Employee Earnings and Hours Survey

Table of contents

• Key points
• Background
  • The EEH Survey
  • Advantages of the EEH Survey
  • Methods of setting pay
• Main findings
  • The overall gender pay gap for Australia
  • Coverage of method of setting pay for women
  • Women's earnings by method of setting pay
  • Gender pay gap by method of setting pay
  • Women's employment status
  • Gender pay gap by various employment characteristics
  • Gender pay gap by employer size
• Conclusions

Key points 

• The latest available data from the Australian Bureau of Statistics (ABS) Average Weekly Earnings Survey (the AWE Survey) point to a steady increase of the gender pay gap in Australia over the last four to five years. While the AWE Survey is the most frequent and commonly used source of data for measuring the gender pay gap in Australia, it provides only limited scope for exploring possible explanations for the widening of the gap.
• The ABS Employee Earnings and Hours Survey (the EEH Survey) is an important alternative source of data for measuring and analysing trends in the gender pay gap in Australia. While only conducted once every two years, the EEH Survey provides more disaggregated and detailed data than those in the AWE Survey, allowing more scope for analyses of potential underlying trends.
• Consistent with recent data from the AWE Survey, the analysis of the results from the latest two EEH surveys, conducted in May 2006 and August 2008, show that, overall, there has been a widening in the gender pay gap in Australia. The EEH Survey data show, in particular, that the gender pay gap, based on the average hourly ordinary time earnings of full-time non-managerial adults, increased from 10.0 per cent in May 2006 to 11.0 per cent in August 2008.
• In terms of the more disaggregated data, a key aspect of the EEH Survey is the information it provides on the coverage and earnings of men and women under different wage-setting arrangements, including awards, collective agreements and individual arrangements, such as Australian Workplace Agreements (AWAs).
• The data reveal key differences between men and women in the wage-setting arrangements that they are generally covered by. In August 2008, women were more likely to be covered by collective agreements (43.3 per cent) and awards (19.9 per cent), whereas men were more likely to covered by individual arrangements (43.2 per cent) and collective agreements (36.5 per cent), with a relatively small proportion on awards (13.3 per cent).
• The data similarly show disparities in the earnings of men and women under the various wage-setting arrangements with distinct differences in the gender pay gap across the pay-setting methods. Specifically they show that in August 2008, the gender pay gap, based on the average hourly total earnings of all non-managerial employees, was wider under AWAs (14.0 per cent) than under registered collective agreements (11.3 per cent) and awards (-6.6 per cent).
• While the negative pay gap for awards indicates that this is the only wage-setting arrangement under which women had higher hourly total earnings on average than men, their earnings under this wage-setting arrangement in August 2008 were 25 per cent less on average than those of women under all wage-setting arrangements ($19.40 per hour compared with $26.20 per hour).
• Further analysis of the data shows clear differences in the gender pay gap by private and public sector, industry, occupation and state and territory, highlighting the importance of going beyond headline, overall measures of the gap and examining differences based on the characteristics of employees.
• The analysis also shows that between May 2006 and August 2008 there were quite disproportionate increases in the gender pay gap for a number of categories.
• More detailed analysis and research of unpublished data from the EEH Survey may help to determine the nature and extent to which various underlying factors, including differences between men and women in the coverage of various wage-setting arrangements, may be a potential driver of these increases.

Return to Top of Section

Return to Top

Background

• The EEH Survey
• Advantages of the EEH Survey
• Methods of setting pay

The EEH Survey

The Australian Bureau of Statistics (ABS) Employee Earnings and Hours Survey (the EEH Survey) is an employer-based survey conducted once every two years, providing detailed statistics on the composition and distribution of employee earnings and paid hours of work. Since 2000, the survey has been the primary ABS data source on methods of setting pay including awards, collective agreements and individual arrangements.

Return to Top of Section

Return to Top

Advantages of the EEH Survey

Although the EEH Survey is only conducted once every two years, it has a number of key advantages over other data sources used to measure the gender pay gap in Australia, including the most widely known source, the quarterly Average Weekly Earnings Survey (the AWE Survey). For example, the EEH Survey is the only source of ABS data on pay-setting arrangements.

Another key advantage of the EEH Survey compared with other ABS sources of earnings data, including the Employee Earnings Benefits and Trade Union Membership Survey (EEBTUM Survey) and the Survey of Income and Housing Costs (SIHC) is that, like the AWE Survey, it is a survey of employers. This means, essentially, that the data is generally taken directly from pay-roll systems. As such, both the EEH and AWE surveys are considered to be more reliable and accurate than the EEBTUM Survey and the SIHC, which as household surveys are both prone to problems of respondent recall.

A third major advantage of the EEH Survey is that the data are highly disaggregated allowing for measures of the gender pay gap to be derived that would not be possible using the AWE Survey. These measures include, in particular, a set based on hourly, in addition to, weekly earnings. The use of hourly measures is a key consideration in measuring the overall gender pay gap, as women generally tend to work fewer hours than men, even when restricted to full-time employees, and this can distort estimates of the gap making them appear bigger than they otherwise would be. While it is possible to derive hourly measures of the gap from the AWE Survey, based on hours worked data from the ABS Labour Force Survey, the combination of data from the two surveys with different sampling methodologies, makes such measures less reliable than those obtained directly from the EEH Survey.

Finally, in addition to providing data that allows for measures of the gender pay gap to be derived on either an hourly or weekly basis for standard categories such as industry and state and territory, the EEH Survey also includes data on earnings by occupation, which are not included in the AWE Survey.

Return to Top of Section

Return to Top

Methods of setting pay

The method of setting pay data in the EEH Survey include data on awards, collective agreements, unregistered individual arrangements and registered individual agreements. The individual arrangements category includes unregistered individual contracts, letters of offer, common law contracts, over-award payments and working proprietors. The registered individual agreements category includes both federal and state registered individual agreements. As the number of state registered individual agreements is considered to be relatively small, compared with those in the federal jurisdiction, that is Australian Workplace Agreements (AWAs), the registered individual agreement category is generally taken to be a measure of AWAs.

Data by method of setting pay has been collected as part of the EEH Survey since 2000. While normally conducted in May, the survey was delayed in 2008, until August. Accordingly, the timeframe between the latest two surveys is May 2006 and August 2008.

Return to Top of Section

Return to Top

Main findings

• The overall gender pay gap for Australia
• Coverage of method of setting pay for women
• Women's earnings by method of setting pay
• Gender pay gap by method of setting pay
• Women's employment status
• Gender pay gap by various employment characteristics
• Gender pay gap by employer size

The overall gender pay gap for Australia

As noted above, the most frequent and commonly used source of ABS data on the gender pay gap is the quarterly AWE Survey. While the gender pay gap can fluctuate considerably over the short term due to the volatility of the AWE data, the most recent data for May 2009, presented in Chart 1, show that there has been a steady increase in the gap over the last four to five years. Specifically, the headline measure, based on the average weekly ordinary time earnings of full-time adults (seasonally adjusted data) has risen from the most recent low of 15.1 per cent in August 2004 to 17.4 per cent in May 2009, the highest it has been since November 1995.

Chart 1: Gender pay gap - headline measure based on the average weekly ordinary time earnings of full-time adults, May 1984 to May 2009

Source: ABS Average Weekly Earnings Survey, Australia (Cat No 6302.0) May 2009, seasonally adjusted data

While not as frequent as the AWE Survey, the EEH Survey has the advantage that it includes more disaggregated data that allow various measures of the gap to be derived, controlling for key differences between men and women that can otherwise distort the size of the gap. These various measures are provided in Table 1, together with comparable estimates, where available, from the AWE survey.

Notes: (a) original data; (b) this measure is derived by dividing average weekly total earnings original data from the AWE Survey by average hours worked of non-farm, full-time employees aged 20 plus original data from the ABS Labour Force Survey; (c) this is the same measure as the headline measure provided in Chart 1, but to allow for comparability with the EEH Survey, is based on original not seasonally adjusted data.

Sources: ABS Average Weekly Earnings Survey, Australia (Cat No 6302.0) February 2009; ABS Labour Force Survey (Cat 6291.0.55.003) data cube E05; and ABS, Employee Earnings and Hours Survey, Australia (Cat No 6306.0), May 2006 and August 2008;

The various measures presented in Table 1 show how different factors such as employees' full-time and part-time status, hours worked and overtime can impact on the size of the overall gender pay gap. Specifically, the data show that in May 2006 and August 2008:

• the smallest gap was based on the hourly earnings of full-time employees and ranged from around 10 per cent to 13 per cent;
• the second smallest gap was based on the average weekly ordinary time earnings of full-time employees and ranged from around 12 per cent to 17 per cent;
• the third lowest gap, again for full-time employees, was based on average weekly total earnings, including overtime, and ranged from 16 per cent to 20 per cent;
• the highest gender pay gap was that based on the weekly total earnings of all employees, including both full-time and part-time employees, and was 31 per cent.

In terms of the changes in the gap between May 2006 and August 2008, nearly every measure showed an increase.

Return to Top of Section

Return to Top

Coverage of method of setting pay for women

The EEH Survey data show key differences in the coverage of pay-setting arrangements between men and women. As presented in Table 2, the data show, in particular, that:

• in August 2008, the most common method of setting pay for women was registered collective agreements (42.6 per cent), while for men it was unregistered individual arrangement (40.7 per cent);
• the proportions of both men and women reliant on awards fell between May 2006 and August 2008, but a significantly higher proportion of women than men remained award-reliant in August 2008 (19.9 per cent compared with 13.3 per cent);
• while the proportions of men and women paid as working proprietors of an incorporated business remained relatively stable between May 2006 and August 2008, the proportion of women covered by this method of setting pay remained considerably less than that of men in August 2008 (2.9 per cent compared with 7.0 per cent);
• the coverage of AWAs was less for women than men in both May 2006 (2.5 per cent compared with 3.6 per cent) and August 2008 (1.8 per cent compared with 2.5 per cent); and
• AWA coverage declined for both men and women between May 2006 and August 2008, with the decline for men greater than that for women (1.1 percentage points compared with 0.7 percentage points).

Source: ABS, Employee Earnings and Hours Survey, Australia (Cat No 6306.0), May 2006 (Table 12) and August 2008 (Table 12)

Return to Top of Section

Return to Top

Women's earnings by method of setting pay

As presented in Table 3, the EEH Survey provides data on men's and women's total hourly earnings under different methods of setting pay for non-managerial employees. The data show, in particular, that:

• across all methods of setting pay, women earned an average of $26.20 per hour in August 2008, $3.90 less than the average of $30.10 per hour for men;

Source: ABS, Employee Earnings and Hours Survey, Australia (Cat No 6306.0), May 2006 (Table 20) and August 2008 (Table 20)

• women's earnings were highest in August 2008 under unregistered collective agreements ($31.20 per hour) followed by registered collective agreements ($28.40 per hour);
• men also earned most under unregistered collective agreements in August 2008, an average of $35.70 per hour, followed by $32.00 per hour under registered collective agreements;
• the lowest total hourly earnings for both men and women in August 2008 were for awards, with women earning an average of $19.40 per hour or slightly more than the $18.20 per hour average of men;
• after awards, men and women covered by AWAs had the second lowest hourly earnings in August 2008,with women earning an average of $25.90 per hour compared with $30.10 per hour for men;
• across all methods of setting pay, women's total hourly earnings increased 12.9 per cent between May 2006 and August 2008, slightly less than the 14.4 per cent increase in the total hourly earnings of men;
• women covered by unregistered collective agreements had the highest increase in their total hourly earnings between May 2006 and August 2008 (50.7 per cent) while women under awards had the lowest increase (4.3 per cent);
• the pattern of increase across the different methods of setting pay was the same for men as for women between May 2006 and August 2008, although somewhat more marked with male total hourly earnings increasing by just 1.1 per cent under awards and over 51.3 per cent under unregistered collective agreements; and
• between May 2006 and August 2008, the total hourly earnings of women on AWAs increased at nearly twice the rate of men, 13.6 per cent compared with 7.1 per cent.

It should be noted that further analysis of more disaggregated data from the survey is needed in order to provide insights as to possible explanations for the changes in the gender pay gap that occurred under different methods of setting pay between May 2006 and August 2008.

Return to Top of Section

Return to Top

Gender pay gap by method of setting pay

The EEH Survey data provide evidence that wage-setting arrangements may have an impact on the overall gender pay gap. As presented in Table 4, measures based on the average total hourly earnings of all non-managerial employees show that:

Source: ABS, Employee Earnings and Hours Survey, Australia (Cat No 6306.0), May 2006 (Table 20) and August 2008 (Table 20)

• in August 2008, the highest gender pay gap was for unregistered individual arrangement (15.5 per cent) followed by AWAs (14.0 per cent), unregistered collective agreement (12.6 per cent) and registered collective agreement (11.3 per cent);
• the gender pay gap for awards was negative in both May 2006 (-3.3 per cent) and August 2008 (-6.6 per cent) and is the only method of setting pay for which women earn more, on average, than men;
• for employees covered by AWAs, the gender pay gap decreased 4.9 percentage points between May 2006 and August 2008;
• awards and AWAs were the only methods of setting pay where the position of women improved relative to men in the period between the two latest EEH Surveys,:
• for award-reliant employees, the gender pay gap changed 3.3 percentage points to -6.6 per cent;
• for AWA employees, the gender pay gap fell 4.9 percentage points from 18.9 per cent to 14.0 per cent but remains above the gap recorded for those on awards and collective agreements;
• the gender pay gap across all methods of setting pay increased 1.2 percentage points between the two surveys, from 11.8 per cent in May 2006 to 13.0 per cent in August 2008; and
• with the exception of AWAs and awards, the gender pay gap increased for all methods of setting pay between May 2006 and August 2008, the highest increase occurring under registered collective agreements (0.8 percentage points), followed by unregistered individual arrangements (0.4 percentage points) and unregistered collective agreements (0.3 percentage points).

It should be noted that, with regard to awards, despite women earning more on average than men under this arrangement, both men and women on awards generally earn much less than those on other industrial instruments. This may have an overall negative impact on the gender pay gap as there are predominantly more women who are award-reliant than men.1

Return to Top of Section

Return to Top

Women's employment status

One factor that has an impact on the difference between men's and women's weekly earnings is employment status. Women are more likely to work in less secure and lower paid part-time and casual jobs, whereas men are more likely to work in higher paying full-time jobs. The EEH Survey includes data on the number of non-managerial employees by employment status. As detailed in Table 5, the data show that of all non-managerial employees in August 2008, women represented:

• 39.5 per cent of those working permanent full-time (59.7 per cent of all non-managerial employees);
• 78.6 per cent of those working permanent part-time (18.1 per cent of all non-managerial employees); and
• 59.3 per cent of those working casually (22.2 per cent of all non-managerial employees).

Source: ABS, Employee Earnings and Hours Survey, Australia (Cat No 6306.0), August 2008 (Table 20)

The EEH Survey also includes data on the number of full-time, part-time and casual employees covered by different methods of setting pay. As the data in Table 6 show, the majority of female non-managerial employees working either permanent full-time or permanent part-time in August 2008 were paid by registered collective agreement (44.5 per cent and 55.9 per cent respectively) while those working casually were predominantly paid by awards (42.8 per cent).

Source: ABS, Employee Earnings and Hours Survey, Australia (Cat No 6306.0), August 2008 (Table 20)

As presented in Table 7, the EEH Survey shows that with regard to the gender pay gap, there are two key findings in terms of women's employment status. Firstly, permanent part-time employees experienced a negative gender pay gap in both May 2006 and August 2008 (-1.3 per cent and -1.2 per cent respectively). In other words, the average total hourly earnings of non-managerial females working permanent part-time were slightly higher on average than those of their male counterparts ($22.80 per hour compared with $22.50 per hour in May 2006 and $25.30 per hour compared with $25.00 in August 2008).

Source: ABS, Employee Earnings and Hours Survey, Australia (Cat No 6306.0), May 2006 (Table 20) and August 2008 (Table 20)

Secondly, while the gender pay gaps for both permanent full-time and casual employees were roughly similar in August 2008 (11.3 per cent and 11.4 per cent respectively), the increase in the gap between May 2006 and August 2008 was considerably more marked for casual employees (1.8 percentage points) than it was for permanent full-time employees (0.2 percentage points).

Return to Top of Section

Return to Top

Gender pay gap by various employment characteristics

As noted above in the background, one of the key advantages of the EEH Survey, compared with the AWE Survey, is that it not only includes earnings data by occupation but also earnings data for other standard categories (including sector, industry and state and territory) on the basis of the average hourly ordinary time earnings of full-time non-managerial adult employees. As such, it allows for measures of the gender pay gap to be derived controlling for key differences between men and women.

These measures of the gender pay gap and the percentage point changes between May 2006 and August 2008 are presented in Table 8 for the key standard categories of sector, industry, occupation, and state and territory. Note that the overall measures presented at the bottom of Table 8 differ from those presented earlier in Tables 4 and 7. Specifically they show that between May 2006 and August 2008, the gender pay increased from 10.0 per cent to 11.0 per cent whereas the measures in Tables 4 and 7 show that the gap was slightly higher, increasing from 10.8 per cent to 13.0 per cent. The reason for this difference is that the measures presented in Table 8 are limited to the ordinary time earnings of full-time non-managerial adults whereas those presented in Tables 4 and 7 are based on the total earnings of all non-managerial employees, including both full-time and part-time employees and both adults and juniors. Despite these differences, the data in Table 8 provide a useful insight into the impact, individually, of the sector, industry, occupation and state and territory variables on the gender pay gap - albeit for a slightly different population.

Private and public sector

• Between May 2006 and August 2008 the difference in the private and public sector gender pay gaps narrowed somewhat, with the private sector pay gap falling marginally from 13.8 per cent to 13.7 per cent while the public sector pay gap rose from 7.3 per cent to 9.6 per cent.

Industry

• The industries with the highest gender pay gap in August 2008 were finance and insurance (27.2 per cent) and health and community services (23.4 per cent);

Notes: (a) In the explanatory notes to the EEH Survey publication the ABS notes that care should be taken when comparing survey estimates based on the ANZSCO occupation groupings and those based on the managerial status of employees. Employees are classified as managerial if they have strategic responsibilities in the conduct or operations of the organisation and/or were in charge of a significant number of employees. Estimates for employees with managerial status include employees classified to ANZSCO categories other than the ANZSCO major group 'Managers'; eg employees classified as 'Professionals' according to ANZSCO may be categorised by employers as having managerial status. Conversely, estimates of non-managerial employees may include some employees classified to the ANZSCO major group 'Managers'.

Source: ABS, Employee Earnings and Hours Survey, Australia (Cat No 6306.0) May 2006 and August 2008

• The lowest gender pay gaps in August 2008 were in accommodation, cafes and restaurants (2.7 per cent), education (5.2 per cent) and government administration and defence (5.7 per cent);
• Approximately half of all industries saw a narrowing in the gender pay gap between May 2006 and August 2008, the largest declines occurring in transport and storage (down 8.6 percentage points) followed by electricity gas and water (down 5.6 percentage points) and retail trade (down 3.5 percentage points);
• Of the eight industries that saw an increase in the gender pay gap between May 2006 and August 2008, the largest occurred in government administration and defence (up 4.6 percentage points) followed by finance and insurance and health and community services (each up 3.9 percentage points).

Occupation

• The occupations with the highest gender pay gaps in August 2008 were technicians and trades workers (19.4 per cent), community and personal service workers (19.2 per cent) and managers and sales workers (both at 18.4 per cent);
• The lowest gender pay gaps in August 2008 were for machinery operators and drivers (7.1 per cent), labourers (13.2 per cent) and clerical and administrative workers (13.3 per cent);
• Between May 2006 and August 2008, most occupations saw an increase in the gender pay gap, notably professionals (up 4.7 percentage points) and sales workers (up 3.9 percentage points);
• Occupations in which the pay gap narrowed between May 2006 and August 2008 were machinery operators and drivers (down 10.2 percentage points), followed by clerical and administrative workers (down 3.0 percentage points) and labourers (down 1.2 percentage points).

State and territory

• In August 2008, the gender pay gap across the states and territories ranged from a low in Tasmania of 5.3 per cent to a high in Western Australia of 20.4 per cent;
• The states or territories with a gender pay gap close to the national average of 11.0 per cent in August 2008 were Queensland (11.7 per cent), the Northern Territory (10.9 per cent) and South Australia (10.8 per cent);
• The remaining states and territories registered gender pay gaps slightly below the national average in August 2008 with the gap in Victoria 10.0 per cent, the Australian Capital Territory 9.7 per cent, and New South Wales 8.8 per cent;
• Between May 2006 and August 2008, the gender pay gap increased across all states and territories, except New South Wales and Victoria where it narrowed by 1.3 percentage points and 0.2 percentage points respectively;
• The highest increases in the gender pay gap between May 2006 and August 2008 were in South Australia (up 7.5 percentage points) and the Australian Capital Territory (up 5.8 percentage points).

Summary

Overall, the data presented in Table 8 reveal a number of key insights. In particular, they point to the importance of going beyond headline, overall measures of the gap and the need to look at more disaggregated measures that take into account the different characteristics of employees. This is particularly apparent in the case of occupation with these measures of the gap generally much higher than those based on the overall average earnings of male and female employees. A key issue for consideration here, and indeed for further research and analysis, is the gender pay differential within occupational groups.

Another key insight to emerge from the data in Table 8 is that changes in the gender pay gap may be just as significant as the actual size of the gap in seeking to account for underlying drivers. More detailed analysis of unpublished earnings and proportions data by sex and method of setting pay, particularly in controlling for any compositional changes in the numbers of employees covered by the different wage setting arrangements under each of the categories, may provide further insights on this issue.

Return to Top of Section

Return to Top

Gender pay gap by employer size

In addition to allowing for measures of the gender pay gap by various standard characteristics, such as sector, industry, occupation and state and territory, the EEH Survey also provides data that allow for measures across small, medium and large-scale enterprises in the private sector. As presented in Table 9 these measures show that:

• in August 2008 employers with 1,000 or more employees had the highest gender pay gap (20.3 per cent), while businesses with fewer than 20 employees had the lowest gender pay gap (9.6 per cent);
• between May 2006 and August 2008, the gender pay gap narrowed across businesses ranging from fewer than 20 employees to between 100 and 499 employees;
• the gender pay gap for employers with between 500 and 999 employees increased considerably between May 2006 and August 2008 (up 12.9 percentage points).

Source: ABS, Employee Earnings and Hours Survey, Australia (Cat No 6306.0), May 2006 (Table 11) and August 2008 (Table 11)

1. For example, of the total 1,457,700 non-managerial employees reliant on awards in August 2008, 861,300 or 59.1 per cent were women compared with 596,300 or 40.9 per cent men.

Return to Top of Section

Return to Top

Conclusions

Data from the quarterly AWE Survey point to a marked widening of the gender pay gap in Australia over the last four to five years. The analysis presented in this paper, based on more detailed data from the most recent EEH surveys conducted in May 2006 and August 2008, provides further confirmation that the gender pay gap in Australia is widening.

There are many possible explanations for the pay gap between men and women. These include occupational and industry segregation, discrimination and women bearing the primary responsibility for care-giving and household work. The gap may also be attributed, in part, to differences between men and women in the coverage of various wage-setting instruments, such as awards, collective agreements and individual arrangements.

As the primary ABS data source on wage-setting arrangements, with highly disaggregated data on earnings, the EEH Survey is a key reference for analysing trends in the gender pay gap and their potential causes. As the analysis in this paper has shown, the data indicate that methods of setting pay do appear to impact on the gender pay gap. Specifically, the gap tends to be wider under individual arrangements but lower under collective agreements and awards.

More detailed analysis and research of unpublished earnings data from the EEH Survey is required, however, to determine the nature and extent to which additional factors (including methods of setting pay) may have contributed to the gender pay gap in Australia.

Return to Top of Section

Return to Top