If I can play a small role in pushing this debate forward so that the Government does start to take more responsibility for job creation and less for plasma TV production then I will feel better.<\/p>\n
A Financial Review journalist asked me today about the participation rates. There is a view out there that the fact participation rates have not fallen since the downturn began is a sign that things are not as bad this time around.<\/p>\n
Since the downturn started (in labour market terms the low-point unemployment rate of 3.9 per cent was in February 2008), labour force participation has risen from 65.4 to 65.5 per cent (persons). For males, it has dropped slightly from 72.6 to 72.2 per cent which for females it has risen from 58.4 to 58.9 per cent. The journalist was advancing the idea that this has been a gender-biased recession to date – against males. The data suggests that so far.<\/p>\n
To put this in perspective, in December 1989 the low-point unemployment rate for that cycle was 5.6 and things became grim after that. By the time the unemployment rate had risen by 1.8 per cent from this low-point (so a similar situation as now) participation had also risen.<\/p>\n
Over the entire recession (December 1989 to December 1992 – where unemployment peaked at 10.9 per cent). male participation fell from 75.2 to 73.6 per cent; female participation remained constant at 51.7 per cent; and total participation fell from 63.3. per cent to 62.5 per cent. Clearly, the participation rates are higher now overall which reflects the length of the growth period we have just exited.<\/p>\n
However, during the 1991 recession (which dragged out in labour market terms until December 1992 and the unemployment rate remained above 10 per cent until March 1994 despite GDP growth resuming in 1992), total participation also rose in the first several months of the slowdown and reached 63.8 per cent in April 1991 as GDP was plunging.<\/p>\n
The following graph shows the labour force participation rates indexed to 100 at the start of the 1991 and 2009 downturns. The actual base months are December 1989 and February 2008, respectively, which represented the low-point unemployment months in the respective cycle. In 1991, labour force participation actually rose more (blue line) than in the current downturn (red line). At a similar point in the cycle (months after the low-point) you would not say there was very much difference between the two episodes. The deterioration set in after several months (about the point we are approaching in the current cycle).<\/p>\n
Labour force participation rate dynamics – 1991 and 2009<\/strong> So the point is that we should not feel comfortable now looking at participation rates given their behaviour in the last recession.<\/p>\n So far the “added worker” effect is dominating the “discouraged worker” effect. The former suggests that households add hours of work (typically the second breadwinner – female usually) as the economy deteriorates to ensure family income is maintained as the primary breadwinner loses hours or their job. The latter suggests that ultimately people give up actively seeking work as vacancies dry up and they “drop out of the labour force” using the official definitions applied by the national statistician. In the early part of the downturn the former effect dominates.<\/p>\n This is likely to be reinforced by the fact that in the early stages of the downturn full-time work drops sharply and firms replace labour with part-time hours. They also try to hang onto labour by cutting hours rather than sacking the workers. This is clearly going on in the Australian labour market at present.<\/p>\n The stimulus package was more focused on the service sector than on the goods-producing sectors (such as manufacturing). The relative fortunes of these two broad sectors is having a gender effect.<\/p>\n The following table shows the employment changes since the low-point month of February 2008 to May 2009 (taken from the ABS Labour Force Survey). It is clear that to date males are losing full-time work faster than they can get part-time work, while females have enjoyed growth in both segments of the labour market.<\/p>\n Employment changes since February 2008<\/strong> To put this is perspective relative to 1991, the following three graphs compares the evolution of employment for males, females, and all persons from the low-point unemployment rate month (noted above) for each cycle. The blue line is the low to high unemployment rate period for the 1991 recession and the red line is the period since February 2008. The horizontal axis are months since the low-point.<\/p>\n You can see that males are following the same downward trajectory as in 1991 although the loss of employment is not yet as severe. Full-time losses were more severe in 1991.<\/p>\n Male employment dynamics – 1991 and 2009<\/strong> The next graph is for females. It shows that at a similar stage in the cycle (measured in months since the low-point which may not be a similar stage defined in other terms), females are faring better this time. In part, this is due to the strange behaviour of female full-time employment this time. In 1991, this fell away sharply along with male full-time work. The current behaviour needs further research.<\/p>\n Female employment dynamics – 1991 and 2009<\/strong> The final graph shows the same dynamics for persons (total employment). It reinforces the fact that something happened around the 9th or 10th month into this downturn that has held employment up relative to the 1991 downturn. The early stages of each downturn saw employers shedding full-time work overall but more than replacing it with part-time employment. It wasn’t until the demand constraints were considered to be worse than first considered that employers starting shedding labour overall in 1991.<\/p>\n But I suspect the stimulus packages which came early into the downturn and were fairly substantial have extended the period in which firms are prepared to carry workers instead of firing them. This is referred to as labour hoarding. It was largely absent in the 1991 recession but then the Government failed to react until it was too late – remember the “soft landing” rubbish that was fed to us then as an excuse for the Government to do nothing.<\/p>\n Total employment dynamics – 1991 and 2009<\/strong> Anyway, with all that in mind, today’s labour market data indicates that the labour market is deteriorating further. While so far employers might have been prepared to adjust hours instead of persons in meeting the waning aggregate demand, the ANZ Job Advertisements Series shows that vacancies (evidenced by job advertisements in newpapers and on the Internet) are falling away sharply.<\/p>\n The total annual decline to June 2009 has been 51.4 per cent with the advertisements declining by 6.7 per cent in the last month alone. It is also the 14th consecutive month that the series has fallen. The following graph shows the ANZ series and the extent of the recent plunge.<\/p>\n ANZ Job Advertisement Series, July 1999 to June 2009<\/strong> The next graph shows the very strong (scatter) relationship between the ANZ series (horizontal axis) and the official unemployment rate (vertical axis) since 1999. So when the job ads fall we would expect the unemployment rate to rise. The simple quadratic regression reported suggests that the unemployment rate will rise marginally in the June (data out soon).<\/p>\n
\n![\"participation_1991_2009_to_May\"](\"https:\/\/billmitchell.org\/blog\/wp-content\/uploads\/2009\/07\/participation_1991_2009_to_May.jpg\" "\\"participation_1991_2009_to_May\\"")
<\/a><\/p>\n
\n<\/a><\/p>\n
\n<\/a><\/p>\n
\n<\/a><\/p>\n
\n<\/a><\/p>\n
\n<\/a><\/p>\n