I’d like to respond to Christopher Joye’s recent blog post (re-posted at The Drum) about inequality.

Why should we care about inequality?

Christopher Joye doesn’t “think there is anything wrong at all with a rise in income inequality if one assumes that we have equality of opportunity; we are committed to combating extreme poverty; and we are vigilant in protecting those members of the community who are fundamentally and irreversibly damaged through, say, mental or physical disabilities”.

This is a common position among small-l liberals of the centre-right, and I see this as a key ideological dividing line between liberals and social democrats.

I place myself on the other side of that dividing line: I do care about inequality, and I think Governments should act to reduce inequality above and beyond ensuring a formal equality of opportunity, or ameliorating extreme disadvantage, or protecting people with disabilities.

I’d like to address a few of Joye’s points in turn to demonstrate where our reasoning differs (and where our ideology differs, too). As an aside, I’d like to note that the question of how much inequality we should tolerate is an inescapably ideological question, one of moral and political philosophy.

Equality of outcomes vs equality of opportunity

I view this as a false dichotomy. I will assume for the purposes of this post that Christopher understands equality of opportunity to mean equal treatment before the law and by public institutions, in other words ‘procedural fairness’, plus perhaps support for public school education. If society ensures equality of opportunity, defined in those terms, should we be comfortable with the dimensions of the inevitable inequality of outcomes that results? I suggest not.

Over time, vast inequality of outcomes erodes equality of opportunity. Wealth, privilege and connections are handed down through generations. Last generation’s meritocrats [fn1] become this generation’s entrenched, quasi-aristocratic elite, able to secure their children’s place in the hierarchy by paying for them to attend expensive schools, or by buying them houses or providing start up capital for entrepreneurial ventures.

If we were to start in one generation with a truly level playing field, with meaningful and substantial equality of opportunity, subsequent generations would nevertheless face unequal opportunities by virtue of the intergenerational transmission of relative status. Significant disparities in parental incomes, and therefore disparities in access to high quality education and health services, stable housing, and the justice system, erode poorer children’s ability to achieve their potential, even if there is procedural fairness. Equality of opportunity is an unstable equilibrium.

Another way of saying this is that social mobility, the likelihood that children will end up in a different earnings quantile than their parents, is eroded by inequality. This makes intuitive sense: the longer the ladder between the bottom and the top, the less likely it is that poor children will manage the journey from bottom to top, whatever their ability, and the less likely it is that rich children will slide down the ladder even if they lack ability.

As it happens, Andrew Leigh and Dan Andrews of Harvard University have quantified this relationship between inequality and social mobility. According to Leigh and Andrews,

“sons who grew up in more unequal countries in the 1970s were less likely to have experienced social mobility by 1999. Across countries, our estimates suggest that a 10-point rise in the Gini coefficient is associated with a 0.07-0.13 increase in the intergenerational earnings correlation. Moving from rags to riches is harder in more unequal countries”.

So, even if you are only concerned about equality of opportunity and social mobility, there are good reasons to be concerned about gross inequalities of outcomes.

What is ‘extreme poverty’?

Christopher indicates that he is concerned about ‘extreme poverty’, and supports its amelioration. He implicitly draws a distinction between extreme poverty (presumably defined in absolute terms) and relative poverty. This distinction has been the basis for the famed ‘poverty wars’ of Peter Saunders (et al)  vs Peter Saunders (et al).

The question for those, like Christopher, who express a concern with absolute or extreme poverty, but not with inequality or relative poverty, is to define what they mean by absolute poverty. Is it a fixed quantity of income, for example $US2/day? If so, that is very low, and is indeed an arbitrary figure. Is it the quantity of income necessary to acquire a fixed basket of goods that are seen as being required for subsistence? If so, what is in that basket of goods?

I think we’ll find that defining which goods and services are necessary to live in dignity in a civilised society is a necessarily relative question; the answer depends on the prevailing attitudes and norms of the society the individual inhabits. Adam Smith famously expressed the same view in the Wealth of Nations:

“By necessaries I understand, not only the commodities which are indispensably necessary for the support of life, but whatever the custom of the country renders it indecent for creditable people, even of the lowest order, to be without. A linen shirt, for example, is, strictly speaking, not a necessary of life. …But in the present times, through the greater part of Europe, a creditable day-labourer would be ashamed to appear in public without a linen shirt … Custom, in the same manner, has rendered leather shoes a necessary of life in England. The poorest creditable person of either sex would be ashamed to appear in public without them. … Under necessaries, therefore, I comprehend, not only those things which nature, but those things which the established rules of decency have rendered necessary to the lowest rank of people.”

In modern times, newer conceptions of relative poverty like ‘capability deprivation’ or ‘social exclusion’ have evolved to describe individuals who lack access to goods or services that are socially defined as necessary, and/or lack the capability to acquire them. Here’s Ken Henry describing Amartya Sen’s conception of capability deprivation:

Sen points to some capabilities that are close to being absolute — ‘to meet nutritional requirements, to escape avoidable disease, to be sheltered, to be clothed, to be able to travel, and to be educated’. Others, like the capability to live without shame or to participate in the activities of the community, are relative to community standards. Being computer literate was not necessary to participate in society 30 years ago, but it is now. Sen views poverty as capability deprivation. And, seen in those terms, poverty is clearly intolerable. It is a form of personal injury that should not be abided in any just society.

In the end, nearly all poverty measures in advanced societies are relative, and they are all ultimately arbitrary. In Smith’s time, extreme or absolute poverty would have been well below the level at which one could afford a linen shirt, yet he rightly saw that lacking a linen shirt would make a person feel, and give them reason to feel, sufficiently impoverished as to be excluded from the mainstream of society. In modern, industrialised society, not having sufficient income to purchase a linen shirt would be likely to put you into ‘extreme’ or ‘absolute’ poverty, but there are many newer goods that we might think of as being equivalent to that linen shirt.

By saying that he is concerned about extreme poverty, rather than relative poverty (inequality), Christopher is really just saying that he is willing to tolerate a higher degree of relative poverty than social democrats. Being willing to accept a higher degree of inequality still implies that you draw the line at some point, and that point is ultimately just as arbitrary as the point at which social democrats would draw it. [fn2]

‘Level effect’ or an ongoing trend?

Joye questions whether rising inequality is likely to be a persistent trend, or whether we have merely taken a one-off ‘step up’ to a higher level of inequality that will remain stable:

I am pretty sure that the latest research shows that rising income inequality is a relatively recent phenomenon (ie, since the 1970s), and it is not clear to me that this will necessarily persist through time. That is, it is just as likely to be a once-off ‘level effect’ that manifested as a result of the relentless centre-right campaign of market liberalisation that took place over the last three to four decades

My reading of the evidence suggests to me that we have not experienced a ‘step up’, but rather a persistent upwards trend in inequality. See, for example, Andrew Leigh’s graph of the top income shares in Australia (below).

This does not appear to be a case of ‘stepping up’ to a higher level of inequality that is then stable over time.

What alternative are we striving for? How much inequality will we tolerate?

Christopher implies that there is something of a slippery slope, that preferring ‘more equality’ necessarily leads one inexorably towards preferring ‘complete equality’. I don’t agree that this is the case. I also believe this is another false dichotomy, this time between those who are comfortable with gross inequalities of outcomes (‘libertarians’, in shorthand) and those who will not settle for any inequalities of outcomes (‘Marxists’, in shorthand).

It’s true that social democrats (myself included) have difficulty pin-pointing the precise degree of inequality we are willing to tolerate. Should we be happy with a Gini coefficient of .28? .25? .2? The judgement is necessarily arbitrary and imprecise, but it is informed by the community’s widely held sense of justice.

That is a nebulous and ill-defined means by which to assess the acceptability of a given degree of inequality, but it is nevertheless important. Joye is dissatisfied with Leigh’s reference to “Australians’ fundamental sense of fairness”, saying that we should not appeal to some “visceral gut feeling about right and wrong”. Leigh here is giving voice to the Rawlsian notion of an ‘overlapping consensus’ in favour of a particular conception of justice, which Joye swats away, perhaps because it is insufficiently quantifiable.

This question, of precisely how much inequality social democrats will allow or prefer, is no less a problem for those whose ideology stresses the primacy of liberty. If you are not a Nozickian minarchist, but you are nevertheless of a pro-market bent, you have chosen to support a degree of ‘freedom’ that is less than complete and yet still substantial. How much freedom do you support? What should be the boundary of state intervention, or of regulation? Would you like Government spending to be reduced? By how much? What is the optimal tax/GDP ratio? What are the valid functions of the state?

These questions are, at the margin, essentially arbitrary in the same way that social democrats’ answers about the preferred degree of inequality are arbitrary. You may support more freedom rather than less, but recognise some limited role for government in particular areas. Where you choose to draw the line has no more or less basis in rational analysis than social democrats’ choice of how much inequity to tolerate and how much state intervention in markets to support. It is a choice; as Ken Henry has said, “societies will choose how much inequity they allow according to the institutions, norms, laws and programs they adopt”.

In the end, the question of how much inequity we allow, and how much liberty we preserve, is one of moral philosophy and of politics. It is difficult to satisfactorily advance a position on these questions using only the tools of the modern economist, but that does not mean they should be abandoned.


[fn1] Michael Young, who coined the phrase meritocracy, foresaw decades ago that a society dominated by a meritocratic elite may be no more desirable than a society dominated by a traditional aristocratic elite.

[fn2] I’d like to note here that there is a clear dividing line you could choose that is not arbitrary: the line between life and death. You could support ameliorating poverty to the extent that individuals’ lives were not immediately imperilled by lack of food and shelter. I am assuming that Christopher’s support for addressing extreme poverty extends beyond poverty which immediately imperils lives, and therefore my critique stands.