Behind the chaste veil

Inequality in Australia, post-Piketty
Christopher Sheil

Advance Australia fair? What to do about rising inequality in Australia is a new report, launched in Parliament House in Canberra on 11 June. Nominally at least, it’s also Australia’s first report on inequality since the publication of Thomas Piketty’s Capital in the Twenty-First Century (Belknap/Harvard, 2014). And indeed, Advance Australia fair? acknowledges Piketty’s title, which will be abbreviated here as Capital21. In this review, I wish take the report as an opportunity to consider the field of Australian study on inequality next to the kind of analysis that Piketty’s Capital21 implies, particularly the inequality of the distribution of wealth. But first, Advance Australia fair? must be described and ought be acknowledged in its own right.

Advance Australia fair?

For a start, the coalition behind Advance Australia fair? is of note. The report was inspired by Andrew Leigh’s Battlers and billionaires: the story of inequality in Australia (Black Inc., 2013), and is the outcome of a collaboration between the Australia Institute, the National Centre for Epidemiology and Population Health at the ANU, and a well-heeled looking group with the rather Piketty-esque name of Australia21. Based in Canberra, Australia21’s website says that it’s a body of ‘leading thinkers from a range of disciplines with diverse executive leadership experience in the public and private sectors’, and directs ‘a not for profit research company which specialises in addressing some of the difficult issues facing Australia’. 

With this institutional weight, the report draws on a roundtable discussion held in Canberra in January this year, which was co-funded by the ACT Labor government, the Social Justice Fund, and the Reichstein Foundation. Leigh hosted the event, which featured 34 invited participants, who considered a discussion paper jointly authored by Sharon Friel, professor of health equity at the ANU, and Richard Denniss, head of the Australia Institute. The participants included reps from the sponsoring bodies, the leading economist Bob Gregory, Laura Tingle from the Australian Financial Review, researchers in the field, and an eclectic variety of other thinkers and specialists, plus people from community organisations, including trade unions. All federal politicians were invited, but only a small group responded, apart from Leigh; namely, Jenny Macklin, Alannah MacTiernan and Andrew Giles from the ALP and Lee Rhiannon from the Greens. Submissions were received from other distinguished personages, including Rosses Garnaut and Gittins.

Advance Australia fair? is thus the report following the discussion paper, as workshopped by the roundtable in light of current research. Listed with Friel and Denniss as authors are Bob Douglas and David Morawetz from Australia21. The report has two forewords, the opener by Paul Barrett, chair of Australia21, the other by Denniss, and both put their emphasis on raising awareness of rising inequality, stirring up debate, and encouraging policy action. They also place the work in the context of the growing worldwide concern over the issue, as is of course exemplified by the blockbuster success of Capital21. As Barrett writes, ‘already in 2014, the Davos Conference, United States President Barack Obama and Christine Lagarde, Managing Director of the International Monetary Fund, have identified inequality as a major risk to the pace and stability of future social and economic growth’.

Emblematic of the broad concern, Advance Australia fair? was launched by John Hewson, who took the moment as an opportunity to roundly criticise the Abbott government’s May budget for its ‘obvious inequity’, warned that ‘the land of the fair go’ was disappearing, and called for an ‘inequality impact statement’ to accompany future policy proposals considered by the federal cabinet. The former Liberal Party leader, once such a hell-bent neo-liberal that he was famously dubbed the ‘feral abacus’ by Paul Keating, warned that inequality is having adverse effects on the nation’s trust, self-image, social mobility, productivity and economic growth.

So far, so impressive. What of the content? The Pope, Oxfam and Lagarde feature in the report’s epigraphs, with Gandhi and Mandela appearing later as inspirations. The work itself draws on Nobel laureates Amartya Sen, Joseph Stiglitz and James Heckman, plus The Economist and venerable institutions such as the ABS, the OECD, the World Bank, and the World Health Organization. Will Hutton is quoted with approval, as is, surprisingly, Rupert Murdoch (for endorsing education as ‘the first civil right of any decent society’). The imperilled Gonski report on school education is often mentioned. Books cited include The spirit level (Penguin, 2012) by R. Wilkinson and K. Pickett and The status syndrome (Holt, 2004) by M. Marmot.

Perhaps the best one-word description of the report is useful. The work is divided into two parts. Part A is the collective report, comprising 24 pages that proceed through sections on definitions of inequality (one page); the current Australian position (two pages); international comparisons (one page, which includes a chart of income inequality from Capital21); the causes of inequality (six pages, including a page on Capital21); the adverse consequences of inequality (one page); the benefits of reducing inequality (four pages); and policies to arrest the present trend (nine pages). The allocation of one page for defining inequality and nine for prescribing ways and means of bringing about a more egalitarian society gives an indication of the report’s general can-do approach to the issue.  

Those who have been following the debate about inequality will find much that is familiar. The causes of inequality, for instance, are listed as the ‘dominant economic paradigm’, the inadequate social security and tax systems, and educational inequalities. The benefits of reducing inequality are summed up as improved wellbeing, more social mobility, and a more efficient economy. The report’s top ten ways to ‘advance Australia fair’ are to promote a national conversation on inequality, improve taxation, implement Gonski, invest early in child development, set social security payments above the poverty line and index them to average wages, create jobs, foster co-operatives, adopt the recommendations of the World Health Organization, charge the Productivity Commission with conducting an inquiry, and establish a national research program to monitor and evaluate progress.

Part B is about the same length and comprises 23 individual comments by participants and invitees, which defy ready summary. Several are shaped by the pre-budget debate. Garnaut observes that there is ‘considerable momentum behind weakening the features of Australian policy that have muted the tendency toward greater inequality in the past’. Denniss bemoans the quality of public debate (‘There are no rules. There is no blow that is too low. There is no sanction for lying or character assassination’). Leigh pitches for preserving the ‘golden thread’ of Australia’s egalitarian ethos. Gittens debunks the alleged conflict with efficiency and finds the case for social equity to be ‘fundamentally a moral one’, as does Giles. Gregory sets out an approach to raising revenue. The ACTU’s Matt Cowill highlights the inadequacy of Australian social security. Others emphasise employment and health programs. The relationship with climate change is raised, as are the special needs of children, the young, the elderly, and the indigenous.          

In conclusion, it would be wrong to expect material to be explored in depth in a multi-authored summary document of this kind, which chiefly aims to attract public attention to the issues. The authors are generally keen to take up the arguments about the relationship between inequality and ill-health and economic growth. The inequitable tax breaks enjoyed by the rich get a good airing (especially the breaks for superannuation). The authors are rightly annoyed by retorts in favour of the status quo that dismiss concerns about social equity as expressions of ‘social envy’ or ‘class warfare’. Of late, inequality has been rising at what could be a dangerous rate, such that ‘a country that prides itself on being “the land of the fair go” should be asking how much inequality is enough’, as Barrett says in his foreword, ‘and how much is too much?’ Conservative apologists for the prevailing order are not so much interested in the answer as intent on disallowing the question. The authors, contributors and publishers are therefore to be congratulated for taking a lead in public debate and providing a useful resource. The only glaring omission in both the report’s parts is women, who remain unequal to men in their incomes and wealth, an omission that will embarrass the authors on reflection and is perhaps a sign that the work was rushed in the end to greet the budget. 

Post-Piketty

For the purpose here, Advance Australia fair? can be taken as a proxy for the current state of our knowledge about inequality. No doubt there are many other specialists and organisations that could have further augmented the work, but the coalition of institutions and individuals is so broad and well qualified that we can assume that the report has picked the low-hanging fruit in the field and more besides.

What, if anything, then, now needs to be done to bring everything up to scratch in light of Piketty’s work? Here I will question the data on the distribution of wealth. My more general point is that Advance Australia fair? has generously acknowledged but not integrated Capital21. This is neither surprising nor a criticism; no one can have studied in depth the 577 pages of text, 76 pages of notes, and 97-page online appendix of supporting data in the few short months since Piketty’s book was published. At the same time, the conjunction of the two works presents a choice opportunity to reflect on the differences and consider directions that might be useful for all of us interested in advancing our understanding of inequality.

Before we look at wealth, it is in order to note that the data on incomes are in good shape. This is no accident, for Piketty's data on Australian incomes rely on the work of our own Andrew Leigh and his Oxford colleague, Anthony Atkinson — and Atkinson is, in turn, one of Piketty’s most influential colleagues, i.e., with respect to income inequality, Australia is in the contemporary research loop. As to the current income distribution, suffice it to say here that Leigh’s research shows that our upper class is pulling away from the rest of us at a fast clip. The social order is changing, not as dramatically as in the United States, but more than in most comparable countries, which renders the protest over the May budget easy to understand. The incomes of 90 per cent of the citizenry are falling behind those of the top 10 per cent, and even the next 9 per cent might feel they are being cheated by the top 1 percenters. The budget fuelled the tension that social restructuring inevitably creates.

Alas, the first-rate quality of the research on Australian incomes is not matched by the data on the distribution of wealth, and here we find something of a contradiction. To be sure, as Advance Australia fair? points out, the top 20 per cent of households account for over 60 per cent of the nation’s wealth, and this presents as a stark imbalance. Yet the proportion does not sit with Capital21’s rule of thumb for the order of inequality in modern society generally, on which Piketty says:

the top 10 per cent of the capital income distribution always owns more than 50 per cent of the wealth (and in some societies as much as 90 per cent). Even more strikingly … the bottom 50 per cent of the wealth distribution owns nothing at all, or almost nothing (always less than 10 per cent and generally less than 5 per cent of total wealth, or one-tenth as much as the wealthiest 10 per cent) (p. 244).

The first embarrassing Australian deficiency to note is that we do not have an estimate of the wealth distributed to the top 10 per cent. Leigh and Atkinson have brought our income data up to standard, but when it comes to wealth, the ABS’s estimate for the top 20 per cent, the top quintile in the hierarchy, is the best we can do — as published in Household wealth and wealth distribution (cat. no. 6554.0), a publication that draws on the ‘SIH’ estimates from the Survey of Incomes and Housing, a sample of 1500 or so households conducted every two years.

Notwithstanding the shortcomings of our statistics, the more salient point is that, however striking the apparent fact that the richest 20 per cent own 60 per cent of the wealth appears, this estimate is almost certainly not large enough to contain a top 10 per cent that owns 50 per cent (given the 21 per cent share owned by the second quintile). In other words, either something is wrong with the SIH survey or Australia’s distribution is actually the least inegalitarian in modern history, by Piketty’s reckoning. It follows that a priority for attention in the inequality field is to either confirm or fix this apparent fact. There are three lines of approach, which are the subject of the balance of this paper. We can refine the current picture, or further develop the historical picture (an issue that raises the question of Australian egalitarian exceptionalism), or we can embrace Piketty’s big picture. Ideally, I hope to show, we should do all three.

The current picture

As a first line of approach, perhaps the SIH distribution should be reduced. In this, we come up against a big limitation of the Australian research, for the official data are presented within the framework of quintiles (20 per cent fractions of the wealth hierarchy, as used by Advance Australia fair?) and deciles (10 per cent fractions). The quintile bandwidths are too wide to tell much, and a curiosity about the deciles is that they conventionally leave the top 10 per cent undefined. The ABS, for example, reports interdecile ratios that divide the wealth owned by the households ranked at the top of the 90 per cent decile (i.e., at the 90th percentile, counting from the bottom) by the wealth owned by those ranked atop the 10 per cent decile (i.e., at the 10th percentile, from the bottom). This ratio is referred to as P90/P10. Using the latest figures, for example, dividing the 90th percentile by the 10th gives a ratio of 54, i.e., the 90th percentilers own 54 times the wealth of the 10th. Ratios are also reported for P80/P20, P80/P50, and so on, producing information on slices of the wealth distribution that sit on a precise quintile, decile or multiple thereof apart, with nothing at the bottom, of course, but also, oddly, nothing at the top.

‘Such indices can be useful’, as Piketty says, but he is critical of their shortcomings:

by construction these ratios totally ignore the evolution of the distribution beyond the ninetieth percentile. Concretely, no matter what the P90/P10 ratio may be, the top decile … may have 20 per cent of the total … or 50 per cent … or 90 per cent ... Conversely, interdecile ratios are sometimes quite high for largely artificial reasons. Take the distribution of capital ownership, for example: the bottom 50 percent of the distribution generally own next to nothing. Depending on how small fortunes are measured — for example, whether or not durable goods and debts are counted — one can come up with apparently quite different evaluations of exactly where the tenth decile of the wealth hierarchy lies (p. 267-68).

If we take Britain in 1910, for example, when the top 1 per cent owned 70 per cent of the wealth, such an intense concentration would be so diluted as to be practically lost in an open-ended measure of the top 10 per cent, let alone a quintile, particularly if the balance owned next to nothing. As well as needing to get a real handle on the wealth of Australia’s top 10 per cent, we must go deeply into the upper distribution, if we are to get a firm grasp of inequality’s elusive shape.

Of more immediate relevance, as Piketty notes, ‘wealth surveys tend to understate top wealth shares as compared to estimates based upon fiscal data’. In particular, Piketty’s above reference to durable goods is apposite. Whereas the SIH survey includes the contents of dwellings and vehicles for private use, Piketty leaves them out in ‘following international standards for national accounting, under which durable goods are treated as items of immediate consumption’ (p. 179). With this adjustment, the value of Australia’s household wealth would fall by some $720 bn, or 11 per cent, which would disproportionately reduce the statistical wealth of Australia’s bottom 90 per cent. Even through quintiles, it’s easy to see that the wealth owned by the bottom 50 per cent would be virtually wiped out.

The ABS also identifies discrepancies between the SIH and the national accounts estimates for household wealth that could boost the holdings of the top 10 per cent. Compared with the national accounts, for example, the SIH only captures 54 per cent of the wealth held in financial institutions, 77 per cent of the value of shares and other equities, 80 per cent of property loans (and only 62 per cent of other loans), and 84 per cent of superannuation. The discrepancies amount to about $1000 bn, or 16 per cent of the SIH’s total, overwhelmingly in forms belonging to the top 10 per cent. There can be no simple adjustment, for in some cases the discrepancies can be accounted for elsewhere in the survey, and in other cases the survey is likely to be the more accurate. Nonetheless, fully worked through in conjunction with the treatment of durables against international standards, it’s not inconceivable that we could find that Australia’s top 10 per cent does indeed own 50 per cent of the nation’s wealth, in which case the top 20 per cent is likely to own somewhere around 70 per cent.

The historical picture ... 

The second line of approach is to build up a picture of the top 10 per cent from the work that Leigh has done on the top 1 per cent. In Battlers and billionaires, Leigh presents the below historical chart of the share of wealth owned by the top 1 per cent.

Wealth inequality in Australia (Leigh)

The present trend is again clear, with the share of the nation’s wealth owned by the top 1 per cent rising from 6 per cent in 1968 to about 11 times their proportionate share today (a little down from a 16 per cent peak in 2006). As can also be seen, the top 0.1 per cent has captured the lion’s share, peaking at almost 6 per cent, some 60 times their proportionate share. One of the most illuminating achievements of Leigh and his colleagues — Piketty and the other Young Turks in the inequality field — has been the uncovering of the upper reaches of the wealth distribution to reveal its extraordinary terrain. If the rungs of a ladder represent a million dollars, the wealth of half Australia’s households leaves them on the ground or at least closer to the ground than the first rung. The entry point for the top 1 per cent is $5 million, five rungs up. Way overhead, more like Jack’s beanstalk than a tall poppy, Gina Rinehardt ‘is nearly ten kilometres off the ground’ (Leigh, Battlers etc., p. 5).

Yet here we also find the same ambiguity when we look down on the top 10 per cent from the upside; the obverse of the view from the top SIH quintile. The current 11 per cent distribution to Australia’s top 1 per cent is consistent with the top 10 per cent owning over 50 per cent of the wealth, the Piketty standard, which would require the other 9 percentiles to own an average 5 per cent. Yet 11 per cent could also be consistent with a distribution to the top quintile of 60 per cent, the SIH standard, which would require the other 19 percentiles to own an average 2.5 per cent. In isolation, both are plausible. By retracing Leigh’s work, or as near as practical, to define the top 10 per cent, we would arrive at an index of the top 1 and 0.1 per cents that could be integrated with the SIH data, as adjusted for international standards and the conflicts in the national accounts. This would show a much sharper picture of the social shape of wealth inequality in Australia today.

... and Australian exceptionalism

The above improvements go to our immediate predicament, but there is another perhaps less urgent although potentially more illuminating issue to be raised about Leigh’s historical index in light of Capital21. This issue is important for Piketty’s big picture of inequality, which we will come to shortly, but it also raises the question of whether Australia really can be distinguished as the ‘land of the fair go’, and whether this fabled land is now disappearing. As I will presently show, when combined with Piketty’s data, Leigh’s index strongly supports the idea that Australia does indeed have an exceptionally egalitarian history, but more confirmation could help and it’s possible that further research will reveal the distinction overdrawn. To be clear, the issue is not whether Leigh’s index is accurate in a technical sense, but how close it takes us to the underlying social reality. The index is from work by Leigh and Pamela Katic, titled ‘Top wealth shares in Australia 1915-2012’ (8 March 2013), and the question is: are the findings strong enough to feel confident that we know most of the story, or about half the story, or have we barely started on the problem? In what follows, I will suggest that Katic/Leigh have dug up a skeleton, but some bones are missing and the body should be fleshed out, if possible.

It is first of all crucial to know that the source material on wealth is thin and tricky, as Katic/Leigh are well aware. Their estimates for the first 63 years draw on a census for the starting position in 1915 and inheritance tax data for each of the 25 years 1953 to 1978. Note straightaway the big gap between 1915 and 1953, which we will return to. The estimates for the remaining 32 years are drawn from a one-off academic survey for 1987, microdata from the survey of Household, Income and Labour Dynamics in Australia (HILDA) for 2002, 2006 and 2010 (conducted by the Melbourne Institute), and supplementary analyses of small fractions of the 1 per cent from rich lists. Now, to return to the big gap, when you look at Leigh’s chart (above), the extraordinary thing about the whole story is that Australia’s exceptionalism is buried in a mystery behind the top 1 per cent’s fall from owning almost 35 per cent of the national wealth in 1915 to owning only (relatively speaking) 15 per cent in 1953. This represents a wealth share-cut of almost 60 per cent, and is shown above as simply a straight line between (just) two data points that cover almost 40 years, from the 1915 census to the 1950s tax data. Consistent with this scanty if however reliable evidence, the top 1 per cent’s share could have, say, grown larger through the 1920s, before being reduced cataclysmic-style to 15 per cent in the Great Depression, or perhaps the real story is staggered in reflection of the wars, or was undulating for no apparent reason, or whatever, i.e., the chart tells us next to nothing about what happened to the top 1 per cent’s share of our wealth between 1915 and 1953, only that it was severely reduced, incidentally setting the scene for the post-war period, the most egalitarian period in our history, capitalism’s so-called Golden Age.

Regardless of how the redistribution occurred, the size is no small fact. Indeed, if we combine data from Katic/Leigh and Capital21, the deconcentration of 1915-53 appears as one the most outstanding redistributions of wealth ever seen in a non-revolutionary context. To show this, below I have plotted the Katic/Leigh data with Piketty’s data for the US, the UK, France, and Sweden. Following Piketty, I’ve converted the Australian data to decennial averages to remove the volatility typically displayed in wealth series over the short-run (most often due to sharp movements in asset prices). Figure 2 shows the comparison with the US, where the concentration of wealth was fairly moderate early in the 20th century, at least relative to the concentration in the upper class of old Europe.

Wealth inequality in Australia v US Katic, Leigh, Piketty

As can be seen, the US top 1 per cent’s share was reduced by about 30 per cent by 1940, at which point the movement was basically all over, until the top group’s share began to increase again in 1970. By contrast, Australia’s top 1 per cent share effectively did not stop falling until 1960, twenty years later, by which time the said share had been reduced by over 70 per cent, more than double the US proportion. There were, moreover, some small additional share-cuts before the present reconcentration began in about 1980. Now compare the European countries.

Wealth inequality in Australia v UK, France & Sweden Katic, Leigh & Piketty

Despite the extraordinary European starting positions, we can see again that Australia’s curve not only goes lower and faster, but also fell further in proportionate terms. By 1960, whereas the Australian top 1 per cent had been reduced by over 70 per cent compared with before the Great War; the United Kingdom’s top 1 per cent had only (relatively speaking) been reduced by 51 per cent and France’s by 47 per cent. By 1960, even the share lost by Sweden’s top 1 per cent was 10 per cent smaller than Australia’s; and because Sweden began with a much higher level of inequality, this left its 1 percenters with more than double the Australian proportion. Against all three countries, Australia’s redistribution had gone further by 1960, despite less to begin with. Yet the big difference in the comparison with Europe is that, whereas the US stabilised in 1940, the wealth-share of Europe’s top percentile continued to fall after 1960, and the wealth hierarchy did not stabilise until 1970-80. Australia followed Europe until 1980, although further redistribution was marginal after 1960. Still, if we measure the full trend up to the respective points of reconcentration, the share-cuts were: Australia, 81 per cent; Sweden, 73 per cent; UK, 67 per cent; and France, 64 per cent. Regardless of an apparently handicapped start, Australia won wealth’s 20th century egalitarian stakes by a decent margin.

Is this true? However unequal, did Australia nevertheless have the least inegalitarian society in the modern world, such that ‘the land of the fair go’ is not entirely mythical? How might we complete the skeleton? How can it be fleshed out? First, the starting point needs more support to ensure it’s not too high. The most prized years for research would be 1910, 1900 and 1890, the last straddling a depression, and data for 1880 and especially 1870 would be comforting. Secondly, if there really are no data points between 1915 and 1953, such a deconcentration at the top must have been visible in other ways. In Battlers & billionaires, Leigh finds many alternate indexes of the current reconcentration (luxury homes, up-scale cars, private aircraft, high-end vintage wine, cocaine). Should we not be able to observe social evidence running in a reverse arc at some stage between 1915 and 1953?

Thirdly, and most importantly, while we have a good general explanation for the so-called ‘great compression’, the reasons for Australia’s exceptionally large and stable version are not apparent. The wealth data for the top 1 per cent do not sit neatly with the income data. Up until 1980, the Australian income inequality curve for the top 1 per cent declined in rough parallel with the top European incomes, which were mirrored in turn by Europe’s declining top wealth shares, all three of which leave the drama of Australia’s wealth the odd pattern out. Why? Australia introduced progressive income and inheritance taxes, but not as progressive as in some other countries (compare Capital21, pp. 499, 503 with Battlers etc., p. 75). Australia was not as exposed as Europe to the physical destruction of capital during the wars, and nor would we assume comparable losses in foreign assets. Australia had strong trade unions, but so did Europe. In Battlers etc., Leigh suggests that import tariffs helped reduce poverty but also reduced Australia’s growth rate, yet Piketty shows that wealth disparities tend to grow wider in low growth scenarios. Except perhaps in the last respect, I don’t suggest that anything Leigh has found is wrong, only that the index has an incredibly wide gap that would benefit from more confirmation and an explanation for Australia’s exceptionalism.

Finally, perhaps flesh can also be added to the skeleton’s fresher head and torso. Katic/Leigh allow that their estimates ‘based on inheritance tax could be biased by tax under-reporting, which could have potentially grown over time’, and this likelihood may warrant more consideration in light of Capital21. Although well known in statistical circles, it’s worth noting that, in general, distributions derived from tax records are of interest ‘mainly as indicators of orders of magnitude and should be taken as low estimates of the actual level of inequality’ (Capital21, p. 282). ‘Even if opportunities for legal tax avoidance and illegal tax evasion have increased in recent years’, Piketty warns, ‘we must remember that income from mobile capital was already significantly under-reported in the early twentieth century and interwar years’, a warning that can be extrapolated to apply to the assets themselves. At the least, the estimates could be fleshed out with a critical assessment of the contemporaneous practices of tax minimisation and evasion. In the most recent period, understatements commonly arise from asset placements in offshore havens, and the invisibility of diversified wealth amid the entrepreneurial bias and haphazard ways of rich lists. Again, I don’t suggest that the picture is likely to change much in light of further research, but the possibility that Australian exceptionalism could be moderated or even disappear doesn’t look able to be ruled out in advance. 

The big picture

This brings us to the core of Piketty’s thesis. The historical index matters because it reflects on Australia’s journey and identity, and could imply, inter alia, that the current resistance to rising inequality runs deeper than we know. But there is another reason to work up a better record. More extensive research will also open a third line of approach to the whole issue, which requires a full series of capital-income ratios for Australia. Boiled down, Capital21 is an analytic history of wealth and income distribution over the last two centuries or so across about 30 countries, as seen through the prism of the capital-income ratio. The ratio is dead simple in concept, very useful in practice, and worth a few words.

At the methodological foundation of Capital21, Piketty divides the value of the stock of national wealth — or ‘capital’, as he calls it — by the flow of national income during every given year. This yields the capital-income ratio, which shows the overall importance of capital (i.e., wealth) in the economy and society, measured in units of years of national income. The ratio is Capital21’s controlling metier, solving countless measurement issues across time and space to provide a universal yardstick for comparisons. The chart below shows the ratio in action, as applied to the world’s eight richest (per capita) countries, including Australia, from 1970-2010.

Private capital in rich countries 1970-2010

As can be seen, the stock of capital in the richest countries was worth between 2 and 3.5 years of national income in 1970, and has since risen to between 4 and 7 years. In Australia, the ratio has risen from about 3.5 years to almost 5.5.

The capital-income ratio introduces a parameter that brings a whole new perspective to the history of inequality. To appreciate the ratio’s significance, imagine, for example, if the wealthiest 1 per cent owned 35 per cent of the nation’s capital, as in Australia in 1915, but the total stock was only worth one year of annual income. Now imagine if the top 1 per cent only owned 6 per cent of the capital, as in Australia in 1968, but the total stock was worth six years of annual income (as it nearly does today). This would mean that the top 1 per cent would have actually been a touch wealthier in 1968 than 1915, since 6 per cent of 600 per cent is worth 36 per cent of the national income (compared with 35 per cent in 1915). Assuming any conventionally limited capital distribution, Australian society in 1968 would have been that touch less equal than in 1915, and we would have a very different picture of the ‘great compression’ and the history of inequality in this country.

In summary, the capital-income ratio allows us to delineate the difference between owning 35 per cent of bugger all and owning 6 per cent of the capital in a modern industrial economy. Now, imagine somewhat more realistically that the capital-income ratio in 1968 was equal to three times the value of national income, as it roughly was in Europe and the US at the time. In this case, the value of the top 1 per cent’s 6 per cent share would have been worth about 18 per cent of the national income (6% x 300% = 18%). Compare this with the top 1 per cent’s (real) 11 per cent share today, which is worth about 60.5 per cent of the national income (11% x 550% = 60.5%). If a rise of this order is true (and it’s probable), then the top 1 per cent has not merely doubled its share of Australia’s wealth since 1968, as Leigh finds, but the weight of the share has increased by more than three times — an appreciably more unequal picture. Leigh’s chart picks up the higher concentration, but misses the rise in the weight. It should be emphasised that these are crude hypotheticals, since we do not have capital-income ratios going back further than 1970.

As a final illustration of the ratio’s utility, if we now allow for Australia’s exceptionalism and imagine that 1968’s capital-income ratio was only worth double the national income (and it’s possible), then today’s top 1 per cent will have increased the relative value of this share by more than five times — and the ‘land of the fair go’ will have long disappeared. Of course, an 11 per cent share of such a large increase in the weight of capital also implies a substantial distribution beyond the 1 per cent, but the balance need not necessarily have gone far beyond the top 10 per cent, and in any event, those without any wealth, perhaps 50 per cent of Australian households, will have been left for dust. There is more that could be said about the balance of the distribution, but for our purpose, all this is only to repeat the main point, i.e., the beauty in the ratio is that it allows us to measure the evolution of capital’s weight in the economy and society over time and space. On reflection, in studying the history of wealth inequality before the advent of the capital-income ratio, it seems fair to say that we barely knew what we were really talking about.

Behind the chaste veil: conclusion

There is more in Capital21 that could advance understanding to inform the national conversation promoted by Advance Australia fair? Piketty’s criticism of interdecile ratios was cited above, but he cuts a wider swathe through the old ways. He argues, for example, that the Gini index, an old favourite of inequality bureaucrats, is simplistic, misleading, overly optimistic, confused, abstract, sterile, alienating, and obscures anomalies, inconsistencies and anachronisms, altogether creating difficult-to-interpret pictures of what’s really going on. Piketty dislikes all synthetic indexes (including the Theil index), for ‘it is impossible to summarize a multidimensional reality with a unidimensional index without unduly simplifying matters and mixing up things that should not be treated together’. For Piketty, quintiles, interdecile ratios and synthetic indexes such as the Gini constitute ‘The Chaste Veil of Official Publications’, which ‘give an artificially rosy picture of inequality’; whereas ‘the official reports of national and international agencies are supposed to inform public debate about the distribution’. As well as obscuring and misleading, the chaste approach ‘inevitably feeds the wildest fantasies and tends to discredit official statistics and statisticians rather than calm social tensions’ (pp. 266-68).

As an alternative, Capital21 more or less directly proposes the wholesale introduction of a new framework for indexing inequality, to which the world outside Australia is now slowly turning, thanks largely to the efforts of the Young Turks. The continuous distribution curve of wealth means that any categorical representation of inequality is doomed to be crudely schematic, as Piketty recognises, but he makes a compelling case for the statistics to be divided into the top 10 per cent (including the top 1 per cent and the upper reaches), the next 40 per cent, and the bottom 50 per cent, which he names for convenience the upper, middle, and lower classes. Perhaps the most compelling part of his argument is that this would more faithfully reflect not only the present but the historical reality. The concentration of wealth in the top decile (historically, usually 90 per cent), and in particular the top percentile (up to 70 per cent), has been a feature of societies since at least the early modern period (i.e., from at least the 17th century). This, of course, also means that the bottom 90 per cent has generally always owned next to nothing, until the 20th century. The great social achievement of the last century was the creation of a middle class, the propertied 40 per cent that sits below the top 10 in the hierarchy, which has claimed between 25 per cent of the wealth (as in the US today) and its full proportionate 40 per cent share (as in Scandinavia in the 1970s and 80s, and perhaps Australia). Everywhere, this structural transformation in the distribution within the modern world has been accompanied by a reduction in the share owned by the top percentile, from over 50 per cent a century ago to around 25-30 per cent today — although here again, we come up against the mystery of Australian exceptionalism, the question of whether the shadow from the legendary land of the fair go can still be seen.

There is another reason for revamping the framework of immediate pertinence for the aim of promoting a national conversation, although I should clarify that the above has not exhausted Piketty’s argument. He points out, for example, that his framework would be more consistent with international accounting standards, and would help ‘humanise’ the national accounts, allowing ‘any observer to see just how much the growth of domestic output and national income is or is not reflected in the income actually received by these different social groups’. His framework also, incidentally, has an affinity with the social tables in vogue before the current forms of atemporal sterilisation, as they too sought to portray the ‘flesh and blood aspects of inequality’. The new tables would allow a ‘more concrete and visceral understanding of inequality, which would make it easy for people to grasp their position in the contemporary hierarchy, always a useful exercise, particularly’, Piketty drily adds, ‘when one belongs to the upper centiles of the distribution and tends to forget it, as is often the case with economists’ (p. 267).         

But conversation-wise, the important reason for bringing Australia’s inequality data into the 21st century is that the current framework not only obscures the underlying reality but tends to kill-off engagement by shunting the discussion into the margins. This is because the least obscure features of the almost wholly obscure current system are the extremes, i.e., the top and bottom quintiles, the 90th and 10th percentiles, i.e., the rich (although not the very rich) and the most poor. Hence, when debates about inequality arise, the majority of the citizenry promptly loses interest (or worse), since the subject is not about them but others, the Battlers and billionaires, as Leigh’s title neatly exemplifies. The 60 per cent that occupy the three middle quintiles, or the 80 per cent between the 10th and 90th percentiles, are out of the picture. And because there seems nothing to be done about billionaires besides taxing them more, and since they have the smartest lawyers and accountants and there never seem to be sufficient to make enough difference tax-wise anyway, the conversation turns squarely to the poor and the utility of welfare programs. By this stage, more than likely the only people still in the conversation will be those who were debating the same issues before the question of inequality was raised.

Capital21 stands this perspective on its head. The lowest decile is scarcely central to inequality when the only real difference between the very poor and the rest of the citizenry in the bottom 50 per cent of the distribution can be counted in household durables, while the top 10 per cent owns 60 or 70 per cent of the income-yielding wealth, as is the case today in Europe and the United States, respectively. The inequality is not as extreme as a century ago, but the fact that the trend is in the same direction is of most acute interest, not to the bottom 50 per cent, who have never really owned much, but to the propertied middle class. In other words, the logical focus of a national conversation about inequality is the top 10 per cent and the next 40 per cent, whose wealth is new in historical terms, and apparently reversible.  The destabilising implications of a progressively smaller share owned by an ever more disaffected middle class is the most glaring reason why the current trend could be dangerous; the corollary being a creeping plutocracy with an ever more limitless capacity to buy and sell the democracy and all she commands.

The problem is of course that we can’t be sure we really know what we’re talking about, for we are having a debate without hard and sharp data. The possibility that the chaste veil over the present distribution is hiding a top 10 per cent with a uniquely small share of the wealth by international historical and contemporary standards cannot be ruled out, even if this has increased dramatically in Australian terms since 1968. But nor, as I hope to have shown, can we rule out the possibility that our distribution right now is much more unequal than we would like to think, perhaps somewhere around European levels, with the top 10 per cent owning maybe 60 per cent of the wealth, the middle class owning somewhere around 35 per cent, and the bottom half with 5 per cent. Hopefully, it also now seems obvious that we need to at least refine and ideally reconstruct the current data and further develop the historical picture, not least to settle the question of Australia’s egalitarian exceptionalism. But most of all, we need a full series of capital-income ratios, which represent the point of unity in Capital21 that attracts the ‘masterwork’ tag. On one side, the ratio will allow us to integrate wealth and income inequality (subject to a rate of return schedule), and by default derive the capital/labour income split and all that implies for further analysis (including the now notorious issue of senior executive pay); while on the other side, the ratio will allow us to integrate inequality with economic performance using the growth and saving rates. The full field of data would also allow for reasoned projections of the long-term evolution of inequality in Australia, and that would certainly supply the basis for a national conversation.


Christopher Sheil is the president of the Evatt Foundation, a fellow in history at UNSW and adjunct professor in social policy at Boston University. The author wishes to thank Peter Bryant for comments. This paper is part of the Evatt Foundation's current research project on inequality.


See also:

Suggested citation
Sheil, Christopher, 'Behind the chaste veil', Evatt Journal, Vol. 13, No. 5, July 2014. <http://evatt.org.au/papers/behind-chaste-veil.html>