The Commonwealth Treasury

Economic Roundup Spring 2007

Close Window

Addressing extreme disadvantage through investment in capability development

Keynote Address to the Australian Institute of Health and Welfare Conference on ‘Australia’s Welfare 2007’, Canberra, 6 December 2007

Ken Henry, Secretary to the Treasury1

Indigenous Australians are the most disadvantaged of our citizens. In addressing this blight on our society, it is necessary to take a broader view of wellbeing than just income inequalities and consider opportunities and capabilities. Improved education and health will be necessary for Indigenous Australians to have the life choices available to other Australians.

Increasing wellbeing will require a reduced role for passive forms of welfare, which do little to encourage Indigenous Australians to invest in education and to participate in employment. Improving education requires seven development platforms: a secure environment, action at an early age, a conducive home environment, ready access to health care, appropriate welfare incentives, realistic job prospects and local engagement in policy development. To achieve these outcomes, program delivery must be targeted to local needs, integrated and delivered in a cost-effective and non-threatening way.


Foreshadowing the content of my speech this afternoon, I would like to begin by acknowledging the traditional owners of the country on which we meet— and I pay my respect to their elders and ancestors.

Thank you to the Australian Institute of Health and Welfare (AIHW), and to Dr Penny Allbon in particular, for organising this conference and inviting me to speak. The AIHW’s report Australia’s welfare 2007 is the eighth in a long-standing biennial series published by the AIHW but is the first under the stewardship of Dr Allbon.

Reflecting the conference theme ‘diversity and disadvantage’, I want to take the opportunity today to talk to you about how policy advisers might conceptualise disadvantage and the means of addressing it. In particular, today I would like to look at the situation of Indigenous people because, without doubt, this is the group of Australians who have experienced the highest levels of disadvantage, however measured, over the longest period of time.

I note that to commemorate the 40-year anniversary of the 1967 referendum, each chapter in Australia’s welfare 2007 provides Indigenous statistics, where available, which illustrate the depth of disadvantage. I am sure these statistics have been highlighted in your discussions throughout the course of today, so I will not repeat them here.

Sharing prosperity and wellbeing

If I were to identify two fundamental roles for government they would be these; first, to provide sustainable macroeconomic growth, with low and stable inflation and unemployment, through sound macroeconomic frameworks and the maintenance of well-functioning markets; and second, to ensure that all Australians share in the nation’s prosperity.

From the Treasury perspective, there is far more to sharing prosperity than simply ensuring that income is redistributed in a way that avoids inequality widening over time beyond some arbitrary level. To our minds, the distributional goals of government must relate to a much broader concept of prosperity, or wellbeing; one that goes well beyond standard inequality measures, or poverty line constructs, based on crude statistical measures of dispersion around mean or median income. These traditional income-based measures of poverty and disadvantage are just too simplistic for the task. The dispersion of money income is of consequence, to be sure, but it is not enough.

Some of you would know that I am generally quite favourably disposed to Amartya Sen’s concept of disadvantage as capability deprivation.2 Sen (1999) emphasises what he refers to as ‘substantive freedoms’ — including political and civil liberty, social inclusion, literacy and economic security — that, of themselves, form ‘constituent components’ of development. Among the capabilities of importance to poverty analysis, Sen (1983) identifies one subset including such things as the capability ‘to meet nutritional requirements, to escape avoidable disease, to be sheltered, to be clothed, to be able to travel, and to be educated’. Poverty lines, defined in income terms for example, that captured these capabilities would not vary much from one community to another and would not, for the same reason, vary much over time. In other words, they might provide the basis for an absolute poverty line measure.

But Sen also notes that a second subset of other relevant capabilities of considerable interest to the classical economists — such as the capability to live without shame, the capability to participate in the activities of the community, and the capability of enjoying self-respect — provides a basis for relative poverty comparisons.

Of course, including all of these elements in an all-encompassing measure of poverty (or disadvantage) — built on a person’s endowment of capabilities, rather than their command over commodities — would be quite a challenge. It’s not surprising that, despite an increasing interest in such a broad measure of disadvantage, no universally accepted measure has been developed. There are, however, many examples of broad conceptualisations of wellbeing and disadvantage being used for various analytical purposes.

For example, we in the Treasury have developed a wellbeing framework as a descriptive tool to provide context for public policy advice; Treasury (2004). It is built on elements of Sen’s capabilities framework within the context of a generalised-utilitarian framework. This quite broad conceptual framework anchors the objective and thorough analysis of policy options that is central to the Treasury’s role.

Another pertinent example is the material contained in the final chapter of AIHW’s Australia’s welfare 2007. While acknowledging that welfare, in its broadest sense, refers to the wellbeing of people and society, AIHW’s primary focus in its Australia’s welfare series is concerned with the system of welfare services and assistance (including specific targeted cash transfers) now operating in Australia, and the people who receive those services and assistance. To give context to the discussion on specific welfare services sectors, summary indicators of wellbeing have been added in recent volumes of the series.3 The conceptual framework underpinning these indicators has three components: healthy living; autonomy and participation; and social cohesion.

The Australian Bureau of Statistics (ABS) has been moving along a similar path, especially with their 2001 publication Measuring wellbeing: Frameworks for social statistics and the series entitled Measuring Australia’s progress.

And several social policy analysts have also expanded the traditional focus on poverty measurement to develop indicators of deprivation and social exclusion.4

Some quite recent work has served to remind us that most measures of disadvantage that are based either at a point in time or on a time-series of cross-sectional data suffer from not being capable of revealing who remains in a situation of disadvantage over time. Nor, generally speaking, do these measures identify the factors that determine whether somebody exits from, or remains in, a position of disadvantage. Increasingly, longitudinal (or panel) data are being used to examine the extent of intra-generational mobility out of poverty.

Significantly, the findings from the first four waves of the Household, Income and Labour Dynamics in Australia (HILDA) survey show that, despite there being consistent levels of income poverty across all households in each year, observing the same households over time shows that income poverty persisted in only a small percentage of households.5 These findings are consistent with results from other high-income countries: the majority of households enter income poverty only temporarily and are able to improve their standard of living over time, while a smaller number of households are at risk of long-term income poverty. As more HILDA data become available we will be able to develop a clearer picture of the determinants of those at risk of long-term poverty, as well as the effects on those who experience shorter durations of poverty.

In addition, inter-generational analyses on the transmission of disadvantage, measured in a number of dimensions, are also being undertaken.6 The general theme that appears to be emerging from these types of studies is that caution is warranted before assuming that income transfers might be an effective means of breaking the generational cycle of disadvantage. Instead, the evidence points toward the need to design income support programs and policies in a way that encourages, or at least does not discourage, active labour market participation; and, in addition, to the importance of a human capital investment strategy, broadly defined to include not only education but also physical and mental health and social development. The focus on human capital development is particularly important for disadvantaged children early in life — a point that justifies a public policy focus on equality of access to capability development opportunities.

Miles Corak has argued, persuasively, that ‘the capacity of children to become self-sufficient and successful adults is compromised not only by monetary poverty, but by poverty of experience, influence and expectation’.7 In other words, there are important, though often subtle, non-monetary factors that determine the outcomes of children within families as well as, arguably, the outcomes of adults within communities. These non-monetary factors include the influence of dysfunctional cultural norms; the demoralising impact of passive welfare and labour market exclusion; and the influence on the cognitive development of children of maternal smoking, alcohol abuse and poor nutrition during pregnancy.8

Several of the multiple causes of disadvantage draw policy makers into difficult areas of social and labour market policy.

Nobody imagines that social policy interventions should seek equality of outcomes. That is just as well, because social policy couldn’t hope to have such potency. No matter how expansive, and expensive, the policy interventions, we will always observe a considerable dispersion in the ‘wellbeing’ outcomes for individuals, both at a point in time and over time, including across generations. Individuals may be provided with true equality of access to materially rewarding opportunities but might choose not to access those opportunities and to live their lives in what would be considered by others in society as a condition of relative poverty, at least in terms of income.9 In my (normative) judgment, policy makers shouldn’t be too concerned by that. Instead, policy makers should be concerned with opportunities. Specifically, they should be concerned to ensure that individuals are endowed with capabilities that allow them the freedom to choose to live their lives in ways that have real meaning and real value.

I endorse strongly Amartya Sen’s view that people who are deprived of such capability endowments may be described as impoverished; as being in poverty.

Addressing Indigenous disadvantage

Measuring the policy effort against that benchmark, Indigenous Australians are entitled to feel especially disadvantaged.

In June this year, in an address to the Cape York Policy Institute’s Strong Foundations conference, I argued there were three key interdependent foundations to Indigenous disadvantage: poor economic and social incentives; the underdevelopment of human capital and capability in general; and an absence of the effective engagement of Indigenous Australians in the design of policy frameworks that might improve those incentives and capabilities.10

In relation to poor economic and social incentives, I argued that perverse incentives — those that encourage undesirable behaviours — are having a negative impact on many Indigenous communities; reducing self-reliance, self-development, aspiration and responsibility, including — in some cases — the commitment to caring for families and communities. In particular, I noted the deleterious effect of the combined incentives in the welfare system, which have resulted in disengagement in some cases and, in many cases, a passive reliance on welfare payments; and which have also done little to encourage Indigenous Australians to invest in education and to participate in employment. And I noted also the resultant breakdown of foundational social norms in many Indigenous communities, as the effects of passive welfare have become entrenched.

Indigenous disadvantage is a regrettable example of income enhancement, in the form of combined welfare payments, not having led to material gains in wellbeing.

Not all welfare is passive. But some is. If we are to make progress, passive welfare, in all its forms, must be addressed.

While incentives are important, they will not be effective in the absence of the human capital — in particular, good health and education — that is needed to take advantage of positive incentives; and to place Indigenous Australians in a position of being able to opt for, indeed demand, the life choices open to non-Indigenous Australians. So a second key component of addressing Indigenous disadvantage involves human capital development.

In my Cape York Policy Institute speech I argued that the third major reason for continuing Indigenous disadvantage has been the limited engagement of, and opportunities for, Indigenous people to shape policies that affect their destiny. A considerable body of international literature suggests that Indigenous engagement in policy development is key to achieving better results — in itself, it reduces the ‘passivity’ of solutions, creating ownership of both the problem and the solution. And it is fundamental to Indigenous self-esteem. Active participation in the decision-making that affects one’s community can be a powerful source of identity, even of pride.11 And it is an obvious means of recognising inspirational role models. Indigenous engagement at the grass roots level has to become the norm.

For all who are engaged in Indigenous policy development, it is not a question of choosing which of these three foundations of disadvantage should be the focus — they must all be addressed, and at the same time.

And yet, as a practical matter, one has to start somewhere. Where should that be?

Today I want to float an idea that addresses specifically the second of these foundations: the underdevelopment of human capital and of capability in general. You will see that, even with that focus, we will quickly get into considerations affecting economic and social incentives and Indigenous engagement in policy development.

Human capital development and the key role of education

Human capital is a term economists talk about quite a bit. Essentially, it refers to the intangible knowledge-based assets people develop that help them become productive members of society. High levels of education and physical and mental health are the hallmarks of strong human capital.

Education can help transform social and economic opportunities, with particularly strong gains for those from disadvantaged backgrounds. People who are better educated are better placed to participate in the labour market and earn higher incomes. Higher levels of education are also associated with lower rates of incarceration and increased engagement in civic life. In short, education is the key to better life opportunities and choices.

As highlighted in chapter 8 (pp 373-4) of Australia’s welfare 2007, the proportion of Indigenous students meeting the benchmarks for reading, writing and numeracy in 2005 were significantly lower than the national rates, in each grade. Moreover, the gap between Indigenous and non-Indigenous students appears to have remained static between 2001 and 2005. The latest OECD Programme for International Student Assessment (PISA) survey results for 2006 confirm there has been virtually no change in Indigenous students’ performance across a range of outcomes since 2000.12

Indigenous education is an important ‘means’ of securing individual and community development. It links so fundamentally to other aspects of community life that educational gains stand a very good chance of leading to improvements in other areas that are also hallmarks of disadvantage.

But, as Sen has suggested, education is not only an instrumental freedom — that is, a means to an end; it is also a substantive freedom — a constituent component of development. People who are educated have greater freedom to choose lives of real meaning and real value. Indigenous education is, therefore, important for its own sake; a valuable ‘end’ in itself. It should be seen as a key component of Indigenous development.

This dual role that is played by education warrants an explicit policy focus on a sustained increase in educational attainment by Indigenous Australians. And in this context, I note the Government has a number of specific targets it has set out to achieve for Indigenous people, including a halving of the gap in reading, writing and numeracy achievement within a decade.

But there is another reason for focussing on Indigenous education — a reason that is both strategic and pragmatic: it provides a clear focus for multiple interventions.

Australian public service leaders have given some thought to the value in approaching Indigenous development in precisely these terms.13 They have come to the view that enhanced Indigenous educational attainment is unlikely to be achieved without seven development platforms being in place.

First, and fundamentally, there must be basic protective security from violence for Indigenous parents and children. Incidentally, Amartya Sen also stresses that the removal of major sources of ‘unfreedom’ — and he specifically instances the lack of effective institutions to deal with crime and violence — are a fundamental pre-condition for development.

Second, there is strong international evidence, to which I referred earlier, that early childhood development interventions, coupled with parental support to develop appropriate at-home learning environments, provide a critical foundational base for young children — especially those from disadvantaged backgrounds — from pre-birth to school.

Third, the home environment needs to be conducive to regular patterns of sleep and study, free from overcrowding and distraction.

Fourth, there needs to be ready access to suitable primary health service infrastructure. In Sen’s terminology, the avoidance of deprivations like starvation and premature mortality are ‘substantive’ freedoms and constituent components of development. But they also play an instrumental role: healthier individuals are physically and mentally more energetic and robust and, as a result, more likely to be active in all areas of life. Healthier children, in particular, are more likely to attend school, and are better able to learn once they are there. Importantly, the instrumental relationship between education and health runs in both directions: better educated mothers are less likely to engage in behaviours that cause low birth weight, putting their babies at greater lifetime risk of a range of diseases, including type II diabetes.

Fifth, particularly in an environment where real jobs are not currently the norm, incentives in the welfare system cannot be allowed to work against the promotion of investment in human capital, particularly of children through the provision of safe and healthy living environments and their attendance at school. Nor can those incentives be allowed to work against the active participation of parents and other role models in communities.

Sixth, there must be a realistic prospect of an educated Indigenous person securing a real job, with the support of appropriate employment services. It is worth observing that almost three-quarters of Indigenous Australians live in cities and regional centres, the vast bulk of which have thriving labour markets. In other places, there is scope for modest and incremental steps towards developing opportunities based on retail and service activities in the local community and, in some places, much bolder steps that would harness genuine commercial opportunities in art, mining, agriculture and tourism, for example. In yet other places it is difficult to avoid confronting the need for mobility. Where remote locations simply cannot produce sufficient job opportunities for local people, there is no point in relying on miracles. A better strategy is to ensure that people have the opportunity to move to take up work if that is what they want to do.

Seventh, governance systems have to support the ‘political freedom’ and ‘social opportunities’ of local Indigenous people (both men and women) to be engaged in policy development.

These seven platforms necessary to support the goal of a sustained increase in educational attainment shouldn’t surprise anyone; they dovetail quite closely with the strategic areas for action and associated indicators contained in the well-known Overcoming Indigenous Disadvantage indicator framework.14

The first four platforms recognise the importance of tackling problems at source and of thinking in causal pathways; points that have been made eloquently, for many years now, by Professor Fiona Stanley.

I am emphasising here the instrumental role played by each of the seven platforms in supporting decent educational outcomes. But it is also the case that each of these platforms, in its own right, tackles an element of disadvantage that we see in many Indigenous communities. So a focus on education would mean addressing the many sources of Indigenous disadvantage. And, as I have noted on the way through, several of the platforms can be viewed as being constitutive components of development; that is, being of more than instrumental significance.

Targeting educational outcomes, therefore, means embracing a holistic Indigenous development strategy.

Policy strategies are one thing; their delivery, on the ground, is another. In recent years we’ve learned quite a lot about models of Indigenous program delivery. No doubt, we have a lot more to learn. However, I reckon we do know this much: that program delivery must be targeted to local needs, integrated and delivered in a cost-effective and non-threatening way. This is important for the proper functioning of government; but it is equally vital for those people for whom the programs exist — Indigenous people themselves. As policy makers and administrators, we understand this at a conceptual level. But we haven’t been very good at allowing that understanding to affect the way in which we implement things. Critically, in our understandable focus on compliance and accountability we have a tendency to insist on paperwork of Himalayan grandeur. And to what end? I have witnessed first hand, in several Indigenous communities, how the mountains of red tape simply bury the limited administrative resources available at the local level.

Concluding remarks

The thought on which I would like to conclude — more by way of a question than an answer — is whether the framework I have outlined as an approach to Indigenous disadvantage has value when considering disadvantage more broadly in Australian society.

I suspect that it does.

Education and the seven platforms required to support it should be seen as capabilities that are critical to development; several of them in a constitutive way.

An individual deprived of these capabilities experiences poverty in a very real and meaningful sense. Certainly, we can say that such an individual is severely disadvantaged. And hopefully, we can agree that such severe disadvantage is unacceptable — whether it manifests itself in a remote Indigenous community, or on the streets of one of our major cities.

Disadvantage is more obvious in remote Australia where it can pervade entire Indigenous communities. In some remote communities, not one of the seven platforms exists. In the cities, if we look hard enough, we see pockets of disadvantage; several of the seven platforms may be mostly in place, with others less developed. There is disadvantage nonetheless.

I have argued here that while poverty assessments based on crude statistical measures of dispersion around mean or median levels of money income are not overly useful, especially because such measures lack a temporal dimension, poverty should, nevertheless, be conceptualised in terms of disadvantage; and, in particular, in terms of capability deprivation. Disadvantage and capability deprivation are concepts that have both absolute and relative meanings. I have argued that education should be accorded special status by policy makers concerned to build capability, and have outlined seven platforms of development that will need to be constructed to support that work. Some of those platforms will be susceptible to measurement; others not so. But whether they can ever be reduced to meaningful quantifiable indicators or not, policy makers cannot be permitted the view that the task of constructing these platforms is too great a challenge. The development of Australia depends upon it.


Australian Bureau of Statistics 2001, Measuring Wellbeing: Frameworks for social statistics, cat. no. 4160.0, ABS, Canberra.

Australian Bureau of Statistics 2006, Measuring Australia’s progress, cat no. 1370.0, ABS, Canberra.

Australian Institute of Health and Welfare 2007, Australia’s welfare 2007, cat no. AUS 93, AIHW, Canberra.

Buddelmey, H and Verick, S 2007, ‘Understanding the drivers of poverty dynamics in Australian households’, IZA Discussion Papers, no. 2827, June.

Corak, M 2006, ‘Do poor children become poor adults? Lessons from a cross-country comparison of generational earnings mobility’ in Creedy, J and Kalb, G (eds) Dynamics of Inequality and Poverty, Research on Economic Inequality, volume 13, Elsevier Ltd, pages 143-88.

D’Addio A 2007, ‘Intergenerational transmission of disadvantage: mobility or immobility across generations? A review of the evidence for OECD countries’, OECD Social, Employment and Migration Working Papers, no. 52.

Daly, A 2006, ‘Social inclusion and exclusion among Australia’s children: A review of the literature’, NATSEM Discussion Papers, no. 62, December.

Henry, K 2007, ‘Creating the right incentives for Indigenous development’, address to the Cape York Institute Conference ‘Strong Foundations — Rebuilding Social Norms in Indigenous Communities’, 26 June.

Henry, K and O’Brien, T 2003, ‘Globalisation, poverty and inequality: friends, foes or strangers?’, Australian Economic Review, vol. 36, no. 1, pages 3-21.

OECD 2007, PISA 2006: Science competencies for tomorrow’s world, volumes 1 and 2, OECD. Paris.

Saunders, P and Adelman, L 2006, ‘Income poverty, deprivation and exclusion: A comparative study of Australia and Britain’, Journal of Social Policy, volume 35, pages 559-84.

Sen, A 1983, ‘Poor, relatively speaking’, Oxford Economic Papers, volume 35, no. 2, pages 153-69.

Sen, A 1999, Development as Freedom, Oxford University Press, Oxford.

Steering Committee for the Review of Government Service Provision 2007, Overcoming Indigenous Disadvantage: Key Indicators 2007, Productivity Commission, Canberra.

Treasury 2004, ‘Policy advice and Treasury’s wellbeing framework’, Economic Roundup, Winter, pages 1-20.

Waldfogel, J 2006, What Children Need, Harvard University Press.

1 I would like to thank a number of my Treasury colleagues, especially Meredith Baker and Peter Robinson, for their help in the drafting of this paper.

2 Henry and O’Brien (2003).

3 See chapter 8 in the volume.

4 See, for example, Daly (2006) and Saunders and Adelman (2006).

5 See Buddelmey and Verick (2007), which defines a household as being in poverty if the equivalised household income is below 50 per cent of the median.

6 See for example, Corak (2006) and d’Addio (2007).

7 Corak (2006, p 171).

8 See also Waldfogel (2006) and  d’Addio (2007, chapter 2).

9 See Corak (2006) and d’Addio (2007) for some other reasons.

10 Henry (2007).

11 The link between identity and behaviour has been explored by many researchers. In the present context, the work of Dr Michael Chandler of the University of British Columbia is especially relevant.

12 OECD (2007).

13 Secretaries Group on Indigenous Affairs, chaired by the Secretary of the Department of the Prime Minister and Cabinet, Dr Peter Shergold. The work was initiated and supported by the Treasury and the Department of Families, Community Services and Indigenous Affairs.

14 See Steering Committee for the Review of Government Service Provision (2007), especially figure 2.2.1 in chapter 2.

Next: Challenges confronting economic policy advisers
Return to:
Table of Contents