RETRENCHED UNEMPLOYED PEOPLE


INTRODUCTION

In February 2010 there were 693,300 unemployed people, of whom 273,000 (or 39.4%) had become involuntarily unemployed in the past two years. These people who 'lost' their jobs due to economic reasons are often contrasted with people who voluntarily 'left' their job.

In ABS statistics, an unemployed person may have lost their job for one of the following reasons:

• being laid off or retrenched from that job;
• their last job was running their own business and the business closed down because of financial difficulties';
• leaving the job because of their own ill-health or injury; or
• the job was seasonal or temporary.

The first two reasons are considered to indicate that a person has been retrenched. It is possible to determine the number of unemployed people who lost their job as a result of being retrenched from information collected in the Labour Force Survey (LFS). By combining information on the duration of unemployment (collected on a monthly basis in the LFS), together with information on the reason for unemployment (collected every three months in the LFS), it is possible to identify retrenchments, and when they occurred (for people who are still unemployed). There are additional retrenchments that will not be identified here, as the LFS only identifies retrenchments for people currently unemployed¹.

From LFS data it is therefore possible to construct three quarterly measures, which this article explores:

1. Measure 1: Retrenchment occurred less than a month ago;
2. Measure 2: Retrenchment occurred in the last 12 weeks; and
3. Measure 3: Retrenchment occurred within the last 12 months.

Data are provided back to May 2001 as it is available on a consistent basis back to this point in time. Data have also been adjusted using the trend unemployed series and have not had their own ABS time series adjustments applied². They are therefore considered to be indicative estimates only.


COMPARISON OF THE THREE MEASURES

The three measures show similar patterns over the period. Each measure clearly shows the robust period in the labour market for much of the decade as a result of sustained economic growth, ending with the impact of the economic downturn from mid-2008. The three measures indicate a turning point in retrenchments occurring after May 2009 (August 2009 for Measure 3).

It is clear from the comparison that Measure 3, that is, the number of unemployed people who experienced a retrenchment within the last 12 months, generally decreased at a faster rate than the other two measures, which reflects that unemployment is usually of a relatively short duration outside of an economic downturn. When the economy is strong many people will only appear in the other measures, given the short duration of their unemployment (sometimes referred to as 'frictional unemployment').

Over the ten year period, Measure 1 generally decreased, to a low of 19,500 in February 2006, before remaining relatively stable until August 2008, where it stood at 23,600. After this point the series increased rapidly to a high of 49,900 in May 2009.

After an initial decrease and a period of relative stability, Measure 2 saw a large increase from 47,200 in August 2008, to a high of 105,300 in May 2009. This peak was consistent with the series high in Measure 1. The peak in Measure 3 occurred a quarter later, in August 2009, at which point it stood at 180,600.


SEX

A comparison of the measures by sex, with a particular focus on the recent economic downturn, reveals a difference in both the magnitude of new retrenchments for unemployed men and women and the length in time before reaching a turning point in the number of new retrenchments.

Retrenchments reported by unemployed men increased noticeably after August 2008 and appeared to plateau from November 2008, until beginning to fall off in November 2009. For example, during this year Measure 1 reveals an average of 28,600 unemployed people who reported having been retrenched less than a month before.

By comparison, the three measures for females revealed a noticeable increase to May 2009 (for Measure 1 and 2) and August 2009 (for Measure 3), after which point they decreased, which was in contrast to the plateau in the males series. At May 2009 there were 19,400 unemployed women who had been retrenched less than a month ago, which was well above the average of 10,100 for the entire period.

OTHER ABS MEASURE OF RETRENCHMENT

A retrenchment indicator is published in Australian Social Trends (cat. no. 4102.0). Data for this measure are drawn from the Labour Mobility Survey, which is a two-yearly survey run in February as a supplement to the Labour Force Survey (LFS), with results published in Labour Mobility, Australia (cat. no. 6209.0). This retrenchment rate is the number of people retrenched during the 12 month period before the survey, as a percentage of all people who had been employed at some time over the same period.


FURTHER INFORMATION

For more information, please contact Bjorn Jarvis on (02) 6252 6552 or email bjorn.jarvis@abs.gov.au.


END NOTE

1. It must be recognised that these measures are confined to those people currently unemployed. There are other groups who will also have experienced a retrenchment who are not in the scope of these measures. These include:

1. Unemployed people who are not unemployed in the quarter months of February, May, August or November, such as someone unemployed in December who finds a job in January;
2. Employed people, who have gone straight from a retrenchment to a new job, or who have been unemployed for a duration short enough for them not to be reflected in quarterly LFS data; and
3. People not in the labour force, who may not be looking for work or available for work following a retrenchment.

2. Trend estimates of unemployed males, females and persons, which are released in Labour Force, Australia (cat. no. 6202.0), have been used to adjust the original estimates in this article. Proportions were derived from the original data and applied to the trend estimates, to reduce the seasonality and irregular influences in these experimental estimates (such as sampling variability). As a result, they are considered to be indicative estimates only.