History of the monthly Labour Force Survey
Each month the Australian Bureau of Statistics (ABS) releases official estimates of employment and unemployment. These estimates come from the Labour Force Survey (LFS), which is the largest household survey conducted by the ABS. It is the Bureau's longest-running household survey, dating back to 1960 and has provided the basis on which the ABS has built an extensive program of labour and social surveys of the population.
Information about the number of employed and unemployed people, the unemployment rate and other labour statistics are important economic and social indicators. Labour statistics are used for a range of purposes by government and business analysts, academic researchers, employee and employer organisations, and the community.
Reflecting the importance of labour statistics, the ABS has been collecting information on employment and unemployment, together with selected characteristics such as occupation and duration of unemployment, since the early-1900s. Until the 1960s the periodic population censuses provided the most comprehensive information about the labour force. Other sources of information were administrative sources and trade unions. The first regular statistical measure of employment dates from the introduction of payroll tax in 1941. Monthly measures of jobseekers, available from administrative data from the Commonwealth Employment Service (CES), also became available in the 1940s.
The first LFS
Household surveys have a number of advantages over administrative sources - they can cover the whole target population consistently over time; questionnaires can be designed to collect relevant and conceptually sound data; and rigorous sample designs and survey procedures can be applied. In the late-1950s the ABS started to develop a program of household surveys, and, in November 1960, the first LFS was conducted as the first ABS household survey.
The initial LFS was restricted to a sample of households in state capital cities only, and included the non-Indigenous civilian population aged 14 years and over. It was conducted quarterly in February, May, August, and November. Initial publication of LFS estimates was in January 1964.
In February 1964 the LFS was extended to cover all geographical areas of Australia. In August 1966 the scope of the survey was changed to the civilian population aged 15 years and over because of changes to the school leaving age and inclusion of the Indigenous population. The survey continued to be conducted quarterly until November 1977.
Increasing the frequency of the survey
The monthly CES registered jobseeker series, which reflected the administrative needs of the CES, was becoming less useful for measuring the level and movement of unemployment in Australia. In response to an increasing demand for more frequent and better statistics of unemployment, as well as for more comprehensive information on the labour market, the frequency of the LFS was increased to monthly from February 1978. The range of information collected was also substantially expanded in the monthly survey. As a result of the increase in frequency, LFS unemployment statistics were adopted as the official measure of unemployment in Australia.
What it collects
The concepts underpinning the LFS have always been aligned closely with international standards for labour force statistics. Although the survey questionnaire has been modified from time to time, there has been relatively little change in concepts and definitions used, reflecting the stability of international standards over this period. Consequently, statistical series from the LFS have maintained a high degree of consistency over time.
The content of the LFS has expanded progressively over the years since 1960. Much of the increase resulted from changes introduced in February 1978. In addition to information on the number of people employed and unemployed, the survey currently collects a range of other information about the population. For employed people, information is collected on whether they work full time or part time, and their industry, occupation, hours worked and status in employment. For people who are currently unemployed, the survey collects information about whether they are looking for a full-time or part-time job, how long they have been unemployed, and the characteristics of their last job (industry, occupation and reason for leaving). The survey also collects personal characteristics such as sex, age, marital status, relationship in household, participation in school and tertiary education, birthplace and year of arrival in Australia.
In addition to collecting regular information on the labour force, the LFS has proven to be a cost-effective vehicle for collecting less regular information on a range of other topics. The first supplementary topic was included as early as the November 1961 survey. The topics, which have become known as supplementary surveys, have covered a wide range of labour market and other issues relevant to particular sub-groups of the population. Some of the topics relevant in the 1960s, such as labour force experience, school leavers entering work, and multiple job holding, are still considered important for analysing today's labour market.
The number and size of supplementary surveys, and the range of topics covered, increased substantially when the LFS became a monthly survey. Supplementary surveys are now run in all months other than December and January. The topics covered reflect the information requirements for monitoring the social and economic well-being of Australians, and for developing and evaluating government policies and programs. In addition to labour issues, the program also includes a wide range of social and economic topics, such as the environment, crime and safety, and child care. Recurrent labour topics include:
- labour force experience, entering and leaving the labour force, job search experience, retirement
- employment, underemployment, multiple job holding, labour mobility, working arrangements, trade union membership, education and work
- earnings, employee benefits
How it collects information
Ever since the LFS was first conducted, the information collected has been obtained from occupants in selected dwellings by specially trained interviewers. Households are included in the survey for eight consecutive months (eight quarters prior to February 1978), and interviews are conducted over a two-week period each month (over a four-week period each quarter prior to February 1978).
Two main changes have been made to the method of interviewing. Between August 1996 and March 1997, the ABS introduced telephone interviewing into the LFS. More recently, between October 2003 and August 2004, computer assisted interviewing was progressively implemented to replace the traditional 'pen and paper' method.
Main changes in content
During the 44 years the ABS has been conducting the LFS there have been several major changes in content - these are summarised below. A description of all changes, including to collection and sampling methods, concepts, data item definitions, classifications, and time series analysis techniques, is available in Labour Statistics: Concepts, Sources and Methods (6102.0.55.001).
A number of additional questions were included in the LFS questionnaire in February 1975 to ask respondents if they looked for work in the last four weeks (rather than in the last week, as asked previously) and about their availability to start work in the survey reference week. Statistics reflecting these questionnaire changes were first released in May 1976, with data revised back to February 1975.
Substantial changes were made to the LFS when it became a monthly survey in February 1978. The questionnaire was revised, with redesigned question wording, structure and sequencing, and it collected a wider range of information. Only minor adjustments were made to the employed and unemployed definitions.
The LFS remained essentially unchanged until April 1986, when the underlying employment definition was changed to align with changes to international standards made in 1982 by the International Labour Organisation (ILO). The definition of employed persons was changed to include persons who worked without pay between 1 and 14 hours in the survey reference week, in a family business or on a farm (i.e. contributing family workers). Prior to this change, these unpaid workers were counted as employed only if they worked at least 15 hours in the reference week.
In 2001 the LFS questionnaire underwent its first major redesign since 1978. This new questionnaire allowed the collection of new or extended data on: future starters; job tenure; underemployment; hours worked; duration of unemployment; and marginal attachment to the labour force. The new questionnaire also incorporated minor definitional changes to employment and unemployment relating to: short term absences; unavailability due to illness; and contributing family workers. These changes were made principally to improve alignment of the employment and unemployment definitions with ILO standards. Most of these changes were reflected in LFS statistics from April 2001.
- not in the labour force, discouraged jobseekers, marginal attachment to the labour force.