Labour Force Survey: The Australian Financial Review (22/1/03, Page 47)


The Editor
The Australian Financial Review

It would be unfortunate if your readers were misled by Barry Hughes' comments on the quality of ABS measures of the labour force(22/1/03, p 47).

After each Population Census, the ABS selects a new sample for the Labour Force Survey, so that it continues to efficiently represent the Australian population. This new sample is introduced gradually, over eight months to minimise any effect on survey estimates. This process is very similar to the usual monthly replacement of about one eighth of the survey sample. The main difference is that the 'new' selections are from a sample design that uses the latest Census information, and that selected households are dispersed slightly from the previous selections. The number of households in the survey for the first time each month remains essentially unchanged, contrary to the suggestions of Barry Hughes. The ABS has no reason to believe that introducing the new sample this way has disrupted the labour force series.

Dr Hughes criticises the ABS for ceasing the "old de-registered banger" of payroll tax based employment measures, twenty years ago. That series had become totally unreliable. It provided no information about employment in businesses below payroll tax thresholds - and the growth in the small business sector meant that measures of change in employment were seriously underestimated. (In 1980 around three-quarters of a million employees, out of a total of five million civilian employment, were excluded because of this). This situation was worsened by state government decisions to change the payroll tax exemption levels and the reporting obligations of business. In 1980, concerns about the quality of the payroll based series were so great that the Australian Statistician suspended publication.

ABS will continue to provide high quality labour force statistics for decision makers.

Susan Linacre
A/g Australian Statistician