4125.0 - Gender Indicators, Australia, Sep 2018  
Previous ISSUE Released at 11:30 AM (CANBERRA TIME) 25/09/2018   
   Page tools: Print Print Page Print all pages in this productPrint All RSS Feed RSS Bookmark and Share Search this Product

This document was added 23/01/2019.



UNDERSTANDING MEASURES OF THE GENDER PAY GAP

Gender Indicators, Australia presents a selection of Gender Pay Gap (GPG) measures. These measures enable an understanding of the relative position of women and men in the economy. This article summarises the considerations behind each measure to help support their use and application.

The GPG is commonly expressed as either a rate ratio (proportion of female to male average earnings) or a percentage gap (difference between male and female average earnings as a percentage of male average earnings). For example if females earned an average of $900 per week and males earned an average of $1,000 per week the rate ratio would be 0.9 (females earn 90% of what males earn) and the percentage gap would be 10% (females earn 10% less than males).

Rate ratio = Female earnings / Male earnings

Percentage gap = 100 x (1 – Rate ratio)%


FIRST… WHAT CAN’T A GPG MEASURE TELL US?

While GPG measures can describe an overall position of women in the workforce, they cannot determine or explain causes of the differences in earnings between women and men.

Where a gap exists, it does not necessarily indicate that a woman is paid less than a man with similar qualifications or experience in a comparable role. The composition of women and men in different industries, different employment arrangements, such as full-time/part-time status and those in casual employment, and rate and type of pay, are all factors that influence the GPG. Some GPG measures therefore attempt to adjust for these factors by specifically referring to certain populations, pay types or pay periods, to put female and male earnings measures onto a more comparable basis.

Consequently, Gender Indicators presents three GPG measures:
  • Non-managerial adult hourly ordinary time cash earnings (mean and median (footnote 1))
  • Adult weekly total cash earnings (mean and median)
  • Full-time adult weekly ordinary time earnings (mean)

These three measures were selected for their suitability in measuring the more common dimensions of the GPG. However, it is possible to use ABS data to produce a number of population, industry, or occupation specific GPG measures, depending on the comparison(s) of interest.


SO WHAT EXACTLY ARE THESE GPG MEASURES?

Non-managerial adult hourly ordinary time cash earnings

In 2018, the rate ratio of female to male average adult hourly ordinary time cash earnings was 0.872, for non-managerial employees; women earned $36.80 per hour compared to $42.20 per hour for men.

Description

This measure is calculated from the Survey of Employee Earnings and Hours (EEH). It is produced every two years and forms the key series of earnings data presented in Gender Indicators.

This measure examines ordinary time cash earnings, which are gross earnings based on the employee’s workplace agreement (footnote 2).
  • Includes amounts salary sacrificed, allowances, regular bonuses, and payments based on performance/targets.
  • Excludes overtime earnings, irregular bonuses, severance/termination/redundancy pay, non-cash components of salary packages (however salary sacrificed amounts are included in before tax payments) and employer superannuation payments.

The population for this measure is non-managerial employees paid at the adult rate (footnote 2).
  • Includes non-managerial employees and some with managerial or supervisory responsibilities.
  • Excludes employees not paid at an adult rate (such as junior, apprentice, or trainee staff).

Strengths

The key strength of this measure is that it accounts for the broad range of employment arrangements in the labour market, and the different hours that are worked. This is important because in 2017–18, 44% of employed women and 16% of employed men worked part-time (See Table 1.9, available from the Downloads tab of this publication), and full-time women worked fewer hours on average than full-time men (36 hours per week compared to 40 hours per week, see Table 1.10).

Another benefit is the exclusion of managerial employees and those not paid at an adult rate, who generally represent the higher and lower earners, and whose inclusion could lead to unusually high or low earning amounts. Consequently, the measure better reflects the bulk of the adult workforce.

Using EEH data also allows median rate ratios and rate ratios by occupation to be produced.

Limitations

The main limitation of this measure is that it is only available every two years. This makes it less useful for monitoring the GPG on a regular basis.

Adult weekly total cash earnings

In 2018, the rate ratio of female to male average adult weekly total cash earnings was 0.689; women earned $1,099.40 per week compared to $1,594.70 per week for men.

Description

This measure is also calculated from EEH. It is produced every two years and presented as a related series of earnings data in Gender Indicators.

This measure examines total cash earnings, which are gross earnings based on the employee’s workplace agreement, plus any overtime earnings (footnote 2).
  • Includes overtime earnings, allowances, regular bonuses, payments based on performance/targets, and amounts salary sacrificied.
  • Excludes irregular bonuses, severance/termination/redundancy pay, non-cash components of salary packages (however salary sacrificed amounts are included in before tax payments) and employer superannuation payments.

The population for this measure is all employees paid at the adult rate.

Strengths

This measure is an important reflection of economic security as gross weekly pay is directly related to disposable income and superannuation.

Using EEH data also allows median rate ratios and rate ratios by occupation to be produced.

Limitations

Given the much larger proportion of the female workforce working part-time (44% of females compared to 16% of males; see Table 1.9, available from the Downloads tab of this publication), the pay gap is increased by the inclusion of both full and part-time employees, when calculated as a weekly rate (rather than hourly).

Full-time adult weekly ordinary time earnings

In 2018, the rate ratio of female to male average adult weekly ordinary time earnings was 0.854, for full-time employees; women earned $1,433.60 per week compared to $1,678.40 per week for men.

Description

This measure is calculated from the Survey of Average Weekly Earnings (AWE). It is produced every year and presented as a related series of earnings data in Gender Indicators. The same measure is also used by the Workplace Gender Equality Agency (WGEA) to calculate the ‘national gender pay gap’ featured in their fact sheets and reports.

This measure examines trend estimates (footnote 3) of ordinary time earnings, which are gross earnings based on the employee’s workplace agreement (footnote 4).
  • Includes allowances, regular bonuses, and payments based on performance/targets.
  • Excludes amounts salary sacrificed, overtime earnings, irregular bonuses, severance/termination/redundancy pay, non-cash components of salary packages and employer superannuation payments.

The population for this measure is full-time employees paid at the adult rate (footnote 4).

Strengths

The strengths of this measure are;
  • Frequency: Data used to produce this measure are collected twice a year.
  • Interpretability: It is easier to interpret the pay gap for full-time employees as there are fewer confounding issues, such as the number of hours worked and employment arrangements.
  • Stability: The trend estimate indicates the underlying level of the pay gap and is more directly comparable at different points in time.

Limitations

This measure omits a large proportion of female employees who, as noted above, are more likely to be employed on a part-time basis.

As a weekly measure reliant on full-time status, this measure does not take into account the variation in the average number of hours worked between women and men employed full-time (36 hours for women compared to 40 hours for men; see Table 1.10, available from the Downloads tab of this publication).

An important consideration when interpreting this measure is in the differences in age for full-time employees, and in particular that earnings generally increase with age. The percentage of employed women who work full-time hours peaks in the late 20’s age group (at around two-thirds of employed women in 2017-18). This percentage drops to between 54 and 61 per cent, up to the age of 60 years. Employed men, by comparison, do not experience the same changes, with the percentage employed full-time remaining at around 90 per cent for the majority of working years.

Restricting the population to full-time employees only, therefore gives greater weight to younger women earners and their influence on the size of the GPG.

Figure 1 - Percentage of female and male employees working full-time hours by age, 2017–18 (a)
Graph: shows that females have consistently lower rates of working full-time but the differences are more pronounced for people in their early thirties through to their late fifties.

Footnote(s): (a) Data averaged using 12 months in the financial year.

Source(s): Data available on request, Australian Bureau of Statistics, Labour Force Survey, Australia, cat. no. 6202.0


SO HOW DO GPG MEASURES COMPARE OVER TIME?

Figure 2 below shows how each measure has been tracking over the last decade and how they compare to one another. Whilst each measure presents a different perspective on the gender pay gap, it is clear that each measure is showing a similar trend.

Figure 2 - Mean female to male earnings rate ratios, selected sources, 2008 to 2018 (a)
Graph: shows that each measure has stayed at roughly the same level for the last decade.

Footnote(s): (a) Data only available for even years for the Non-managerial adult hourly ordinary time cash earnings series and the Adult weekly total cash earnings series.

Source(s): Data available on request, Australian Bureau of Statistics, Employee Earnings and Hours, cat. no. 6306.0; and Australian Bureau of Statistics, Average Weekly Earnings, cat. no. 6302.0


The ABS presents ‘Non-managerial adult hourly ordinary time cash earnings’ as the key series in Gender Indicators. It represents most employed people and allows direct comparisons to be made (i.e. employed women were paid 12.8% less per hour than men, on average).

The ‘Full-time adult weekly ordinary time earnings’ measure is used by WGEA. Given the consistency shown between each measure, this provides a more frequent and stable method to show ‘whether the gender pay gap is closing over time’.

It is important to note that over the past decade there has been little change in the gender pay gap. Since the measures are based on survey data, which always have an unavoidable and inherent level of sampling variability (footnote 3), the ABS recommends caution in focusing on very small period-to-period movements in the EEH and AWE data, and in the corresponding GPG measures.


DATA SOURCES FOR GPG MEASURES

The ABS collects earnings data across a number of household and business collections. Gender Indicators features GPG data based on two collections:

Survey of Employee Earnings and Hours (EEH)

The EEH is conducted every two years and designed to provide detailed statistics on the composition and distribution of employee earnings, hours paid for and the methods used to set employees' pay.

Information is collected using a sample survey at both the employer and employee level. Characteristics of both the employers (such as industry and sector) and their employees (such as occupation, employment arrangement, and method of setting pay) are obtained. The statistical units of this collection are employees.

As information is collected at the individual employee level, it is possible to derive measures of distribution (e.g. medians, deciles, earnings ranges) and provide some information on individual characteristics of employees based on data items collected.

Survey of Average Weekly Earnings (AWE)

The AWE is conducted biannually, with reference to May and November each year, and information is collected using a sample survey of employers. The statistical unit produced by the collection is employers.

The AWE collects total earnings that employers pay to their employees, and the total number of employees in the business.

The measure of GPG is calculated by dividing aggregate estimates of weekly total earnings by estimates of the number of employees, which means that distributional measures are not able to be calculated, or factors such as the composition of the labour force. Variations in the rates of full-time and part-time employees are able to be controlled for.


DIFFERENCES IN SCOPE BETWEEN EACH GPG MEASURE


Non-managerial adult hourly ordinary time cash earnings
Adult weekly total cash earnings
Full-time adult weekly ordinary time earnings
Data sourceSurvey of Employee Earnings and HoursSurvey of Employee Earnings and HoursSurvey of Average Weekly Earnings
Unit of earnings comparedHourlyWeeklyWeekly
Unit level for informationEmployeeEmployeeBusiness
Payment typeWorkplace AgreementWorkplace AgreementWorkplace Agreement

Population inclusions
Full-time employeesYYY
Part-time employeesYYN
Managerial employeesNYY
Owner managers of incorporated enterpriseNYY
Apprentices and traineesNNY

Earning inclusions
OvertimeNYN
Salary sacrificeYYN
Non-cash benefitsNNN
Regular bonusesYYY
Irregular bonusesNNY
Severance, termination or redundancy paymentsNNN
Employer contributed superannuationNNN
Voluntary superannuation contributions (post-tax)YYY
Workers’ compensation payments (made through payroll)YYY



FOOTNOTES
  1. For explanations of the terms mean and median, refer to Measures of Central Tendency in Statistical Language.
  2. For further information, refer to the Glossary in Employee Earnings and Hours, Australia, May 2016 (cat. no. 6306.0)
  3. For an explanation of the term trend data, refer to Labour Statistics Explained Frequently Asked Questions in Labour Statistics: Concepts, Sources and Methods, Feb 2018 (cat. no. 6102.0.55.001)
  4. For further information, refer to the Glossary in Average Weekly Earnings, Australia, May 2018 (cat. no. 6302.0)