Australian Wages Series - Sources, Methods and Uses


Diane Hutchinson

University of Sydney

Revised June 2016


 Florian Ploeckl

University of Adelaide

  1. Average weekly earnings, 1860/1 to current.


1860/1-1905/6: ‘Composite Estimates of Selected Long Run Labour Series 1860-1984’, in Glenn Withers, Tony Endres and Len Perry, Australian Historical Statistics: Labour Statistics, , ANU Source Papers in Economic History, No 7, 1985, Revised Series, 1993 (supplied by the first author.)

1906/7-1968/9, ‘Composite Estimates of Selected Long Run Labour Series 1860-1984’, in Glenn Withers, Tony Endres and Len Perry, Australian Historical Statistics: Labour Statistics, ANU Source Papers in Economic History, No 7, 1985, pp. 203-5.

1969/70-1995/6: ‘Average weekly earnings, all employees, quarterly’, Reserve Bank of Australia, Historical Statistics, G06: Labour Costs, available at This data was derived from the ABS[1] which is also the source of data for subsequent years:

1996/7-2010/1: ‘Weekly average, all employees, total earnings, annual’, Australian Bureau of Statistics, Australian Economic Indicators, ABS: 1350, Table 6, Table 7.5

2011/2-2014/5: “Average Weekly Earnings, Australia”, ABS 6302, Table 3

Methods and Uses

The foundation for this long run average wages series is the estimates by Withers et. al. which were constructed by estimating gross wages and employees numbers for each year. Employee numbers are an average over the year to 1918, and full-year equivalent employees (part time and full-time) thereafter. Their series to 1899/1900 is derived from wages data for five occupations from different sectors and represents average earnings for all employees. From 1900/01, their series is for average earning for manufacturing employees only. This break in series was made because the data in the manufacturing census returns allowed more accurate estimates for that sector than for the economy as a whole. There also are other less substantial breaks in the series due to changes in data sources. 

Withers et. al.’s source paper provides a full explanation of the sources and methods employed to construct their initial series. The authors subsequently constructed a revised series for the period 1860/1 to 1905/6 which we use here. [2] The main revision was to replace the use of N. G. Butlin’s wage estimates for 1861-1900 with new estimates derived by first estimating wages for five occupations and then weighting them according to their importance in the workforce in each year. Their series is for annual earnings and it is converted here to a weekly amount by dividing by 52, for consistency with later Australian Bureau of Statistics (ABS) data. For the pre-1901 period, this is simply a reversal of the authors’ conversion of some daily and weekly wage rates to an annual amount, by assuming that workers worked a full week every week. It appears from the source paper that the estimates for 1907 to 1918 are for calendar years, but no attempt has been made to convert these to financial years here to match the remainder of the series.

Some users may be concerned about the reliability of average manufacturing earnings as an indicator of wages for the economy as a whole from 1900/1. Manufacturing did employ workers with a wide range of skills, and Withers et. al. note that there was a close correlation between their all-industry and manufacturing series in the overlap years around 1900.[3] We have no direct evidence on which to evaluate whether earnings in manufacturing remained a good indicator for the first half of the twentieth century, but for the period from 1954 the ABS has published indexes of male average earnings in manufacturing and in all sectors. These two indexes show a remarkably similar rate of growth over the 14 year period to 1967/8, at which point the male all-sector index was around 1 per cent higher than the male manufacturing index.[4] New ABS all-sector all-employees average earnings estimates from 1969/70 also show that the level of average earnings across the economy was a little below but comparable with Withers et. al.’s estimate of the level of average earnings in manufacturing in 1969/70.[5]

However from the end of the 1960s, as manufacturing’s share of GDP began to decline and the sector slowly re-structured towards higher value-adding industries, manufacturing earnings did behave differently from the all-industry average. The divergence was most marked in the tumultuous 1970s, but there was also a clear divergence in the period 2003 to 2006 when, on occasions, the quarterly estimate of manufacturing earnings was almost 20 per cent higher than the all-sector average. The gap has subsequently narrowed, and average earnings in manufacturing in 2011 was around 10 per cent higher than the all-sector measure.[6] At the cost of creating a major break in our series in 1969/70, we decided it was advisable to switch to the new ABS estimates of average earnings for all employees in all sectors, described below. The Withers et. el. manufacturing series for 1860/1-1968/9 was re-based using a ratio of .9337, to link to these ABS estimates.

The ABS all employees’ average earnings data for the years 1969/70-1995/6 can be found in the Reserve Bank’s collection of on-line historical statistics, and thereafter directly from the ABS web-site.[7] The Reserve Bank data for 1969/70-1995/6 is quarterly and an average of the four quarters of the financial year is used here. Earnings estimates to 1981 were derived primarily from payroll tax returns so are not strictly comparable with those derived from surveys from 1981. There were also changes in definitions that create a break in the series in 1981.

Our final average earnings series complied from the above sources is best thought of as an estimate of employers’ average direct wages and salaries costs per worker; of course, it is also an estimate of the average gross earnings of workers, but it should not be interpreted as representing the earnings of the average worker, and particularly not as representing changes in how the average worker fared, because the average worker in this series is a statistical artifact – part adult and part junior, part male and part female, part skilled and part unskilled, part full-time and part part-time - a weighted average of all the components of the employed workforce. The composition of the workforce, and therefore the weighting of each component in determining the level of average earnings, has undergone substantial changes over the 150 years covered by this series. Some of these changes, such as the increase in relative importance of adult workers, were gradual but there have been two major, well-documented changes in workforce composition since 1950. The first was the increase in female participation. Women comprised just under 20 per cent of the workforce in 1911; by 1954 this had only increased a little to just under 23 per cent but thereafter female participation rose rapidly and stood at 36 per cent in 1976.[8] Since average female wages were lower than average male wages, even after state and commonwealth equal pay decisions in the late 1960s and early 1970s[9], this rapid growth in female participation has clearly impacted on the level of average earnings. The second major recent change in workforce composition is the increase in the proportion of part-time employees. We do not know how many employees worked part-time in the early twentieth century but in 1964 part-time workers comprised less than 10% of adult employees. After slow initial growth, the proportion started to rise rapidly from the late 1970s until, by 2011, part-time workers comprised almost 30% of adult employees.[10] Some were part-time workers by choice but their ranks were swelled by those who would have preferred substantially more hours of work per week, and who are classed as under-employed.[11] The speed and magnitude of the growth in part-time employment from the late 1970s, like the growth in female participation from the 1950s, has had a substantial impact on the average weekly earnings series.

  1. Average weekly compensation per employee, 1949/50 to current.


1949/50-1970-71: ‘Wages’, Reserve Bank of Australia, Australian Economic Statistics 1949-50 to 1978-79, Occasional Paper 8A, 1980, Table 4.16, p. 103

1971/2 to current: Australian Bureau of Statistics, ABS 5206, Australian National Accounts: National Income, Expenditure and Product, Table 34, Key Aggregates and analytical series, Annual, (Series ID: A2304919R), Non-farm ; Average non-farm compensation per employee: Current prices ;

Methods and Uses:

There is a break in this series in 1960/1, with the switch from all sector average compensation to the non-farm sector average compensation. Non-farm compensation estimates are not available prior to 1960/1, but it is considered the more reliable indicator because of the difficulty in determining the role of family members (employee or entrepreneurial?) on family farms.

This series provides a broader, more inclusive indicator than the average earnings series. The wage supplements such as employer contributions to superannuation and long service leave have become an increasingly important component of employee compensation since 1950 but it is notable that, even in 1950, the average compensation series was well above the average earnings series. Which series is most useful depends on the user’s purpose. For users interested in the average costs of employing a worker since 1949/50, this more inclusive average compensation series is a better indicator. It may not be so relevant for users interested in employees’ average spending power, because employees do not have discretion over some supplements, and even with superannuation they do not derive immediate spending power. Yet average earnings is a gross measure from which taxation is deducted, funding government services over which employees also have no direct discretion. As with average earnings, this series should be used as an indicator of changes in average compensation per employee, rather than changes in the compensation of the average employee, because this series is also affected by changes in the composition of the workforce.

  1. National minimum wage, weekly, 1907-1982; 1997 to current.


1907-1982: Withers, op.cit., ‘Commonwealth Basic and Minimum Wages, 1907-1981, pp.138-9

1997-2000:, p.45

2000-2011: Fair Work Australia, Statistical Report- Annual Wage Review 2011-12, Table 8, p. 27

2012-2015: Fair Work Australia, National Minimum wage Orders

Methods and Uses:

This series presents the official, widely publicised national minimum wage rates for adult men to 1974 and for all adults thereafter. Unlike the other Australian Measuring Worth series, this is a calendar year series. To 1922 it is an average for the year; thereafter it is the minimum wage rate in force at, or near, the end of the calendar year. We made the decision not to convert this series to financial years by averaging as this would dilute the immediate effect of some important minimum wage decisions. The most problematic issue raised by the minimum wage rate series is the extent to which these rates should be considered a universal wages floor, or whether some workers were paid less than this minimum. Since this is central to the use of this series, it is discussed in detail below, as is the gap in the series from 1893 to 1996.


The first national minimum wage emerged from a broader system of wage regulation which developed in the 1890s with the creation in most Australian colonies of wages boards with the power to determine or ‘award’ wage rates. With federation, the Commonwealth Court of Conciliation and Arbitration was also established to deal with disputes and determine award wages for the Federal jurisdiction. It was in the Commonwealth Court, in the 1907 Harvester case, that Justice Higgins laid the foundation for a needs-based national minimum award wage. He defined a ‘living wage’ for unskilled males as the amount sufficient to meet the normal needs of the average (male) employee, regarded as a human being living in a civilised society, and determined that 7 shillings a day, or 42 shillings for a 6 day week, was the appropriate amount to meet this criterion. The average male employee was defined as one who provided for a wife and three children. 

Higgins was not the first Australian wage regulator to make reference to living costs in making award wage determinations but he was the first to explicitly award a ‘living wage’.[12] The Act under which Higgins made this determination was subsequently ruled unconstitutional but he applied the principle of a needs-based wage to other unskilled wage cases. Other Federal and state wage regulators followed his lead, and the ‘living wage’ became the cornerstone of Australia’s system of wage regulation. It was referred to as the ‘basic wage’, and formed the base rate to which skill loadings and allowances were added to determine male award wage rates for more skilled or difficult jobs. The female basic wage was defined as a proportion of the male basic wage[13] until the 1974 equal pay case when a single adult minimum award wage was introduced. 

The wage regulators’ method of adjusting for price changes initially was ad hoc and, even after a price index was used from 1914, adjustments to the wage usually lagged prices until automatic quarterly cost of living adjustments were incorporated in 1922. Colin Forster found that the Harvester 7 shillings (adjusted for inflation) did not become the standard unskilled award wage in both federal and state jurisdictions until 1921.[14] Even thereafter, it is important not to exaggerate the impact of ‘needs’ in determining the level of the basic wage. On a number of occasions the basic wage was increased by more (or less) than was necessary to maintain its buying power, on the grounds of the economy’s capacity to pay more (or less). Then, in 1953, indexation was abandoned although cost of living, along with capacity to pay, remained the primary criteria in determining adjustments.[15]

Perhaps more important for users of this series is the question of coverage, or the extent to which this minimum award wage set a wages floor for all unskilled male workers. Award wages were not mandatory, and not all unskilled male workers were paid the minimum award, although the reasons for deviations from the award differed over time. Initially there was the issue of jurisdiction – most industries and occupations came under the state systems of wage regulation which took time to converge to the Commonwealth Court’s award determinations. Additionally, the award wage system was a union-mediated system: unions had to apply to the relevant jurisdiction for an award for their members’ occupation or industry, and then ensure it was implemented in its members’ workplaces. Union membership was relatively low in 1906, but escalated thereafter.[16] As more applications for awards were made, it seems likely that many more workplaces were subject to the award system by the 1920s and therefore that the national minimum male award wage set a floor for unskilled male wages in these workplaces. In fact, Colin Forster argues that the only industries not covered by awards in the 1920s were rural ones; in urban areas he also argues that militant unions ensured there was little evasion of award rates, concluding that the minimum wage was an almost universally standard for unskilled urban workers, with the exception of the aged and infirm. Forster’s limited evidence and arguments, however, seem tailored to larger establishments. Despite rapid growth in membership, only approximately half the workforce belonged to a union in the 1920s. Small establishments were common in some industries and those with a non-unionised workforce – retail shops employing one or two assistants or carriers employing one or two labourers - might well have paid less than the minimum award wage, especially in times of high unemployment. And there is ample evidence of relatively high unemployment among the unskilled, and especially those unable to do heavy work, both before 1915 in the 1920s, while unemployment touched many more workers with the onset of, and slow recovery from, the 1930s depression.[17] In view of evidence of non-compliance with the current mandatory minimum rate discussed below, it seems sensible for users to assume that at least some unskilled adult workers were also paid less than the national minimum award wage prior to the 1939.

A related issue is the frequency of work, and payment, for the unskilled.[18] The ‘living wage’ or the national minimum award wage was introduced in a context where many workers, and most especially the unskilled, faced the risk of intermittent work. Under-employment is a modern concept but it was a fact of life for many in the nineteenth century, although the form of underemployment differed from that faced today. In the nineteenth century some workers were engaged on an hourly or daily basis with no assurance of a full week’s work, let alone a full year’s work; even those engaged on a weekly basis faced the risk of under-employment due to a reduction in the number of working days, known as short time, or temporary shutdowns. Jenny Lee and Charles Fahey identified widespread sources of intermittent employment in the nineteenth century, even in the boom years of the 1880s; wages books assembled by Fahey show that that short time and intermittent employment continued into the early twentieth century.[19] Both identify the unskilled as the most vulnerable but Fahey’s analysis of surviving wages books demonstrates this most dramatically with highly skilled workers more likely to be retained, albeit on short-time, compared to their unskilled counterparts. 

Higgins’ ‘living wage’ was a product of this time. It specified a daily wage rate, along with the maximum hours per week before overtime rates were payable. This daily minimum award wage rate is conventionally multiplied by six to impute a weekly full-time wage, as we have done in this series, but this does not imply that workers had the security of a full week’s work. Higgins in fact acknowledged the risk of intermittent employment by including in workers’ living costs the need to make provision for hard times. As the award wage system spread, many other awards also specified daily rates. NSW awards for many blue collar, male-dominated industries and occupations were still daily, or sometimes hourly, rates of pay in the late 1920s.[20] Of course those with weekly award rates were not immune from short-time work, but the widespread acceptance of daily award rates is indicative of the nature of the employment relationship. Occupations such as wharf laboring where employment was universally organized on a casual basis were accorded a higher hourly or daily rate to compensate for intermittent under-employment. Yet, as the Development and Migration Commission’s Report on Unemployment in 1928 noted, for unskilled workers in many other industries the real employment relationship was also essentially a casual one.[21]

There is considerable evidence that under-employment also remained a reality for some in the 1920s. The Development and Migration Commission’s Report on Unemployment in 1928 is especially revealing. It drew heavily on the only continuous source of data on unemployment - the quarterly reports by some trade unions to the Commonwealth Statistician on their membership and the number of members unemployed. [22] The definition of unemployed in this data – those without work for at least three of the six working days in the reporting week – includes both those with no work who would be defined as unemployed today, as well as well as those who were substantially under-employed. The report identified long term sources of this unemployment such as the contraction of the NSW mining industry.[23] It also identified cyclical influences on unemployment, most especially the downturns in 1921-2 and 1927-8. But the report focused on intermittent location and industry-specific unemployment. The extent of this type of unemployment was clearly demonstrated by the a wide differences between states in unemployment rates for the same industry, and wide difference between industries within the same state, as well as and by the substantial changes in the level of unemployment between quarters. The report offered many of the same explanations for this intermittent unemployment as those identified by Lee and Fahey for the nineteenth century – seasonal or weather-related variations in production affecting demand for labour in rural and construction industries and related service industries as well as in manufacturing industries such as canning, fertilisers, agricultural machinery and building materials. The report also highlighted the relatively high incidence of intermittent unemployment in manufacturing more generally, which it attributed in part to small or poorly managed or marginal manufacturing establishments which produced a vast range of products to order, resulting in erratic bouts of short time work or temporary shutdowns. The NSW manufacturing census confirms the relatively high rate of total shutdowns in manufacturing - establishments accounting for over ten per cent of manufacturing employment shut down for one or more months over the 1927-28 financial year.[24] While these shutdowns affected all workers, the Development and Migration Commission Report concluded that the unskilled still bore a disproportionate share of the short time and employment cut backs in the 1920s. Those unable to do heavy unskilled work were the most vulnerable. [25] With the severe downturn in the late 1920s and slow recovery in the 1930s, it seems firms adjusted by cutting back employment rather than by further reductions in hours - Gregory et. al. concluded that the aggregate data, especially for manufacturing, showed little evidence of a decline in hours of work.[26] The broad conclusion users should draw then is that, even in workplaces where the minimum wage did set a daily wage floor for the unskilled to 1939, an unknown but certainly non-trivial proportion of those unskilled workers who had jobs received less than a full weeks’ work each week at this rate. In short, the weekly minimum award wage in our series is likely to have remained an aspiration rather than an income floor for some – perhaps most - unskilled workers in the 1920s and 1930s.

By contrast, in the full-employment environment of the 1950s and 1960s, those seeking full-time work were more likely to have been able to find it, and unskilled workers were also likely to have been paid more than the minimum award wage. For those whose workplaces were subject to award wages, there were industry-specific award allowances (dirt, danger and shift allowances) or industry- specific productivity-based award wage increases, or over-award base-rate wages. Those in workplaces were wages were not determined directly by awards were also likely to have benefited, since the widespread publicity given to minimum wage decisions ensured all workers knew the wages floor they could expect by moving to an award wage workplace.

In 1967, the concept of a ‘basic wage’ was abandoned when a new system of determining total wages for each award occupation was introduced. A male and female minimum award wage was retained, at the unions’ request, until 1974 when they were replaced by an adult minimum award wage’, but adjustment of the minimum wage rate became an ad hoc process.[27] From the mid-1970s as policy makers grappled to deal with high inflation and rising unemployment, the minimum award wage gradually became, first a parsimonious wages floor, and finally an irrelevancy. National wages cases from 1975 awarded cost of living adjustments for all awards, but these usually provided only partial compensation for cost of living increases. Specific occupations and industries gained additional award wage increases on the grounds of work value, productivity gains or structural efficiency, but those still on the national minimum award wage, and most likely those in non-award workplaces, started to fall behind.


As labour market policy and regulation focused on workplace reform and award re-structuring, including establishing a minimum rate for each award, the national minimum award wage (along with the n.e.c. classification in the metals industry award to which it was tied) became defunct. The Reserve Bank dates this from 1983 when its historical statistical tables first define the minimum award wage as “n.a.” or not applicable.[28] Some researchers have used the minimum award rate for unskilled metal workers as an alternative, since this also provided the foundation for determining new awards for unskilled workers in other industries. However this award did not have the same standing as a national minimum award. As researchers have increasingly recognised, there are workplaces where wages are not determined by awards or collective bargaining, but national minimum awards do set a benchmark and put pressure on employers to ‘do the right thing’.[29] In the absence of a national minimum award wage from 1983 to 1996, we have elected to follow the Reserve Bank and record no values for our series for these years. Below we do present an implied national minimum award wage from 1983 to 1996, derived by adding national wage increases and, from 1993 national safety net increases, to the 1982 national minimum award rate. It could be thought of as the unskilled wage prevailing in a hypothetical workplace that paid the publicly announced rates - the national minimum wage in 1982 with adjustments thereafter for national wage cases and safety net reviews. 

Table 1: Imputed Minimum Wage


AIRC national wage case decisions,

See also































1997 to current

From 1993 there were fundamental changes in Australia’s system of wage determination, with a shift away from collective (industry or occupation-wide and union-mediated) bargaining with award determinations, to negotiated enterprise level agreements. This shift raised important concerns about the relatively powerless position of some workers in making negotiated agreements – concerns that were magnified when it was found that low wage workers remained disproportionately reliant on the award system despite a deliberate policy of minimising award wage increases to encourage the shift to negotiated agreements.[30] In 1997, after the peak union body, social welfare agencies and others made submissions to the 1997 safety net review, the federal wage regulator, then called the Australian Industrial Relations Commission (AIRC) introduced a new federal minimum wage for full-time adult employees. This wage was later re-named the national minimum adult wage, and the wage rate is subject to annual review.[31] 

Like the earlier living wage, the new national minimum wage was seen as providing a living wage – one that would meet the needs of low paid workers - but there were some important differences. With the new minimum wage, the AIRC did not attempt to estimate the cost of providing those needs; additionally, the new minimum was only intended to support the worker, with redress for the cost of dependents provided through various social security mechanisms. The most important difference was that the new national minimum wage was mandatory, and applied to workplaces outside both the award system and the new enterprise bargaining system. As a result users may have more confidence that this new national minimum wage provided an effective floor to wages for all unskilled workers from 1997. However, there are two important qualifications. First, recent evidence indicates that even a mandatory minimum has not ensured compliance, with 5,600 substantiated complaints of underpayment in 2007-8.[32] And second, almost a quarter of the growing number of part-time workers are classed as under-employed, or unable to obtain sufficient work.[33] In some respects, this represents a return to the pre-1939 pattern of work but it has taken a more modern form, with specific jobs designated as casual or permanent part-time.[34] For many unable to get a job with longer hours, under-employment has also been relatively long term, with a median period of under-employment of 30 weeks.[35] While the minimum wage provides a wages floor for these workers too, it is a pro-rata wages floor which inevitably delivers the under-employed a lower income than the weekly minimum wage in our series.

[1] The Australian Bureau of Statistics (ABS) had a number of earlier names but where the data is on the current ABS web-site and has a current ABS series code, the name ABS is used here to avoid confusion.

[2] Construction of this series is discussed briefly in David Pope and Glenn Withers, ‘Wage Effects of Immigration in Late Nineteenth Century Australia’, in T. J. Hatton and J. G. Williamson eds., Migration and the International Labour Market (London,1994).

[3] Their series used here can also be compared with their nineteenth century series for average earnings in manufacturing presented in Wray Vamplew (ed.), Australians Historical Statistics, (Collingwood, Victoria: 1988), Table LAB181-185, p. 160

[4] With both manufacturing and the all-industry index starting from a base of 100 in 1954, the male manufacturing average earnings index rose to 195.6 in 1967/8, while the all industry index rose to 197, a difference of less than 1%. See ABS 6302: Average weekly earnings, December 1968, but see also the index published in December 1970, which shows the beginning of a divergence discussed below.

[5] The difference was approximately 7 per cent.

[6] See ABS 6302, Table 10I, Average weekly earnings, industry, Australia, persons, total earnings.

[7] In the 1960s, the official earnings measure was male-equivalent earnings, and the government statistician repeatedly stated it was not possible to calculate all employee earnings, but a revision to the method of constructing the male-equivalent measure in 1970 may have generated data that allowed retrospective estimation once the ABS adopted an all-employee indicator.

[8] See Vamplew, Australians, op. cit, LAB15-32 for male and female workforce numbers.

[9] See ABS 6302 Table 3 for a comparison of adult men and women ‘s full-time average wages since 1994.

[10] Reserve Bank of Australia, ‘Employment by Full-time and Part-time Status by Sex’, Australian Economic Statistics, Occasional Paper 8A, 1949-50 to 1980-81, 1981, Table 4.8, p. 97; Australian Bureau of Statistics, 2012 Labour Statistics in Brief, ABS6104.0, 2012.

[11] While there is no broad-based data on average hours prior to the 1960s, there is wealth of more indirect evidence indicating that under-employment also was relatively common to 1939. Since the unskilled, and especially those unable to do heavy work, were most vulnerable, this issue is discussed more fully in the final section of this document dealing with the implications of a minimum daily award wage.

[12] See David Plowman, ‘Protecting the Low Income Earner: Minimum wage determination in Australia’, Economic and Labour Relations Review, 6(2), 1995: 253-286 for discussion.

[13] That proportion was initially at 54% of the male rate, and was increased to 75% in 1949.

[14] Colin Forster, ‘An Economic Consequence of Mr Justice Higgins’, Australian Economic History Review, 25(2), 1985: 95-111.

[15] For an overview of these decisions, see Plowman, op.cit., and also Sambit Bhattacharyya and Timothy Hatton, ‘Australian Unemployment in the Long Run, 1903-2007’, Economic Record, 87 (277), 2011: 202-220 for a model that incorporates the tension between these two principles of award wage determination. Users can also see the effect of these two principles by using the Measuring Worth calculator to convert the minimum wage series to 1907 prices, but note that our CPI is a revised version of the initially rudimentary price indexes used by the wage regulators, so the results are a little different from those the regulators intended.

[16] Union membership was estimated at 175,500 in 1906 and, with more reliable measures, stood at 433,200 in 1912 but this increased to 703,000 by 1921. See Withers, op. cit., p. 172

[17] See Charles Fahey, ‘The Aristocracy of labour in Victoria, 1881-1911, Australian Historical Studies, 26, April 1994, pp. 77-96; Forster op.cit.; and Australia, Development and Migration Commission, Report on Unemployment and Business Stability in Australia, Government Printer, Melbourne, 1928 which is discussed below.

[18] A separate but related issue is the question of whether wage regulation increased unemployment by setting wages above the level justified by productivity. See Forster, op. cit., and Bhattacharyya, op. cit., for a recent and more long run analysis. See also See David Preetz, ‘The Safety Net, Bargaining and the Role of the Australian Industrial Relations Commission’, Journal of Industrial Relations, December 1998: 533-553

[19] See Jenny Lee and Charles Fahey, ‘A Boom for Whom? Some developments in the Australian labour market. 1870-1890’, Labour History, 50 (1986): 1-26; Fahey, op. cit.

[20] See New South Wales Statistical Register, 1927-28: Award Wages.

[21] Development and Migration Commission, op. cit, p. 28

[22] Returns were not required from unions whose members were considered to be in permanent employment and those whose members were clearly casual employees, and so cover only about half the unionized workforce or about a quarter of the total workforce. Unemployed numbers include those who could not find (more) work and those who could not work more than three days due to illness, accident or other cause (excluding industrial disputes). The cut-off of three or more days without work to qualify as unemployed in itself indicates a society in which more moderate under-employment was considered a normal risk.

[23] This provides the context for Sheilah Gray’s study of Newcastle which shows work for many was intermittent in the 1920s.Sheilah Gray, ‘An Evil Long Endured: Newcastle’s depression’, in J. Macinolty ed., The Wasted Years( Sydney, 1981)

[24] See New South Wales Statistical Register, 1928: Factories and Works - No. 18, Time worked in each class of industry, 1927-28. For instance construction, ship-building and wharf labouring’ awards were hourly rates. Most awards for the stone, glass and clay industries, and those for metal processing and simple fabrication industries were hourly or daily rates, as were those for water and sewage workers. Awards in male-dominated food and drink industries such as meat processing and in heavy textile industries such as sail-making were also daily, while mining awards were a mix of tonnage rates and daily rates.

[25] See also Forster, op. cit. on the relatively high rate of unemployment among unskilled workers in the 1920s.

[26] See R. G. Gregory et. al., ‘Sharing the Burden; the Australian labour market during the 1930s, in R. G. Gregory and N. G. Butlin eds., Recovery from Depression: Australian and the world economy in the 1930s (CUP). They do note that hours of work were cut on the railways. This was an industry for which unions had not been required to report unemployment and underemployment in the 1920s because workers were considered to be in permanent full-time employment.

[27] See Plowman, op. cit. for an overview.

[28] See

[29] See Peetz, op. cit., for example.

[30] See John Buchanan and Ron Callus, ‘Efficiency and Equity at Work? The Need for labour market regulation in Australia’, Journal of Industrial Relations, December 1993: 515-537, for an early example. 

[31] The new federal adult minimum wage was set at $359.40 – well in excess of the defunct minimum wage shown in Table 1 which, with the 1997 safety net adjustment, would have stood at $260.08. Minima for youth and disabled adults have also been introduced.

[32] This was in addition to other monetary breaches such as underpayment of overtime or penalty rates . See Lucy Neims et. al., ‘Employees earning below the Federal Minimum Wage: Review of data, characteristics and potential explanatory factors’, Fair work Australia, Research Report 3/2011. Two industries - retailing and accommodation and food services – recorded respectively 19 per cent and 18 per cent of the wage rate breeches, and both were also over-represented among those earning below or just above the minimum wage.

[33] Australian Bureau of Statistics, Underemployed Workers, Australia, ABS: 6265.0, Summary of Findings, September 2011, released 2/3/12. See also John Buchanan et. al., ‘Changing employment portfolios and inclusive growth in Australia: redistributing risks at work’ in Paul Smythe ed., Inclusive Growth in Australia (Allen and Unwin, forthcoming) for analysis of the origins and implications of this recent increase in under-employment.

[34] The main distinction between these two classifications is that permanent part-time workers have leave entitlements while casual workers are paid a 20 per cent loading. 

[35] Australian Bureau of Statistics, Underemployed Workers, Australia, ABS: 6265.0, Summary of Findings, September 2011, released 2/3/12.,

Please read our Note on Data Revisions.

Copyright Notice

Copyright © 2010 MeasuringWorth. All rights reserved. This work may be copied for non-profit educational uses if proper credit is given. For other permission, please contact