Tag Archives: labour market

Class In Australia: Everything and Nothing?

A book, a red book (of course), simply titled Class in Australia. A front cover emblazoned with Sally McManus proclaiming that it is “a powerful and vibrant study of the complex realities of class in modern Australia”, and a back cover announcing an examination of class rooted in the specifics of Australian settler-colonialism which also takes account of race and gender relations. A big promise from Monash University Publishing about Steven Threadgold and Jessica Gerard’s book which was published in February this year.

Book Cover: Class in Australia, by Steven Threadgold and Jessica Gerard

With this advance advertising, I pre-ordered a copy, but I am afraid I was ultimately disappointed in the purchase. As an edited collection of essays, it is a hard ask to generate a coherent picture of the complexities of class (and that was probably not the aim), but from the opening chapters I was not sure who the audience was for the book.

Much of the work plotted issues or the various authors’ research in relation to existing academic literature, but without a knowledge of that literature it was hard to evaluate the arguments and contributions. But for an academic audience, the short generalist pieces lacked the data and detail to be convincing. My reading was somewhere in between, and I was left wanting more.

Theory

Threadgold and Gerard’s introduction argued for the importance of class as a concept, and against arguments of the “death of class”. They argue that

“class is necessary for understanding how Australian society functions, how the powerful maintain their interests, and how social and cultural institutions work to reproduce inequality”.

No argument from me on that, but they neither define class or a particular approach to class analysis, beyond emphasising the need for an open analysis of the complexity of class which takes account of gender, sexuality, race, ethnicity and the particular context. From that atheoretical (or at least non-structural) starting point I was not sure what “class” meant or was grounded in.

The first chapters designed to “situate class analysis” within the specifics of Australian experience were vague and disconnected – one leaping from Poulantzas to the class contradiction of one working class man’s love of classical music, while another described property relations in settler colonial society, but appearing fairly dated in its sources. The most theoretical of the chapters in this section set out its key definitions and assumptions, and adopted a categorisation of class based on income from paid work for owners (employers and petty bourgeoisie) and labour distinguished by control of operational skills and managerial rights (expert managers, managers, experts, workers). There was data on the numbers of people in each of these classes, and some discussion of the interplay of income, assets and culture. IMHO, it was a too dismissive of housing as a class asset (for reasons discussed here), but in any case, the chapter was too short to develop its key themes and, in an edited collection, this framework did not necessarily apply to other chapters.

Race/Aboriginality

Beyond the early chapter on settler colonial society, there were various references to race and the experience of Aboriginal people, but few were developed. For instance, in the concluding interview, Raewyn Connell contrasts Australian colonialism with South African settler society in that:

“Except in the pastoral industry and especially in Northern Australia, colonialism in Australia did not subject the Indigenous population as a labour force … That produced a different pattern of racism in Australia which we still have elements of today – exclusionary rather than hierarchical.”

This struck me as in important entry point to understanding an intersection of class and race, but I wanted a more detailed analysis of how these geographic differences played out, and how the situation changed over time. In 2016, 51% of Aboriginal and Torres Strait Islander adults were employed. This was still well below the 76% of the non-Indigenous people, but it shows that the exclusion from employment/class is not total. So how are we to understand the class processes and differences for both Aboriginal employees and non-employees?

Similarly, the interview with Larissa Behrendt was a story of exclusion in highlighting the discrimination she has experienced in her career. While her story is inspirational, I was not sure what it says about class that is not simply captured by the notion of discrimination (with “class” being redundant).

Industrial Relations

For me, one of the most interesting chapters was an analysis by Tom Barnes and Jasmine Ali of an industrial dispute over retrenchments in a Woolworth’s warehouse in suburban Melbourne. The analysis adopted Erik Olin Wright’s multilevel synthesis of Marxian, Weberian and Durkheimian theory (which I considered in an earlier post) to show the divisions within the warehouse staff. Wright’s work in fact appeared several times in the book, but Barnes and Ali’s chapter was a great example outlining the (Weberian) distinctions between entitlements of full-time, casual and labour-hire workers and the (Durkheimian) situational differences within the formal and informal workplace culture and hierarchy. This was framed within a Marxian logic of the power of capital in deciding production location and warehouse closure.

In the end, the union got a good outcome (much improved redundancy and rights) based on identifying the unity of class interests against capital. While that may be good news, I would have liked to have known more about how those institutional and work-floor divisions were navigated – i.e. how class was mobilised. The article also said little about race or gender intersections, so while it was a good exposition of Wright’s methodology, it did not fully situate class in the current context.

Conclusion

There is not space here to comment on each chapter of Class in Australia, which was something of a smorgasbord (or at least a tasting tray) of class discussion. Suffice to say that the cultural studies chapters analysing an SBS documentary (Struggle Street) and rural romance novels failed to convince me of the generalisability or importance of the topic. And I did not read the chapters on class and education because …

Throughout the book (and somewhat in contradiction to Threadgold and Gerard’s statement cited above), I got no sense of one (or more) classes accumulating wealth and power from their class position or at the expense of other classes. There was a sense of inequality based on class and hierarchy between and within classes, but not really a sense of exploitation or of class processes as drivers of macro-economic structures or of social change or stability. Rather, (and perhaps because they generally reject a priori theory in favour of class forming in context) class appears as the wash-up of other economic and social processes. This is unsatisfactory both analytically and politically as it robs classes of agency.

There is much to say about class in Australia, and Threadgold and Gerard set out to raise rather than answer questions. But I would argue that the class processes and conflicts which determine (or at least influence) the distribution of income and wealth at the macro-level are more important than the musical tastes, and even the education levels or voting patterns, of the players in those processes. Ultimately, that is why I am drawn to political economy rather than sociology, even while acknowledging the importance of other analysis.

The Gender Wage Share: History and Implications

This post traces the gender wage share and women’s increasing share of the total wage pool in Australia since the mid-1980s. Women as a whole currently earn just 38% of all wage earnings in Australia. This is a product of the aggregate of the gender wage gap and the difference in labour force participation, and by my calculation amounts to a difference of around $200bn a year.

In a previous post I argued that the magnitude of the difference constituted a significant macroeconomic flow with an important role in the reproduction of society – and of gender relations in particular.

The rationale for the use of gender wage share data and the conclusions drawn are set out in that previous post. Here I want to consider the changes over time in those gender economic aggregates, and the implications of those changes for our understanding of inequality.

Time Series

The graph below shows the female share of “total earnings” in the ABS Average Weekly Earnings data from 1984 to the present. This is a simple calculation based on average employment earnings multiplied by the number of workers.

It is important to note that “earnings” here refers only to employment income. In the ABS data it is called “total earnings” because it includes overtime – as opposed to ordinary time earnings (which is also in the data set). However, that should not be confused with a total of all earnings, which could include social security payments, investment income or other mixed income. The World Economic Database now has this all-earnings data for Australia, but the time series is more limited and contains a bold assumption that mixed income is shared in the same proportion as other income. In any case, the results are similar with the data for 2019 showing a female share of 36.6% of total earnings, while my data has the female share at 38%.

Line graph of Female Share of Total Earning 1984 - 28.6% 2020 - 38%

The graph shows three phases in a history of an overall increase in women’s share of the total wage pool over the last 36 years. From the mid-1980s through to 1992, there was a significant increase in women’s share going from 28.6% of wages to 33% of the total wage pool. This was based on a small narrowing (2 percentage points) of the full-time gender wage gap, but a more significant increase in women’s share of jobs – going from 37.9% of employment in 1984 to 42.6% in November 1992.

After 1992, women’s share of the wage pool continued to increase, but at a slower rate until 2012. There is a change in ABS data series here so some data discontinuity, but women’s share of total wages has grown significantly since then from 34.9% of all wage earnings in 2012 to 38.5% in May 2021. This has largely been on the back of a decrease in the full-time gender wage gap (4.5 percentage points) and a more modest (1.9 percentage point) increase in the share of jobs.

These drivers are shown in the following graph which creates indexes showing changes in the gender wage share alongside the proportion of the workforce who are women (participation) and the changes in proportionate remuneration (that is, average female full-time earnings as a proportion of male full-time earnings). This last index is just a different presentation of the commonly-cited gender wage gap.

As can be seen, the increased gender wage share tracks most closely with increased participation. However, between through the 1990s and early 2000s the wage share is dragged down below the participation rate increase by a stagnation of the remuneration gap (evident in the F-T wage proportion) from 1992 to 2007, followed by a widening of this gap after the onset of the global financial crisis in 2007. From 2014 this dynamic largely reversed with the decreasing remuneration gap accelerating the female wage share faster than the increase in participation.

Index of Gender Wage Share, Gender Wage Gap (F/T Ordinary) and Participation (female proportion of jobs).

The overall trend of an increase in women’s share of the earnings is not unique to Australia. The World Inequality Report 2022 data shows that women’s share increased between 1990 and 2020 in most regions of the world (with China being the notable exception).

Implications

This gender wage share data has implications for how we understand and speak about inequality. With the female share of the total Australian wage pool growing significantly and (relatively) steadily since the 1980s we have seen a move towards greater gender equality (at least in terms of labour market incomes). Yet, particularly following Piketty’s work, it is now a fairly standard claim on the left that inequality has increased since the early 1980s.

This claim of increased inequality is certainly true based on the usual measures of household income (the data is well summarised by ACOSS/UNSW), but given the data on the female wage share we need to recognise that claims about increasing inequality are gendered, or at least gender-blind, statements. They are not wrong, but they are privileging particular data and the gender-blind category of the household over other standpoints and data which focus on women’s income.

Or to put it another way, such measurements of increasing inequality are based on (and promote) views of society as households stratified along a continuum, rather than as structured by gender (and other) inequalities.

Further, the gender wage share data puts a different light on the left critique of neoliberal or right-wing labour market policies. The standard argument is that the removal of labour protections, penalty rates and working conditions, and the increasing precariousness of work are likely to impact disproportionately on women who are in the most marginal and disempowered jobs. (See for instance Alison Pennington’s excellent critique of last year’s Industrial Relations Bill).

There is no argument from me with these critiques of the neoliberal reforms of the last 30 years. But what the gender wage data shows is that these neoliberal advances have been counterbalanced and ultimately outweighed by the movement of women into the labour market in greater numbers and some closing of the gender wage gap.

This is not an argument for complacency, but rather to argue for a more nuanced and multi-level analysis of our analysis of inequality. There are other forces beyond neoliberalism which are also shaping economic outcomes.

Caveats and Conclusion

While greater gender equality would (and should) normally be seen as a good thing, the increased female wage share is not unproblematic. Firstly, it should be recognised that this is an increasing share of a proportionately decreasing pie as the labour share of total income has been decreasing over much of the period. (See the Journal of Political Economy’s Special Issue on the Declining Labour Share).

Further, it must also be recognised that, in an era of stagnating wages and increasing cost of living (in particular, rapidly rising housing costs), one of the drivers of increasing female workforce participation is the need for households to have two incomes to stay afloat.

Unequal symbol with words "inequality - not what you think"

Ultimately though, regardless of these caveats on the increasing gender equality story, what the gender wage share data shows is the importance of standpoint and the questions we ask about inequality (and much else). Asking questions about structural inequality (such as gender, class, race, geography) provides different perspectives and different conclusions than the traditional focus on household income – a theme I will return to in future posts about other structural inequalities.

And of course, there is also the sheer size of the $200bn gap in the gender wage share, which is important in its own right (and not visible in the mainstream statistics).

The Gender Wage Share – a $200bn gap

It seems to me to be an important fact that men receive $4bn a week more than women in the Australian labour market. This is the aggregate of the gender wage gap, but at $200bn a year it represents more than a gap in individuals’ earnings. It is a significant macroeconomic flow shaping both the economy and gender relations.

The Gender Wage Share in the Australian Labour Market Men 62% Women 38%

The Gender Wage Gap

The gender wage gap is a common measure of the difference between the average earnings of women and men in the workforce. It is generally expressed as a percentage of men’s earnings based on average total remuneration for full-time workers (a gap of 20.1%), full-time base salary (15%) or full-time average weekly earnings (14.2%) (WGEA data).

These measures are all based on full-time work. This is useful for comparing like-with-like (i.e. full time work) and can be important for highlighting differences in pay rates with women on average in less senior jobs and clustered in low paid jobs and industries. However, the full-time data ignores the actual work of just under half of the female workforce where around 45% of female jobs are part-time. Further, given that only 19% of male jobs are part-time, the focus on full time work centres and normalises male-work patterns and underestimates the differences in wages actually taken home.

These problems are corrected somewhat by calculating the gender wage gap based on the average weekly earnings of all workers. This blows the gender wage gap out to 31.3% because women are not only being paid less by a straight comparison (full-time earnings), they are working fewer hours so their average earnings are lower. Yet this adjustment only goes part of the way to recognising differences in labour force participation because the average weekly earnings do not tell us how many men or women are earning that income.

It would be theoretically possible to have an economy with no gender wage gap, but very few women employed. Indeed, if there were no women employed there would be no gender wage gap!

Alternatively, (and more realistically), women’s participation in the labour force may increase over time, which would be important in terms of the income, independence and economic power of women. However, if that increased participation simply replicates existing patterns then the gender wage gap data would remain unchanged.

The gender wage gap does not tell us how many people are impacted by the gap.

Gender Wage Share

Alongside these disaggregated averages and the expression of the gender wage gap in individual terms, I think it is important to record and express the aggregate outcome of the gender wage gap. This can be done by calculating the share of the total wages pool taken home by men and women. This will be a function of both earnings and participation and is used in the United Nations’ Gender Development Index.

The calculation in the UN index is complicated by the need to ensure international comparability, but in Australia the gender wage share can be calculated fairly easily. Using the ABS Labour Force and Average Weekly Earnings data, it is simply a matter of multiplying the average male wage by the number of male workers and the average female wage by the number of female workers. This gives an aggregate wage pool, and the male and female share of that pool. The table below shows the data for May 2021.

Data table showing female and make no. of employees, average weekly earnings, and difference and share of total wage pool.

There are caveats on this data in broader gender terms given that it relates to labour earnings only, and does not include investment or other income. Nor does it deal with the distribution of non-market production income, or the redistribution of wage income within households, or the range of other inequalities which are tied up in wage gaps.

However, wages are the main source of income for most Australian households, and the data above shows that women take home 38.5% of the total wages pool, a gap of 23 percentage points to the male share – or 11.5 points to an equal share.

But for me, the standout figure is the differential in total earnings of $4bn a week between men and women. That is, the ABS data shows that in any week men as a group earned around $4bn a week more than women as group.

To put that in perspective, it is over $200bn a year – an annual figure which is about the same size as the entire federal government expenditure on social security (Commonwealth Budget Paper No.1, Statement 6, Table 3).

In a recent submission to the Federal Government’s review of the Workplace Gender Equity Agency (WGEA), I contrasted this difference wage share with the annual budget of the agency. Obviously responsibility for addressing the gap in women’s wages and workforce participation does not lie solely with the WGEA, but it seems fairly optimistic to expect an expenditure of $6m to have much impact on a $200bn problem (quite apart from the limitations of the liberalism of the agency – which I did not mention in the submission!).

Macroeconomics

But beyond the individualism of liberal workplace strategies, understanding the quantum of the difference in the gender share of wages is important because of its macroeconomic reproductive role. I noted above that $200bn a year was roughly equivalent to the entire federal government expenditure on social security. It is also around ten times the size of the entire South Australian state budget.

It would seem fairly uncontroversial to say that the social security system and state governments have an important role in total income distribution, and in the stability, direction and reproduction of the economy and society. Could we not then also see that a $200bn a year income differential between men and women is an important income distribution in itself? And that it has a role in the reproduction of society – and specifically the gendered inequalities across society?

For those familiar with structural analysis, this should seem logical. But for those with a more individualist starting point, it may be challenging to posit gender groups as collective economic entities. After all, the social security system or a state government are institutions with formal rules, lines of authorities and someone in charge, whereas “men as a group” or “women as a group” are just collections of individuals making their own decisions with no governing force or collective interest.

However, in macroeconomics we happily talk about a $120bn tourism industry, when it is a myriad of small and large operators competing to maximise their individual business profits. We talk about flows of money to and from a “finance sector” which is a range of entities responding to the savings made available by and the financial needs of individual actors outside of its realm. At a more abstract level, we lionise the disparate decisions of many people into a collective entity called “the market”, so the concept of women or men as a group should not be too much of a stretch.

There are of course differences within gender categories, issues of intersectionality and critiques of such a binary construction of gender. These are big and important issues, but beyond the scope of this post. My point here is simply that if we are to talk about a gender wage gap, then it seems reasonable to also talk about it at a collective and aggregated level.

At this aggregate level, the shares of total wages going to men and to women, and the substantial difference between the two is important because money is not neutral, or simply a conveniently exchange mechanism. It is a store of wealth and an enabler of access to goods and services.

It matters to gender relations and the structures of society that men as a group have access to $200bn a year more cash than women as a group (or that women’s access to that extra money is mediated by men at home or by the state at large). That money enables men as a whole to do more things (or more economically rewarded things) than women, to occupy public space and power, and to command greater aggregate purchasing power to be met by market production.

A gendered labour market not only reflects inequalities, it begets further inequality.

Concluding Directions

I am not claiming that gender differences are limited or reducible to the economic sphere, or determined by economic forces. But I am trying to move away from individualist economics, and in this instance, mainstream theories of the gender wage gap as being a result of differences in human capital or household income maximisation strategies. I am arguing for a more macro-level and structural analysis.

In a future post I may use the gender wage share data to track changes over time (spoiler: the female wage share has increased with greater labour market participation, but tracks a bit differently to the more traditional gender wage gap data). For now though, I am simply putting the $200bn figure out there and challenging anyone to say that an aggregate flow of that amount of money is not important.