- Telephone
- Email
Over the past couple of days there has been some discussion regarding the ‘true’ picture when it comes to employment in Townsville. The discussion started on LinkedIn and has subsequently made it into the pages of the Townsville Bulletin (see story here).
The essence is that over the course of the past 6 months the ABS (after some gentle prodding from us and others) have started to produce a new set of regional labour force statistics which address many of the issues we have had to contend with from the original Labour Force Survey data. When narrowed down to small regional areas the sample size of the survey results from the LFS are so small that they exhibit extreme volatility. The traditional way that this has been handled has been to simply average the monthly data over a 12-month period, and that is the series the Queensland Govt Statisticians Office uses. As a result, their ‘official’ data shows Townsville having lost 6,400 employed people in the past 12 months.
However, the problem with this approach is that reliance on a comparison of two simple 12-month averages essentially attaches as much significance to the original monthly data from almost 2 years ago as it does to the data from this month! It therefore provides a very lagged response to actual movements in the labour force.
For many years we have been working to rectify this problem by producing the Conus/CBC Staff Selection Trend series, which estimates the seasonally-adjusted trend based upon the original LFS data. More recently the ABS made a large step forward by producing modelled ‘Enhanced Regional Labour Force’ data which uses the original LFS survey data but supplements it with administrative data such as JobSeekers and One-Touch-Payroll data. The ABS are currently advising that “The ABS recommends using the modelled estimates in {the Enhanced data} over the direct survey estimates whenever possible in reporting on employment and unemployment rates for regional labour markets.” The QGSO are opting at this stage to ignore that recommendation.
Although at the moment this data is only available at the higher level, and therefore does not include data split by sex, age or full-time and part-time distinctions we welcome it as an extremely positive step in the provision of better regional data. The new Enhanced Data is available back to 2020 (for employment) and 2016 (for unemployment) and since it was first released, we have been using this data to estimate our higher-level Conus/CBC Staff Selection Trend series. We are then using the original LFS for age, and full/part-time data and adjusting this for consistency with the estimated Trend series from the Enhanced series.
The impact of this change can be clearly seen in the graph below.
The original LFS data, and the 12-month average derived from it, showed a dramatic ramp up in employment through 2023; that increase has now turned negative and as a result we see the large decline noted in the ‘official’ data. The ‘Enhanced’ series, and the Conus/CBC Staff Selection Trend derived from it, shows no such ramp up and consequently no sharp decline now. Indeed, the Trend suggests an increase in employment of 4,500 rather than a decline of 6,400 (even though their actual levels are only about 2,000 different).
So, what’s the ‘true’ picture? It’s important to say that none of these interpretations are necessarily ‘right’ or ‘wrong’. They are merely using available data to try and paint a picture of the labour force situation in Townsville (or any other regional area) that is as realistic and useful as possible. Obviously, we believe that our Trend analysis based on the ABS Enhanced data provides a far less volatile and realistic version than a very lagged 12-month average approach. The story it tells is not as ‘exciting’ or ‘newsworthy’ but it’s worth pointing out that those highlighting ‘large job losses’ in Townsville now would presumably have been highlighting ‘massive jobs gains’ a year ago??
We should also note the ABS in their response to questions posed by the Townsville Bulletin say…
“The new modelled estimates provide the best indication of change in the regional labour markets. These estimates also align more closely with the seasonally adjusted and trend estimates that local economists are also producing, such as Conus Consultancy in Far North Queensland.”
The full Conus/CBC Staff Selection Trend estimates for all QLD, NSW and Victoria regions are available on our website. Download them here.
We are proud to have CBC Staff Selection as supporters of the Conus/CBC Staff Selection Trend Employment and Jobseeker data series