TRENDS IN HOURS WORKED


INTRODUCTION

Are Australians working more than ever? How many hours do we work, especially during times of economic downturn? These are questions often posed in relation to work-life balance and overall labour market performance.

Hours worked estimates over time can provide rich insights into the relative strength of the labour market during different stages of the economic cycle, and the extent to which different groups within the labour market are working different numbers of hours.

This article examines trends over the past 32 years in average actual hours worked per week, and complements and extends the analysis presented in the article “Aggregate Monthly Hours Worked Per Employed Person”, published in the July 2010 issue of Australian Labour Market Statistics (cat. no. 6105.0). That article examined trends over the past 32 years in the aggregate monthly hours worked per employed person by full-time and part-time status, and by sex. The average actual hours worked measures used in this article also allow a more detailed analysis of long-term trends in the distribution of hours worked.

While it may seem as if Australians are working longer hours than ever, the average actual hours worked per employed person have decreased over the past 32 years. However, the average actual hours worked by full-time and part-time employed people have both increased (although average actual hours worked by full-time employed people have been decreasing since 2000). This total decrease, but component increase, can be attributed to the changing full-time to part-time composition of the workforce. These trends in the average actual hours worked for full-time and part-time employed people, and for all employed people, are briefly discussed in this article, along with how average actual hours worked differ for men and women. Trends in the actual hours worked by hours worked cohorts are explored, in particular for those working few (1-15) hours and those on the cusp (30-34 hours) of full-time employment.

Detailed information about the labour force in Australia has been collected on a monthly basis in the Labour Force Survey since February 1978. This article uses estimates from this survey, and from the supplementary survey Forms of Employment. Estimates quoted "over the past 32 years" refer to estimates in February 1978 and July 2010, except where noted.


HOURS WORKED MEASURES

There are three measures of hours worked produced from the Labour Force Survey: actual hours worked; usual hours worked; and aggregate monthly hours worked. See the Glossary for definitions of these three measures. For more detail on these measures, as well as on full-time and part-time employment, please refer to the "Employment Measures and Classifications" chapter in Labour Statistics: Concepts, Sources and Methods (cat. no. 6102.0.55.001). This article uses estimates of actual hours worked.


AVERAGE ACTUAL HOURS WORKED

Average actual hours worked per week in all jobs have generally decreased over the past 32 years, from approximately 35.5 hours in early-1978 to approximately 33 hours in 2010. Hours worked peaked in 1979, which roughly coincided with full-time employment peaking at around 86% of employed people (which occurred in January 1979). From this high, average actual weekly hours worked fell during the early-1980s economic downturn, driven by a sharp fall in the hours worked of full-time employed persons. Since then, hours worked have, in general, decreased: from 1984 to 2000, average actual weekly hours worked per employed person remained between 34 and 35 hours. Average actual weekly hours worked then trended down after 2000, falling below 34 hours. A rapid increase in hours worked occured between mid- and end-2007, just before the economic downturn of 2008-09. Since the economic downturn there have been further falls in average hours.

Full-time and Part-time Status and Sex

While average actual hours worked for all employed people have generally decreased, average actual hours worked for both full-time and part-time employed people have generally increased over the past 32 years. However, since 2000, average actual hours worked per full-time employed person have been falling.

Average actual weekly hours worked per full-time employed person generally rose following the early-1980s economic downturn, and reached a peak of approximately 41.5 hours in 2000. From there, average hours worked per full-time person decreased, until an increase in 2007 (driving the increase in average hours worked for all employed persons). Since 2008, average full-time hours have decreased again.

Average hours worked for part-time employed people have demonstrated a relatively steady, albeit small increase, from around 15 hours in August 1978 to approximately 16 hours in January 2010. The average hours of part-time employed people exhibited a steady increase between 1990 and 2005, fell slightly between 2005 and 2007, then rose to the beginning of 2008, before falling again during the economic downturn of 2008-09. Additionally, compared to hours worked for full-time employed people, average actual hours worked per part-time employed person have recovered more rapidly following (recent) economic downturns.

The trends in average actual hours worked are markedly different for men and women, and this is largely driven by differences in full-time and part-time employment. From February 1978 to July 2010, the percentage of employed men who worked part-time rose from 5% to 17%, compared with a rise from 33% to 46% for women. The average actual hours worked of all employed men (both full-time and part-time) are similar to the average actual hours worked of full-time employed women, especially over the last ten years, largely due to the relatively small proportion of men working part-time.

Full-time employed men have, over the past 32 years, worked higher hours than full-time employed women. In July 2010, full-time men worked 41.0 hours compared with 35.8 hours for full-time women. On average between February 1978 and July 2010, full-time men worked 4.1 hours more than full-time women. In April 1999, the difference was at its greatest (5.9 hours), and the smallest difference occurred in January 1983, when full-time men worked an average of just 1.4 hours more than full-time women.

While there is a large difference in the average hours worked by full-time men and women, since February 1978, the average hours worked for part-time workers have been quite similar for both men and women. The difference was at its greatest in January 1984, when part-time men worked 2.4 more hours than part-time women, and has since converged such that since approximately 1990, part-time men and part-time women work almost the same hours.


Occupation

Of all occupation major groups, Managers had the highest average actual hours (as seen in graph 3), however, their average actual hours worked have also exhibited the largest overall decrease in hours, from 47.6 hours in August 1996 to 43.3 hours in August 2010.

While average actual hours worked by all persons have decreased from around 34.5 hours (in August 1996) to around 32.5 hours (in January 2010), a decrease of almost 6%, this fall was not uniform across occupation groups. For example, average actual hours worked fell by only 2% for Machinery operators and drivers and Clerical and administrative workers.

Sales workers had the lowest average actual hours worked in August 2010 (26.8 hours), down from 29.4 hours in August 1996 (a decrease of 9%). The relatively low average hours of Sales workers can partly be explained by the high concentration of part-time work amongst Sales workers (55% in August 2010).


Employment Type: Employees and Owner Managers

Estimates from the Forms of Employment Survey (FOES), collected in November each year, allow analysis of average actual hours worked by the nature of a person's employment relationship or contract, or employment type. Information on the 'Employment type' classification can be found in the Employment Measures and Classifications chapter in Labour Statistics: Concepts, Sources and Methods (cat. no. 6102.0.55.001).

As shown in table 4, people who own and work in their own business (which comprises Owner managers of incorporated enterprises (OMIEs) and Owner managers of unincorporated enterprises (OMUEs)) work considerably longer hours than do employees (excluding OMIEs). In November 2009, the average hours worked by employees (excluding OMIEs) was 33.0 hours, compared with 39.2 hours for Owner managers (43.0 hours for OMIEs and 37.2 hours for OMUEs).

The high average actual hours worked for all Owner managers is largely driven by full-time Owner managers, with an average of 48.5 hours per week (OMIEs at 50.1 hours, and OMUEs at 47.5 hours), while full-time employees (excluding OMIEs) worked an average of 39.6 hours. In contrast, part-time Owner managers worked fewer hours, on average, than part-time employees (excluding OMIEs), at 15.5 hours and 17.1 hours respectively.

4. Average Actual Hours Worked, by Employment type and Full-time and Part-time status - November 2009
		Owner Managers
	Employees (excluding OMIEs)	Owner managers of incorporated enterprises	Owner managers of unincorporated enterprises	All owner managers	Total
	hours	hours	hours	hours	hours
Full-time	39.6	50.1	47.5	48.5	41.3
Part-time	17.1	16.9	15.0	15.5	16.8
Total	33.0	43.0	37.2	39.2	34.1
Source: Forms of Employment Survey.

Whether Usually Work the Same Number of Hours Each Week

While this article examines trends in the number of hours worked, it is also interesting to analyse those workers whose hours vary from week to week. Table 5 (using data from FOES) shows, for different occupations, the percentage of people who usually work the same number of hours each week.

5. Whether Usually Works the Same Number of Hours Each Week, by Occupation - November 2009
	Usually works the same number of hours	Does not usually work the same number of hours
	%	%
Managers	73.5	26.5
Professionals	76.1	23.9
Technicians and trades workers	71.3	28.7
Community and personal service workers	67.5	32.5
Clerical and administrative workers	83.2	16.8
Sales workers	69.2	30.8
Machinery operators and drivers	69.8	30.2
Labourers	68.2	31.8
Total	73.4	26.6
Source: Forms of Employment Survey.

Occupations such as Community and personal service workers, Labourers, and Sales workers have the highest proportion (around one third) of people who do not usually work the same number of hours each week. Workers in these three occupations also had the lowest average hours (in November 2009 of 27.1, 28.8 and 26.4 hours respectively).

For those occupations with a high percentage of people usually working the same number of hours each week, the converse (ie. that they work a larger than average number of hours) does, in general, not hold. Clerical and administrative workers have the highest percentage of people usually working the same number of hours per week (83%), yet their hours worked are moderate (at an average of 30.0 hours in November 2009).


DISTRIBUTION OF HOURS WORKED

Trends in average actual hours worked for employed people are not only influenced by the relative contribution of full-time and part-time employment, but also by changes in the distribution of hours worked over time. Several trends are evident: a growing percentage contribution of persons working either longer or shorter hours; and a declining proportion of persons working 'traditional full-time hours' (ie 35-40 hours).

Graph 6 below shows the percentage of persons in the different hours worked ranges, and how this has changed over the past 32 years. While the largest proportion of people have average actual hours worked between 35 and 40 hours, this proportion has declined significantly over the past 32 years (from around 43% in 1978 to around 30% in 2010), with a low of around 28% between 2001 and mid-2007. More recently, however, the proportion working 35-40 hours has remained relatively steady, with an average between January 2000 and July 2010 of 29%.

In the part-time (ie. 1-34) hours ranges, the proportion of people working 1-15 or 16-29 hours has increased (from around 8% to around 12%, and around 9% to almost 15% respectively), while the proportion working 30-34 hours has fluctuated between approximately 8% to 12% over this period.

While the proportion working 41-49 hours has remained relatively stable around 10% to 13% over the past 32 years, the proportion working over 50 hours increased from 13% in 1978 to 19% in late 1999 and early 2000, before falling to around 15% in 2010.


Gender

Although the highest proportions of both men and women are working 35-40 hours, as seen in graphs 7 and 8, the proportions have decreased over the past 32 years.

In 1978, the proportion of employed people working these hours was around 43% for men and approximately 40% for women. This has now fallen to approximately 32% for men and 27% for women. For both men and women, there was a decline until 2007, then a slight increase during the most recent economic downturn.

Across the other hours worked ranges, men are more highly concentrated in the higher full-time hours worked ranges (41-49 and 50 hours or more), while women are more highly concentrated in 1-15 and 16-29 hours worked ranges. Over time however, the percentage of men working part-time hours has grown, and the proportion of women working full-time hours has also increased.


People "On The Cusp" Of Full-time Employment

It should be noted that while the threshold for full-time employment is 35 hours, many people who are defined as full-time can record actual hours worked in the reference week of less than 35 hours. Full-time employed people are those who usually work 35 hours or more per week in all jobs (regardless of how many hours they worked in the reference week), or, although usually working less than 35 hours a week, actually worked 35 hours or more during the reference week.

As a result, people who work these "on the cusp" (of full-time) hours are comprised of two groups: the part-time employed (ie. those who usually work less than 35 hours per week, and did so in the reference week); and full-time employed people who worked between 30 and 34 hours in the reference week.

In May 2010, just over half of those who worked between 30 and 34 hours in the reference week were full-time employed people, a marked fall from the 75% recorded in 1978, highlighting the large increase in the proportion of part-time employed people working 30-34 hours. In 1978, approximately 15% of part-time employed people worked 30-34 hours. By July 2010 this had increased to just under 30%.

This growing percentage of part-time employed people who are "on the cusp" of full-time employment largely accounts for the increase in average actual hours worked per part-time employed person since 1978.


People Working Fewer Hours

The proportion of employed people working 1-15 hours has increased 3.4 percentage points from 7.5% in February 1978 to 10.9% in July 2010. This has largely been driven by increasing proportions of both younger and older workers working 1-15 hours (up from 8% to 24% for those aged 15-24, and 8% to 13% for those aged 55 years or over).

The highest proportion of persons working 1-15 hours are those aged 15-24 years, and this age group has become over-represented amongst people working 1-15 hours. For example, while people aged 15-24 comprised 26% of all employed people and 27% of those working 1-15 hours in 1978, in 2010 they comprised 17% of all employed people and 33% of those working 1-15 hours. For people working 1-15 hours, the proportion aged 55 years or over has increased from 13% in February 1978 to almost 20% in July 2010.

For men working 1-15 hours per week, those aged 15-24 comprised the the largest share, with on average, more than 45%, with a peak of around 59% in May 1990. The distribution of the other employed men working 1-15 hours is spread relatively evenly across the other age groups, and has been relatively constant over the past 32 years (although there was an increase in the share of those aged 55 and over in the last ten years: from around 16% in 2000 to approximately 23% in 2010). Unlike men, for women there is no one particular age group which overwhelmingly contributes to those working 1-15 hours. The largest group are those aged 15-24, which has increased from 23% in February 1978 to 29% in July 2010. The proportion of employed women 55 and over working 1-15 hours per week has shown a similar pattern to men in the same age group, increasing steadily since 1995, from around 9% to approximately 17%.

In conjunction with the the high proportion of young (15-24) people who work few hours, young people are also over-represented among the underemployed (see Glossary for definitions of underemployed workers and the underemployment rate). Furthermore, this over-representation has increased over the last 20 years. Over the past 32 years, the underemployment rate in the 15-24 year old age group rose 10.0 percentage points from 3.1% in February 1978 to 13.1% in July 2010.

Since the economic downturn of the early-1990s, the underemployment rate for people aged 15-24 has been considerably higher than in all the other age groups.


FURTHER INFORMATION

For more information on the estimates, analysis and methodology used in this article, please contact Matt Dillon on (02) 6252 5183 or email Matt.Dillon@abs.gov.au.


END NOTE

The average actual hours worked estimates are original terms, and are not available as seasonally adjusted or trend estimates. ABS investigations have determined that actual hours worked are not suitable to be seasonally adjusted or trended, and instead that aggregate monthly hours worked provides the most accurate seasonally adjusted measure of hours worked. For further information, please refer to Information Paper: Expansion of Hours Worked Estimates from the Labour Force Survey (cat. no. 6290.0.55.001).

As average actual hours worked original estimates are strongly influenced by seasonality, they are limited in their usefulness in monitoring movements. However, for the purpose of exploring long-term trends in actual hours worked in this article, indicative estimates are produced by applying a 13-point symmetric moving average to monthly original estimates. This procedure is not equivalent to an ABS time series analysis of seasonally adjusted or trend estimates. These indicative estimates can only be produced for original estimates more than six data points from the series ends. Graphs plot these indicative estimates, and thus have a range of August 1978 to January 2010. Estimates quoted in the article are both indicative estimates (qualified by the terms "approximately", "about" and "around") and original estimates.

Underemployment estimates are given as trend.