6291.0.55.001 - Labour Force, Australia, Detailed - Electronic Delivery, Feb 2020 Quality Declaration 
Previous ISSUE Released at 11:30 AM (CANBERRA TIME) 26/03/2020   
   Page tools: Print Print Page Print all pages in this productPrint All RSS Feed RSS Bookmark and Share Search this Product

SPOTLIGHT: INSIGHTS INTO DETAILED LABOUR FORCE HOURS WORKED DATA


This spotlight highlights detailed Labour Force data that will be useful in analysing the impact of the coronavirus (COVID-19) on the Australia labour market in the coming months.

Assessing February 2020 for an early impact from COVID-19

Analysis of February 2020 Labour Force data did not identify any notable impact from COVID-19. It is important to note that the reference weeks for the monthly Labour Force Survey fall in the first half of each month. During the first two weeks of February 2020, there was only a relatively low number of confirmed COVID-19 cases within Australia, and it was yet to be declared a global pandemic.

Changes in hours worked often lead changes in employment

As with other major disruption to the economy, early labour market impacts are usually most evident in the monthly hours worked in all jobs series, which can be found in Table 19 of Labour Force, Australia (cat. no. 6202.0). Additional information, such as the reason people worked fewer hours than usual, and the number of people working in different categories of hours, are available within the release of detailed Labour Force data, a week later.

Hours worked can change more quickly than employment, given the variability in individual circumstances for people and businesses from one week to the next. During an economic downturn, reducing hours is often an early response taken by businesses, often with the view to avoiding people losing their jobs.

Given the variety of ways in which people work and the conditions they are employed under, the ABS uses a longstanding, comprehensive and international best practice framework for determining whether someone is employed. This framework also effectively covers employment arrangements that are more common in an economic downturn. Based on this framework some of the common examples of how people could be categorised during the COVID-19 period include:

    • If a person takes any kind of paid leave while not working, they will be classified as employed.
    • If a person is away from a job, business or farm for four weeks or less without pay for any reason, and believes they still have a job to go back to, they will be considered to be employed.
    • A person away from a job, business or farm for one month or more will only be considered employed if they were paid for some part of the previous 4 weeks.
    • If a person believes they no longer have a job, business or farm to be absent from, they will be asked additional questions to determine whether they are unemployed or not in the labour force. That is, the person must be actively looking for work, and available to start work. Those people that do not meet this criteria will be classified as Not in the Labour Force.

It is for this reason that hours worked analysis can provide an early indication of aggregate labour market impacts from any major disruption to the economy, ahead of changes in other Labour Force indicators.
    Trend and seasonally adjusted hours worked data: an aggregate view

    Measuring changes in the number of hours worked is critical to an effective understanding of how the labour market is changing over time. However, hours worked data reflect a high degree of systematic seasonal factors and effects, many of which are closely related to public holidays, school holidays and other major events in the year (eg. Christmas).

    For this reason, the ABS produces trend and seasonally adjusted series to provide insights into how aggregate hours worked within the labour market are changing over time. These series adjust for the seasonality in the underlying original data, and also convert the weekly hours measures into hours representing the entire month.

    This enables analysis of total hours worked in the month alongside employment. For example, the three charts below show annual growth rates in trend employment and trend monthly hours worked for all people, males and females.

    These charts show that hours worked growth rate for females has remained relatively strong over the past three years, usually around or above the growth rate for employment. In contrast, the growth rate for hours worked for males has been slowing more quickly than employment for the past two years, with trend hours worked now declining.

    Chart 1: Year on year trend growth in total employed and hours worked
    This graph highlights trend year-on-year growth of total employed and monthly hours worked between Feb 2017 and Feb 2020 for people. Both monthly hours worked and total employed year-on-year growth rates have been trending down slowly since June 2019.
    Chart 2: Year on year trend growth in male employed and hours worked
    This graph highlights trend year-on-year growth of total employed and monthly hours worked between Feb 2017 and Feb 2020 for males. Both monthly hours worked and total employed year-on-year growth rates have been trending down slowly since May 2019.
    Chart 3: Year on year trend growth in female employed and hours worked
    This graph highlights trend year-on-year growth of total employed and monthly hours worked between Feb 2017 and Feb 2020 for females. Both monthly hours worked and total employed year-on-year growth rates have been trending down slowly since October 2019,
    Source: 6202.0 Tables 1 and 19

    Original hours worked data: a compositional view

    Released in the week following the headline Labour Force aggregate measures, the detailed monthly and quarterly Labour Force releases (6291.0.55.001 and 6291.0.55.003) contain additional hours worked data, in original terms. Care should be taken in interpreting month-to-month changes in the detailed original data, given the systematic seasonality in hours worked data. Detailed data for specific sub-populations will also be inherently more volatile than higher level aggregates, and the ABS generally recommends using smoothing techniques when using very detailed data.

    Much of the analysis presented in the remainder of this spotlight focuses on February data, and how February 2020 compares with February data from the 2015-2019 period. The year-on-year comparison approach helps to control for the effects of seasonality in the original data.

    Every month the ABS releases information on the number of people working in different groups of hours worked, by various characteristics. Table 1 below shows the distribution of employed males and females across the hours worked categories, over the past 5 years. The data shows that, over the past 5 years:
      • the proportion of males working very long hours (60 hours or more) consistently fell, while the proportion working part time hours (under 35 hours) rose.
      • the proportion of females working 35-44 hours per week increased, while those working longer full time hours (45 hours or more) fell.
    It is important to note that this underlying pattern largely continued from February 2019 to February 2020, with the weakness in male hours worked also evident in the aggregate measures.

    Table 1: Proportion of hours worked by male and females

    Year
    Males, hours worked in all jobs, Original
    Females, hours worked in all jobs, Original
    1-19 hours
    20-34 hours
    35-44 hours
    45-59 hours
    60+ hours
    1-19 hours
    20-34 hours
    35-44 hours
    45-59 hours
    60+ hours
    Feb 2015
    10.0%
    13.5%
    43.0%
    22.6%
    11.0%
    21.6%
    29.2%
    35.4%
    10.5%
    3.3%
    Feb 2016
    9.2%
    14.1%
    43.5%
    22.2%
    10.9%
    21.1%
    29.1%
    36.3%
    10.1%
    3.3%
    Feb 2017
    9.6%
    14.5%
    44.1%
    21.7%
    10.1%
    21.7%
    29.7%
    34.9%
    10.4%
    3.3%
    Feb 2018
    9.4%
    14.8%
    43.7%
    21.6%
    10.5%
    20.7%
    29.7%
    36.3%
    10.3%
    2.9%
    Feb 2019
    9.7%
    14.4%
    44.5%
    21.9%
    9.6%
    19.8%
    30.5%
    37.0%
    9.8%
    2.9%
    Feb 2020
    10.3%
    15.8%
    44.3%
    20.8%
    8.8%
    20.5%
    29.7%
    37.0%
    9.9%
    2.9%
    Source: 6291.0.55.001 Table 9

    Information is also available on the number of people working less hours than usual, which may provide an early indication that hours declined due to COVID-19. Chart 4 presents this information as an original data time series for the past 5 years, and shows that there was not a notable change in February 2020.

    Chart 4 : In original terms, total employed who worked less than usual hours
    This chart displays the total number of employed people who worked less than usual hours. The chart shows this number of people over a 5 year time series from Feb 2015 to Feb 2020.
    Source: 6291.0.55.001 Data Cube EM2a

    In the Labour Force Survey, when someone responds that they worked less than their usual hours, they are asked for the reason they worked fewer hours. Table 2 below shows that the most common reason people worked fewer hours in February 2020 was because they were taking annual leave, holidays, flextime or long service leave. 15% of people were working fewer hours than usual due to illness, injury or sick leave, while 17% of people cited standard work arrangements (including shift work) as the reason they were working fewer hours. This is similar to proportions in February 2019.

    Table 2 : Percent of employed who worked fewer hours than usual, February 2020
    Reason worked fewer hours than usualPercentage of Total
    Annual leave, holidays, flextime, or long service leave
    28%
    Standard work arrangements or shift work
    17%
    Own illness or injury or sick leave
    15%
    No work, not enough work available, or stood down
    14%
    Personal reasons, study, caring for sick or injured family
    10%
    Worked fewer hours than usual for other reasons
    6%
    Bad weather or plant breakdown
    5%
    Maternity, paternity or parental leave
    4%
    Began, left or lost a job during the week
    2%
    Source: 6291.0.55.001 Data Cube EM2a

    People affected by the impacts of COVID-19 may fall into the following categories in the coming months:
    • Annual leave/holidays/flextime/long service leave (eg. if a person was self-isolating and taken paid leave);
    • Own illness or injury/sick leave (eg. if a person had cold/flu/COVID-19 symptoms or a confirmed diagnosis and had taken sick leave);
    • Personal reasons/study/caring for sick/injured family (eg. if a person was caring for a family member with COVID-19, or had self-isolated without accessing paid leave);
    • No work/not enough work available, or stood down (eg. if a person had been told not to attend work and they were not receiving any paid leave); and
    • Began, left or lost a job during the week (eg, if a person had changed jobs but had less hours, no longer had a job with an employer, or their business closed down)
    • Worked fewer hours than usual for other reasons (that is, a person may have felt that none of the other categories described their circumstances, and chose to select 'other').
    When analysing these categories for potential COVID-19 impacts in February 2020, no discernible impact was detected. Table 3 below shows the relative monthly movements between January and February for the past six years, for three of these categories. The relative movements in February 2020 were comparable to those seen in February 2019.

    Table 3: Reasons people worked less than usual hours - monthly movements
    Year
    No work, not enough work available, or stood down
    (%)
    Own illness or injury or sick leave
    (%)
    Personal reasons, study, caring for sick or injured family
    (%)
    Feb 2015
    -6.2
    47.1
    48.9
    Feb 2016
    -10.9
    45.0
    69.1
    Feb 2017
    9.7
    65.8
    96.0
    Feb 2018
    10.8
    95.8
    108.7
    Feb 2019
    -11.8
    34.2
    54.9
    Feb 2020
    -14.0
    31.9
    50.3
    Source: 6291.0.55.001 Data Cube EM2a

    Looking at the time series in original terms can provide additional insights into patterns in the labour market over time. For example, Chart 5 shows distinct seasonal patterns related to the cold and flu seasons in Australia in the winter months. If people begin to take sick leave due to COVID-19, then there may be an increase in the number of people in this category and a more pronounced seasonal pattern.

    Chart 5 : Employed who worked less than usual hours - Own illness or injury or sick leave
    This chart displays the total number of employed people who worked less than usual hours due to the reason "own illness or injury or sick leave". The chart shows this number of people over a 5 year time series.
    Source: 6291.0.55.001 Data Cube EM2a

    Quarterly hours worked information

    The quarterly detailed labour Force release (cat. no. 6291.0.55.003) contains hours worked information by industry, occupation, and range of other characteristics.

    Analysis of hours worked data by industry did not identify any unusual movements in February 2020 that would suggest an early indication of COVID-19 impacts.

    Chart 6 below shows the hours worked for three industries which may potentially be impacted early by COVID-19. For each of these three industries, the hours worked series largely reflects existing patterns. This was also true for all industries in February 2020.

    Chart 6 : Number of hours actually worked in all jobs, Original
    This chart displays the total number of hours actually worked in the three industries, "Retail Trade", "Accommodation and Food Services" and "Transport, Postal and Warehousing". The chart displays these total hours actually worked as a time series spanni
    Source: 6291.0.55.003 Table 11

    In future quarters, the ABS will also undertake analysis of reasons people worked less than usual hours by industry or groups of industries (following a similar approach to that taken inTable 3).

    The ABS will also be undertaking analysis of hours worked by industry in the quarterly Labour Account (cat. no. 6150.0.55.003), given it is the best source of industry information. March quarter Labour Account estimates will be released in June 2020.

    Additional data

    Beyond hours worked information, other detailed Labour Force products will also be important in monitoring and understanding impacts to labour force status. This data includes information on:
      • redundancies (quarterly, in Table 29a and 29b);
      • the extent to which people expect to be with an employer for the next 12 months (quarterly, in Table 17);
      • duration of job search (monthly, in Table 14a, 14b, 14c, 14d, 14e); and
      • the gross flows product.
    The ABS will include analysis from some of these in future months.

    Other products beyond the core Labour Force products may also provide important information over the coming months, or assist in understanding the number of people who may be more affected by the labour market impacts of COVID-19. For example Characteristics of Employment (cat. no. 6333.0) contains information regarding employee earnings and working arrangements.