There was much made of the decrease in the seasonally-adjusted unemployment rate last month; and there was a lot of uninformed commentary on the topic because the outcome defied the expectations of most.

Luckily Australia has a very capable statistics commentator, Possum Comitatus, who writes at Crikey http://www.crikey.com.au/ and managed to dispel a lot of the myths surrounding the result: http://blogs.crikey.com.au/pollytics/2009/06/01/fair-suck-of-the-sav/ and http://blogs.crikey.com.au/pollytics/2009/06/10/a-warning-on-tomorrows-u...

This month’s increase to 5.7 means that there should be less scrambling for explanations and unfair criticism of the ABS methods. However, it should be remembered that the unemployment rate, as always, comes with a 95% confidence interval. This month the ABS is 95% confident that the actual unemployment rate is between 5.2% and 6.0%.

Today’s contributions to the EconArticles’ Chart Pack also look at unemployment. The unemployment rate considered here is the seasonally-adjusted figure smoothed by way of a three-month backward-looking average i.e. the figure that is reported for August 2005 is the mean of June 2005, July 2005 and August 2005. The reason for looking at the three-month average is that unemployment tends to be quite volatile.

There are a few points worth noting. Focussing on the past three years it can be seen that unemployment has been increasing since the early months of 2008. The increase from about 4.0 to about 4.5 during 2008 meant that about 50 000 people became classified as unemployed during that year. The increase since then from about 4.5 to about 5.7 means that an additional 135 000 people have become classified as unemployed so far this year. Note that these figures treat a full-time job the same as a part-time job and so those employees whose hours have decreased are not represented in these figures.

From the broader sample, 1978 to 2009, changes in the difference between the male and female unemployment rate can be observed. Notice the points where female and male unemployment rates cross. The switch from higher female unemployment to higher male unemployment occurred during the 1990 recession and it was not until 2003 with its booming conditions that this reversed.

The recent data appears to have a similar tendency with male unemployment rising faster than female unemployment. This is concerning because of the potential for these previously employed males to give up searching for a job – leave the workforce – and become part of the long-term unemployed. The lesson may be that to keep the overall unemployment rate low in times of recession requires concentrating on minimising male unemployment.