Payroll Jobs methodology

STP data is reported on a cash basis (the time when the payment was made) rather than an accrual basis (the time when the payment was earned). Production of estimates require the conversion of STP data from a cash basis to an accrual basis. 

The initial transformation step is called "calendarisation". STP transactions are converted into common 'daily' pay events for individual employees, enabling their aggregation to weekly reference periods.

The calendarisation method includes the following steps:

• calculation of the periodicity (pay cycle frequency) using the start and end date of the payment period,
• calculation of a daily pay rate by dividing the total payments by the pay cycle frequency (for example, weekly pay is divided by seven), and
• an adjustment to the periodicity for a job to exclude the days before commencement (or after termination), where the start or termination date for a job occurs within a payment period.

As at the end of March 2024, STP data indicated that 33% of employees are paid weekly; 53% are paid fortnightly; 12% are paid monthly; and 2% are paid quarterly or infrequently.

While a fortnightly pay cycle frequency is the most common amongst all industries, the dominant pay cycle frequency varies across industry (seen in the graph below).

The STP reporting year is based on the July – June financial year. Within any given reference week, STP reporting may be incomplete for a range of reasons including, but not limited to: 

• different business reporting habits, and
• different pay cycle period reporting.

Imputation is a form of jobholder level non-response adjustment, applied to account for partially reported and/or short term lagged reporting. Imputed data remains in place until 16 weeks have elapsed to account for the majority of delayed business reporting (for example quarterly reporters).

In this release, the time between the reference week and release date was extended to around 18 weeks (131 days) to allow for the receipt of substantially more complete data and results in the data included in the release being outside of the imputation window.  Taking advantage of more complete STP reporting reduces volatility and revisions and provides a more accurate time series of weekly change in the labour market. Further information on imputation applied in the compilation of payroll jobs estimates for previous releases can be found in the associated methodologies. 

Once STP data is transformed via calendarisation and imputation, the data is aggregated for each week (ending Saturday) to the output classifications of state/territory, industry division and subdivision, employment size groups and age groups.

Individual STP transactions are not amended, however aggregate levels of payroll jobs may be adjusted at a state/territory by industry level to remove the effect of: 

• reporting anomalies which are significant at the state/territory by industry division level, and
• large organisation onboarding to STP reporting. 

Aggregate adjustments are applied to the applicable state/territory by industry division levels. The adjustments in each release are: 

• calculated individually for each state/territory by industry division level, and
• removed as the underlying data becomes more complete, or the adjustment is no longer required. 

Currently, it is not possible to calculate individual adjustments for all output groups, hence the adjustments applied at a state/territory by industry division are also applied equally to all component indexes (such as industry subdivision, state/industry by age group or employment size group). Aggregate level adjustments can also affect the relative proportion of characteristics. The effect is more pronounced around large organisation STP onboarding events (where they are dominant in their state/territory and industry).

When existing large employers commenced reporting through STP for the first time, the ABS determined when the employer first appeared in the dataset and adjusted underlying jobs to the start of the time series. This ensures that weekly estimates best reflect change in the labour market over time through payment activities, rather than changes in the uptake of STP reporting.

To protect the confidentiality of individuals, data may be subject to suppression where a small number of individuals contribute to an individual estimate. At times, a detailed index may be suppressed throughout the time series, if the quality or interpretability of an index are affected by the compilation process.

Payroll jobs estimates are supplied as indexes to provide an indication of movements rather than levels. In order to compare change over time, the week in which Australia recorded its 100th confirmed Coronavirus case (i.e. the week ending 14 March 2020) is used as the reference period for constructing the indexes and given an index value of 100.0. These indexes differ from the ABS' suite of price indexes (including the Wage Price Index) which measure changes in price over time unaffected by quality or quantity and should not be directly compared.

Indexes allow comparison of data between two points in time. The points in time can be adjacent (this week and the previous week) or many weeks apart. Movements in the index from one period to another can be expressed as either index points or percentage change and these are rounded to one decimal place. The following example illustrates the method of calculating change in index points and percentage change between any two periods:

The following example illustrates the method of calculating a recovery percentage change between any two periods, which can be useful in understanding the relative strength in payroll jobs increases at a point in time after a low point.

STP data is linked to other information held by the ABS to derive jobholder and employer characteristics such as age, state/territory, industry and employment size. If a specific characteristic cannot be linked or derived, and does not appear on STP datasets, it is assigned an 'unknown' category (for that characteristic).

Records with 'unknown' characteristics are included in the calculation of index totals for that category. As 'total' and 'component' indexes are calculated independently, the inclusion of records with 'unknown' characteristics in a 'total' index can result in independent movement from 'component' indexes.

Records with ‘unknown’ characteristics are excluded from the calculation of all proportions included in Table 4. Payroll jobs distributions of characteristics found in the Data downloads section.

The proportions of ‘unknown’ characteristics at mid-June 2024 are:

• almost no payroll jobs have an unknown state or territory,
• around 0.3% of payroll jobs have an unknown age,
• around 0.8% of payroll jobs have an unknown industry, and
• around 0.8% of payroll jobs have an unknown employment size.

Weekly payroll jobs indexes are available for a combination of employer characteristics (industry division/subdivision and employment size groups) and jobholder characteristics (state/territory of residential address and age groups). These characteristics are populated from variables found in the snapshots of the ABS Business and ATO Client Registers. These Registers are described in more detail in the Glossary and How data is collected sections of this release. 

ABS Business Register snapshots

In the 14 March 2024 release, employer characteristics of payroll jobs were updated using a December 2023 ABS Business Register (ABSBR) snapshot. These updated characteristics were applied back to the week ending 30 December 2023.

Once an employer's payroll job is allocated to an employer characteristic from the ABSBR snapshot, it is held constant until the next snapshot update, when missing values will be populated and existing values refreshed (if they have changed). Employers with industry and employment size characteristics that can't be linked or derived from the latest ABSBR snapshot are assigned an 'unknown' category for the relevant employer characteristic. As a snapshot ages, the proportion of records with unknown industry and employment size increases.

The date at which characteristics change from one snapshot to another, called 'transition points', can be visible in the levels of some indexes. Users should always exercise caution when comparing week-on-week changes and index levels across transition points. The table below shows the reference periods each ABSBR snapshot applies to and identifies the transition points.

Refer to the historical Update of employer characteristics section for more detailed information on changes resulting from the last ABSBR update.

The ABS will refresh ABSBR snapshots when possible to update employer characteristics and minimise the number of unknowns. 

Determining employer characteristics of newly reported payroll jobs

Newly reported payroll jobs are in one of two ABSBR population groups (profiled and non-profiled), and the assignment of employer characteristics depends on which population group the employer has on the ABSBR. For non-profiled businesses (businesses with a simple structure), employer characteristics match those of the business’ Australian Business Number (ABN). The remainder of payroll jobs are part of the profiled population (businesses with a complex structure). These payroll jobs are assigned to a Type of Activity Unit (TAU), and conceptual modelling is used to predict and assign employer characteristics (industry and employment size) to each payroll job.

Determining jobholder characteristics

Jobholder characteristics used in these estimates (age and residential state/territory) are sourced primarily from the annual ATO Client Register which reflects a historical point-in-time. When age and residential state/territory are not available from Client Register data, they are sourced from STP data which are received and updated weekly. 

The system and processes used to produce payroll jobs estimates are no longer able to accommodate ATO Client Register updates. STP records whose jobholder characteristics cannot be determined from the Client Register snapshot or STP data are assigned an ‘unknown’ category for the relevant jobholder characteristic. As persons join the workforce or register with the ATO for the first time, the Client Register snapshot becomes outdated, increasing the number of payroll jobs with a characteristic derived from the STP data. Overall, the proportions of jobholders with an unknown age or residential state/territory remain relatively small over time.

To reduce the volatility in fluctuating jobholder characteristics, once assigned these characteristics are held constant. While the month and year of birth variable is held constant, the jobholder’s age (as derived from month and year of birth) continues to be updated at the start of a jobholder’s birth month each year.

ATO Client Register snapshots updates and transition points

A mid-2019 ATO Client Register snapshot assigned the initial jobholder variables and has been updated two further times, in the 13 April 2021 release and then in the 10 March 2022 release.

As noted above, payroll job systems and processes are currently unable to support further updates from the ATO Client Register. As the sex jobholder characteristic was not able to be updated from STP data, the quality of sex indexes diminished over time resulting in a growing proportion of 'unknowns' and ultimately the withdrawal of sex indexes in this release.

The ABS is currently exploring alternative methods to produce STP employee jobs estimates with more up-to-date jobholder characteristics.

The following table shows the reference periods each ATO Client Register snapshot applies to and its transition points.

The transition point for the last two Client Register updates was aligned with the change in financial year as:

• the STP reporting is financial year based, ensuring jobholder characteristics are held constant across each STP reporting year, and
• the annual ATO Client Register snapshot is taken in July each year, aligning the update point to that of the snapshot.

Some jobholder characteristics indexes have a visible change in levels at the transition point between ATO Client Register snapshots. This is particularly visible for state/territory, where changes in residential addresses from across a year will all be reflected in a single week.

Revisions are a change to a previously published value and may arise due to a variety of reasons. They can reflect both the complexity of measurement, and the need to balance precision and timeliness to maximise the utility of the estimates. Over time, as more employer-reported STP data is received, the size of revisions usually decreases. 

In this release, the extension of the time between the latest reference week and the release date takes advantage of significantly more complete STP reporting and will reduce subsequent revisions in future releases.

The magnitude of revisions will vary between reference periods and may relate to one or more of the following factors:

• updates in the STP data including receipt of updated, more complete or resubmitted data
• a greater incidence of longer STP pay cycle periods (such as quarterly)
• higher levels of reporting seasonality
• removal of previously imputed data
• changes in implemented methods or processes
• updating of jobholder or employer characteristics 
• changes in historical data adjustments
• graduated STP onboarding (the effect of which was more prevalent up to July 2021)

The size of revisions and the period over which they apply can vary between indexes. Some component indexes are subject to higher than usual levels of revision, over a longer period. For example, the employment size group indexes show both a higher incidence and duration of revision. This is most noticeable in the small employers group (those with under 20 employees). Users should exercise caution when focusing on change in the most recent weeks of data, as this period sees the greatest levels of revision in subsequent releases, however extending the time between the latest reference week and the release date will reduce subsequent revisions in future releases.

From the September 2023 release, the ABS ceased the application of revisions prior to January 2022, except in exceptional circumstances. In this release, revisions have been applied prior to January 2022 to support the update of employment size group estimates.

These estimates are presented as an ‘original’ data series, and do not include seasonally adjusted or trend data time series found in other labour statistics releases (e.g. Labour Force, Australia). 

STP is a relatively new program (and data source). Generally, three to five years of data are required before good seasonally adjusted data can be produced, hence it is not yet possible to produce a seasonally adjusted series (with seasonal elements removed) or trend series (with both the seasonal elements and irregular fluctuations removed). This means that variations in these estimates may reflect seasonal changes in the labour market in addition to other labour market shocks (such as COVID-19 related impacts) in the economy.

Seasonal characteristics may be seen:

• ahead of the Christmas period, where considerable labour market activity in a range of industries can temporarily increase payroll jobs.
• across school holiday periods, where some industries experience lower business activity. This is particularly pronounced in the holiday period after Christmas, resulting in a decrease in payroll jobs across industries.
• in industries where there is more pronounced peaks and troughs in employment due to changes in demand for labour, such as the Agriculture, forestry and fishing and the Education and training industry.

Each year, wages and salaries data during June and July see increased variability in employer reporting, due to:

• a reset of the financial year in payroll systems, and
• employers finalising their year to date employee earnings information.

These factors can translate to greater variability in estimates over this period, as well as larger revisions in payroll jobs around this period in subsequent releases.

Estimates of change throughout this release are calculated using un-rounded index values. They may differ from, though are more accurate than, change calculated from rounded values.

The accuracy of these estimates may be subject to the following sources of error:

• Conceptual misalignment - the Australian tax system is purpose-built and complex, and in some cases, it is difficult to determine how a particular STP item should be used to describe impact on payroll jobs. While all care is taken, some income items are subject to this type of validity error. Coherence with other sources indicates that this has a low impact on the aggregates.
• Reporting error - this is likely to be present in both employee and employer information used. Given the magnitude of STP records submitted every month, a significance approach is taken to identify and adjust for reporting errors that have a significant effect on high level aggregate estimates. Some reporting errors will remain. These errors are considered to have a low impact on the aggregate series and may be corrected in subsequent employer reporting.
• Updating of employer characteristic - the ABSBR receives updated information when a business entity reports changes to either the Australian Business Register or the ATO, such as new registration, revised classificatory information, or when the business entity reports changes in its business structure or activity. These reporting changes can impact upon the business entity’s industry classification (ANZSIC), and to a lesser extent its employment size. There may be a delay between the real world timing of these changes, and the update of this information in these statistics. In addition, users should exercise caution when comparing week-on-week change and index levels across ATO Client Register and ABS Business Register update transition points.

Refer to the Updating characteristics section of How data is transformed for more detail on ABSBR snapshot transition points.

Week ending 17 February 2024

• Update of employer characteristics - ABS Business Register snapshot update to industry, employment size and sector variables

Week ending 15 July 2023

• Withdrawal of wages and sub-state geography estimates - wages and sub-state geography indexes removed

Week ending 11 February 2023

• Update to STP processes - changes to the processes supporting the use of STP data for statistical purposes

Week ending 15 October 2022

• Update of employer characteristics - ABS Business Register snapshot update to industry, employment size and sector variables

Week ending 12 February 2022

• Update of jobholder characteristics - revisions to all sex, age and geography indexes as noted in Methods review

Week ending 13 November 2021

• Update of employer characteristics - refinement of Job-TAU allocation process as noted in Methods review

Week ending 11 September 2021

• Update to imputation retention - the imputation retention threshold updated from 8 pay periods to 16 weeks

Week ending 31 July 2021

• Change in lag between reference week and release date - reducing imputation in the latest weeks of data, as noted in the Imputation section

Week ending 19 June 2021

• Update of employer characteristics - ABS Business Register snapshot update to industry and employment size variables

Week ending 10 April 2021

• Historical imputation - period over which historical imputation is removed for terminated employees changed from 3 to 8 pay periods

Week ending 27 March 2021

• Update of jobholder characteristics - Client Register snapshot update to sex, age and geography variables

Week ending 30 January 2021

• Imputation update - to deliver a reduced magnitude of future revisions, particularly in the most recent week's data

Week ending 16 January 2021

• Age derivation update - a person is aged at the start of their birthday month instead of redefining age each reference week

Week ending 2 January 2021

• Update of characteristics variables - Employers with unknown industry included in estimates, as noted in Data limitations and revisions

Week ending 14 November 2020

• Week on week revisions - persons aged under 15 years excluded from the under 20s age group

Recording wages when they are earned, accrued or incurred regardless of when payment is made or received.

A register of all Australian businesses and organisations maintained by the Australian Bureau of Statistics (ABS) for the purpose of producing statistical frames and business demography outputs. It contains identifying and classificatory data for each business and organisation.

Information to populate the ABS Business Register is largely sourced from the Australian Business Register.

The ABS Business Register consists of two subpopulations, the profiled population and the non-profiled population. The ABS Business Register uses an economic units model to describe the characteristics of businesses and the structural relationships between related businesses.

A unique identifier. To be entitled to an Australian Business Number (ABN), an organisation must be one or more of the following:

• a company registered under the Corporations Act 2001
• an entity carrying on an enterprise in Australia
• a government entity
• a non-profit sub-entity for Goods and Services Tax purposes
• a superannuation fund.
   

A non-resident entity may be entitled to an ABN if they are carrying on an enterprise in Australia and/or, in the course of carrying on an enterprise, the entity makes sales that are connected with Australia.

The data store containing details about businesses and organisations that have registered for an Australian Business Number. More information can be found on the ABR website.

A register of persons that have interacted with the Australian Tax Office (ATO) and have been issued with an Australian Tax File Number (TFN). The ATO Client Register comprises demographic information such as sex, month and year of birth, and state/territory of residential address.

Recording the wage payment in the pay period when the payment was received by the employee.

Commencement and termination dates associated with each job as reported through Single Touch Payroll (STP).

Persons who work for a private or public sector employer, where the employee has received payment in the reference week through Single Touch Payroll (STP) enabled software and reported to the Australian Taxation Office (ATO).

An organisation with an Australian Business Number (ABN) that provides employment income to one or more people, and reports through the ATO STP system.

An institutional unit covering all the operations within Australia's economic territory of legal entities under common control. Control is defined in Corporations legislation. Majority ownership is not required for control to be exercised. An enterprise group may have one or more Type of Activity Unit.

State/territory based on jobholder's residential address as sourced from either the ATO Client Register or Single Touch Payroll (STP) data. 

A homogenous grouping of economic activities undertaken to produce goods and services. The Australian and New Zealand Standard Industrial Classification is used to classify an entity to an industry based on its dominant activity.

The broadest grouping of industries within the Australian and New Zealand Standard Industrial Classification. The main purpose of the industry division level is to provide a limited number of categories, which give a broad overall picture of the economy. There are 19 mutually exclusive divisions. For more information see the Australian and New Zealand Standard Industrial Classification (ANZSIC).

The second broadest grouping of industries within the Australian and New Zealand Statistical Industrial Classification. Industry subdivisions are built up from the industry groups which, in turn, are built up from industry classes. For more information see the Australian and New Zealand Standard Industrial Classification (ANZSIC).

See payroll job.

The vast majority of businesses included on the ABSBR are in the non-profiled population. Most of these businesses have simple structures that are suitable for ABS statistical purposes at the Australian Business Number (ABN) level. For the non-profiled population, one ABN equates to one business.

Statistic is not available. This can be to protect the confidentiality of data providers or to prevent misinterpretation of statistics due to poor quality.

A person who operates their own unincorporated enterprise, which does not possess a separate legal identity to that of its owner(s), or engages independently in a profession or trade.

OMUEs can also be referred to as self-employed. OMUEs are not included in these estimates as they are not in scope of STP-enabled software reporting to the ATO.

Unincorporated enterprises are further defined in the Type of Legal Organisation (TOLO) classification section of the Standard Economic Sector Classifications of Australia (SESCA) release.

A payroll job is a relationship between an employee and their employing enterprise, where the employee is paid in the reference week through STP-enabled payroll or accounting software and reported to the ATO. Where an employee is paid other than weekly, the established payment pattern is used to identify jobs in weeks outside the payment week.

For a small number of businesses, the ABN unit is not suitable for ABS economic statistics purposes and the ABS maintains its own economic units structure through direct contact with businesses. These businesses constitute the profiled population and are represented by Type of Activity Units (TAUs). This population consists of large, complex and diverse groups of businesses (known as Enterprise Groups).

The Single Touch Payroll (STP) system sends taxation and superannuation information from a business' STP-enabled payroll or accounting software to the ATO as a business runs its payroll.

A TAU consists of one or more business entities, sub entities or branches of a business that can be grouped according to production activity, and can report a minimum set of data items. TAUs are classified according to the industry of their main activity. The relationship between TAUs and their associated legal entities (ABNs) may be one-to-one, many-to-one, or one-to-many.