the regional data and we therefore prefer the Conus Trend series for both the monthly Jobs data and the quarterly Industry Jobs data. Nevertheless, when we consider employment we feel confident that the ABS (and derived Conus) Trend series give us the best available indicator as to actual levels at the national, state and regional levels.

Certainly the ABS agree. To quote from their own information, provided with the Census release yesterday, “The Labour Force Survey produces the most authoritative and recent estimates of labour market information, including employment and unemployment”.

It is therefore worth considering how this data tallies with, or differs from, the Census data. And when we do that we see issues that force us to consider how much importance we should be attaching to the Census data in this area.

Employed
2011
Employed
2016
Change
11-16 %
Variance
ABS-Census %
Census ABS/Conus Census ABS/Conus Census ABS/Conus 2011 2016
Aus 10,058.3 11,222.7 10,683.8 11,941.5 +6.2 +6.4 11.6 11.8
QLD 2,039.3 2,279.7 2,136.5 2,342.4 +4.8 +2.8 11.8 9.6
RoQ 1,024.8 1,156.4 1,059.5 1,163.9 +3.4 +0.6 12.8 9.9
Cairns 102.9 114.1 106.1 109.0 +3.1 -4.5 10.9 2.7

Of particular note here, for our own area, is the fact that the Census suggests an increase of 3.1% in the number employed in Cairns SA4 over the period, while our own Conus Trend (derived from the ABS data) shows a 4.5% decline. Indeed for the Rest of Queensland as a whole the Census has a 3.4% increase while the Trend suggests virtually no change.

The Census also provides detail about industry of employment and here too we see some significant differences when considering the regional data.

  • While both Census and Trend agree that Healthcare was Cairn’s largest employment sector in August 2016, the Census suggests a 15% increase since 2011 while both the ABS annual average (-0.7%) and Conus Trend (-2.0%) have the sector falling slightly.
  • Education & Training has seen rapid growth but the Census has it running at +15.7% since 2011 while the ABS annual average suggests +112.4% and Conus Trend +81.5%
  • Accommodation & Food is an important sector for the region which the Census suggest is up 6.4% while both the ABS and Conus have it down (44% and 33.6% respectively) since 2011.
  • Manufacturing exhibits some wide variances. The Census has the sector falling by 20.1% since 2011 while the ABS annual average is up 2.6% and the Conus Trend up 39.9%. Recent rises in the sector (which are more rapidly reflected in the Trend than the annual average are likely part of the reason for the seeming paradox).
  • Despite well documented increases in Public Administration jobs, the Census has the sector in Cairns virtually unchanged while the Conus Trend estimates an 18.0% rise.
  • Retail Trade has seen employment fall across all measures since 2011 but the Census has just a 8.7% drop, the Conus Trend suggests down 20.7% and the ABS annual average is a thumping decline of 33.4%.

So who do we believe? I would suggest, as the ABS themselves do, that we focus on the ABS Labour Force data (and, where appropriate the derived Conus Trends) when we are discussing employment and industry of employment. The Census comes into its own when considering issues such as hours worked, unpaid work done and other indicators not adequately covered within the standard monthly or quarterly ABS data releases.

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *