Social policy

Greg Marston and Zoe Staines

Key terms/names

civil society, distributive justice, evidence-informed policy, liberalism, paternalism, social policy, welfare state, workfare


Social policy is guided by questions of distributive justice: who gets what in society, under what circumstances, and how. Social policy is generally considered to be synonymous with the ‘welfare state’, which sees government as a direct provider or regulator of private and not-for-profit social welfare services including education, health, community services, social housing, occupational welfare and income support. These forms of welfare are usually underwritten by social insurance schemes, which redistribute funds accumulated through taxes and levies to those in need of support. They can also take other forms, like the provision of tax expenditures and informal care provided through civil society.1

The various forms of welfare, provided through a combination of direct government service delivery, markets, non-profits and civil society, are referred to as a mixed economy of welfare. Social policies tend to interact and intersect in a number of different ways, forming a complex web of enabling and disenabling structures and systems. For example, the right to access and engage in meaningful education is deeply linked with other areas of social policy, because policy structures can either establish the architecture that enables or denies this access and engagement (like promoting access to quality early childhood care).2 Social policy can, therefore, have direct and substantial impacts on how people live their lives, the choices available to them, and their overall standard of living. Social policy decisions are also often highly politicised and deeply contested as the meeting of human needs has both a moral and material dimension.

This chapter explores some fundamental characteristics of social policy including a brief history of social policy in Australia, how it is made, by whom, and key debates. As you will see, social policy plays a powerful role in shaping how society operates, how it redistributes wealth, how it cares for and controls its most disadvantaged members, and provides tax breaks for middle and upper income Australians. Summing up the role of the conflicted welfare state in an essay titled ‘What is social policy?’, the pioneering scholar of social administration Richard Titmuss wrote that ‘what is “welfare” for some groups may be “illfare” for others’.3 In this chapter we seek to draw out these dilemmas and contradictions.

Social policy in Australia: recent history

Prior to Federation in 1901, social welfare in Australia was largely the purview of non-government charitable organisations. Smyth described Australian colonial society as being ‘cool’ when it came to government-provided welfare, but ‘hot’ on promoting equal opportunity.4 The focus was on supporting citizens to be self-sufficient rather than looking to government for poverty relief. This represented a contrast to the (often stigmatised) provision of state welfare under the Poor Laws in Britain.5

Dickey described the years following Federation as being characterised by a transition away from an age of charity to an age of rights.6 The Harvester Judgement, handed down by Justice Higgins of the Court of Conciliation and Arbitration in 1907, established the first minimum ‘living wage’ for Australian workers.7 In 1908, the new Australian government also introduced a fixed-rate old-age pension, though this was not universal. For example, only those who passed the means test qualified and some groups, like Indigenous Australians, were explicitly excluded.

Thereafter, around the Second World War, Australia’s welfare system was dramatically redefined and expenditure increased exponentially.8 As Shaver explained, ‘Australia entered World War II with only fragmentary welfare provision: by the end of the war it had constructed a “welfare state”’.9 This was partly prompted by the formation in 1941 of a Joint Parliamentary Committee on Social Security by the Menzies government (1939–41), which reviewed existing social policies and recommended new measures to improve postwar life.10

A series of new social policies were introduced during and in the aftermath of the Second World War, including widows’ pensions, uniform income tax in 1942, and the National Welfare Fund in 1943–44, which funded the national unemployment benefit.11 The provision of national unemployment benefits was, at least in part, a response to the anticipated demobilisation of military personnel in the post–Second World War period.12 However, the benefits were means-tested rather than universal – a contrast to the approach then taken in Britain.13

In 1945, then-Treasurer (and later prime minister from 1945–49) Ben Chifley referred to Australia’s growing social security system as a safety net much like that used by a trapeze artist: ‘The net is not, of course, part of the main show … The more competent the performer, the less the net will be used’.14 The underpinning belief was that the best form of welfare was a job and, thus, the emphasis was on ensuring equal access to fair employment. The social security system was perceived as a ‘fall-back’ measure only. It is for this reason that Australia’s postwar welfare state came to be characterised in the literature as ‘a wage earner’s welfare state’, or more precisely given the nature of the labour market during this period ‘a white, male wage earner’s welfare state’.15

During the 1950s and 1960s, there was only incremental social policy reform, perhaps in part because of very strong employment throughout the period.16 This preceded further widespread expansion of the welfare state during the 1970s, when unemployment rates began to increase with the 1974 global recession.17 A series of inquiries was also initiated to examine social welfare, including the Commission of Inquiry into Poverty in Australia, or the ‘Henderson Inquiry’, established by the McMahon Coalition government (1971–72) and whose terms of reference were later expanded under the Whitlam Labor government (1972–75). Following this (and other) public inquiries, a raft of changes were implemented, which had the effect of moving welfare from being viewed as residual, as per Chifley’s description, to becoming a fundamental aspect of citizenship.18

The Whitlam government’s reforms were undertaken on the basis that domestic social policy should focus on achieving a more ‘just’ and ‘liveable’ Australia.19 They involved an extensive program, including increased expenditure on public housing, a revision of school funding rules to recognise level of need, the removal of all fees for tertiary students, and the introduction of the country’s first universal health care, Medibank.20 Whitlam also established an Indigenous land rights scheme in the Northern Territory and announced an explicit shift in social policy focus for Indigenous Australians from ‘protection’ to self-determination.21

In the late 1980s and early 1990s, Australia experienced some further fundamental social policy shifts, including major pension reforms and the introduction of the Working Nation policy, which signalled a transition away from the public provision of unemployment support services to a quasi-market model.22 This was coupled with the introduction of ‘active labour’ policies from 1986 onwards, which placed increased conditions on unemployment benefits.23 Social policy under recent Australian governments has been comparatively conservative and routinely underpinned by neoliberal arguments about reciprocity, the benefits of market provision and individual responsibility.24 However, the overall size of Australia’s welfare expenditure envelope has nevertheless increased.

Recent social welfare expenditure in Australia

Australia’s welfare expenditure increased from $117 billion in 2006–7 to $157 billion in 2015–16, representing a growth of 3.4 per cent per annum.25 Simultaneously, expenditure on health between 1989 and 2014 increased from 6.5 per cent to 9.7 per cent of Australia’s Gross Domestic Product (GDP).26 In 2015–16, the Australian government also lost a total of $47 billion in tax expenditures, which is additional to the total welfare expenditure reported above.27 Finally, the sheer number of people involved in delivering welfare services has also increased, with the ‘welfare workforce’ growing by 84 per cent since 2005 and representing 4.1 per cent of the total Australian workforce in 2015.28

Klapdor and Arthur frame these increases as being largely a result of ‘population growth, population ageing, labour market changes and economic circumstances as well as policy changes relating to eligibility requirements’.29 They are also partially due to the addition of the National Disability Insurance Scheme (NDIS), which is expected to increase expenditure on disability services from $4.7 billion in 2015–16 to around $24 billion in 2019–20.

Notwithstanding these overall increases, Australia’s social welfare expenditure as a percentage of GDP continues to be comparatively lower than most other countries in the Organisation for Economic Co-operation and Development (OECD).30 Welfare in Australia, in the form of income support payments, is also highly targeted. In 2017, about 80 per cent of Australia’s spending on cash welfare benefits was means tested, making Australia the highest means-testing country in the OECD.31

Against this backdrop, the gap between rich and poor in Australia is growing. The wealthiest 20 per cent of households accounted for 59 per cent of total household wealth in 2004–5 and 63 per cent of total household wealth in 2015–16.32 In contrast, the poorest 20 per cent of Australian households only accounted for around 1 per cent of total household wealth in 2004–5 and also in 2015–16. (The gap did, however, remain largely stable from 2013–14 to 2015–16.) Wealth also tends to be distributed unequally across geographical regions and for different groups. For instance, median disposable household income for Indigenous households in urban areas between 2011–16 increased by $57 per week, but fell by $12 per week in very remote areas where incomes were already far lower.33 In general, Indigenous Australians are more likely to experience absolute as opposed to relative poverty.34 Addressing these disparities requires political and policy change. Ongoing policy debates about the means and ends of welfare are underpinned by competing moral arguments, as the following section illustrates.

How is social policy made, and by whom?

The question of who is involved in making social policy is important, because different policy actors will have different worldviews, moral beliefs, experiences and agendas. This is particularly apparent when thinking about the different goals and objectives of state actors (i.e. bureaucrats and elected politicians) and non-state actors (e.g. individuals, collectives, not-for-profit and private-sector organisations).

State actors shape social policy through their direct role in the policy-making process. Policy advisers in the public service or in ministerial offices have the role of exploring social policy ‘problems’ or ‘issues’ as they arise, gathering research and evidence around different options, undertaking consultation, and providing information and advice to elected officials about possible courses of action. In Australia, elected officials then have the final decision-making power over government social policy; it is up to members of parliament and Cabinet to decide which policies are to be pursued, how, when and why.

Non-state actors also have an important role to play in shaping, delivering and sometimes also designing social policy. For instance, they may:

  • seek to influence the focus and development of government social policy through advocacy, lobbying, and participation in consultation
  • deliver government-led/designed social policy, particularly through new public management (NPM) contracting arrangements
  • initiate and develop social policy themselves, either with or without the involvement of the state.

Processes for making social policy will inevitably differ between these groups. There are, however, recognised standard processes for policy making in the public sector – often articulated through the concept of policy ‘cycles’.35 Althaus, Bridgman and Davis’ Australian ‘policy cycle’ proposes eight stages of policy development: (1) issue identification, (2) policy analysis, (3) policy instruments, (4) consultation, (5) co-ordination, (6) decision making, (7) implementation, and (8) evaluation. The authors argue that policy makers do not necessarily step through these stages consecutively, but that the policy process may instead be haphazard: a kind of ‘policy dance’.36

Others have critiqued the Australian ‘policy cycle’ on the basis that it represents an overly technocratic view of policy making and does not adequately grapple with the complexities of real life, including political dimensions and other constraints.37 Indeed, the process of identifying and framing social policy ‘issues’, choosing which issues demand a response (and which do not), identifying and interpreting evidence, and making recommendations is inevitably political, demanding a series of subjective and collective value judgements. Below, we turn to two aspects of this process – the role of evidence, and the cyclical and iterative nature of policy making – to illustrate its somewhat messy nature.

Evidence-informed social policy

As indicated by the ‘analysis’ and ‘evaluation’ steps of the Australian policy cycle, social policy is not merely the end product of a contest between different ideological perspectives. It is also influenced and informed by empirical evidence. Although this is not a new concept, it gained prominence with the growth of the ‘evidence-based policy’ movement in the 1990s and early 2000s.38 The movement grew out of the UK and was also taken up in Australian policy circles, supported primarily on the basis that it provides a foundation for improved public policy decision making grounded in objective ‘truth’ rather than ideology, and can also improve efficiencies and outcomes.39 However, what this commitment means in practice remains somewhat murky. For instance, evidence-based policy raises questions about which types of evidence should be relied upon, how well they approximate ‘truth’, how they should be used, and to what extent they can or should influence policy outcomes.40

There is a great deal of contestability regarding the value of various types of evidence for policy development. For instance, there continues to be much debate regarding the utility of ‘evidence hierarchies’, which generally place greater value on evidence produced through experimental and quasi-experimental quantitative studies41 over evidence produced through qualitative inquiry.42 This preference for particular research methods over others, however, shapes and moulds the types of knowledge that are able to be produced, often favouring positivist ontologies (that is, those that perceive the world as objectively ‘knowable’ and ‘measurable’, typically through rational scientific means). As Marston and Watts argued:

If knowledge operates hierarchically, we begin to see that far from being a neutral concept, evidence-based policy is a powerful metaphor in shaping what forms of knowledge are considered closest to the ‘truth’ in decision-making processes and policy argument.43

There are many arguments for the inclusion of richer forms of knowledge in social policy inquiry and design: forms of knowledge that do not necessarily appear at the top of evidence hierarchies. This is particularly the case in the social sciences, because despite common assumptions that particular policies or programs can directly produce certain anticipated outcomes, trajectories of change are rarely simple or linear when dealing with humans’ lived experiences.44 Any change in an individual’s life is more likely to be the result of the intricate interplay of personal and environmental factors rather than any policy or program alone.45 Thick qualitative description can be particularly useful in teasing out these complexities and providing fuller accounts of the impacts of different policies, particularly when individuals are exposed to and affected by multiple policies and programs at once.

The process of developing and framing research studies also involves implicit assumptions that inevitably influence subsequent findings. For instance, the types of research questions being asked in a study permit the researchers to ‘discover’ some forms of evidence, but ignore (or fail to discover) others. Furthermore, social constructivists argue that all ‘truths’ are not necessarily discoverable or knowable. Thus, there is inevitably an implicit bias in the types of knowledge that are able to be produced through common and accepted research methodologies, regardless of their specific methodological leanings.

The extent to which evidence influences policy outcomes is also a cause for debate. As Colebatch and others have discussed, the policy development process is inevitably a contest between different types of knowledge, different ideologies and diverse ideas.46 It is not driven by evidence alone. Thus, it is unclear what role evidence does and should play within this contest, and whether this depends on changing contexts and circumstances (e.g. evidence availability).

In response to the potential shortcomings of evidence-based policy, some have advocated a shift to the more pragmatic aspiration of being evidence informed.47 This involves an acknowledgement that there are various forms of evidence, that evidence is not neutral, and that policy making is also guided by factors other than evidence alone.

Social policy as an ongoing, iterative process, rather than an end ‘product’

There is an often-held misconception that once social policies are designed, they move along the policy conveyer belt to be implemented in an apolitical and exacting manner. However, this view treats social policy as an end ‘product’, which arguably overemphasises the linearity of the relationship between agenda setting, policy design and implementation. Ewig and Palmucci stated:

We know from previous studies of implementation that one cannot assume that policies will simply be implemented as designed, nor is the process of implementation a linear one from policy passage to simply successful or unsuccessful. Instead, implementation is an interactive political process involving political calculations and negotiations among diverse parties who often have competing political stakes.48

Most social policies continue to be iteratively designed and redesigned, even if only in an incremental sense, during their implementation and throughout their delivery – a sort of continual moulding and reshaping to suit different local circumstances, match different stakeholders’ needs, respond to changing contexts, and/or to align with the views and needs of those responsible for social policy delivery. This can happen in a bottom-up way, where those delivering social policy – the ‘street-level bureaucracy’ of social workers and other human service professionals – adapt policies to suit local needs.49 It may also occur through top-down changes as a result of feedback loops in the policy cycle (e.g. in response to policy evaluation findings) or in response to other changes in context and circumstances.

Moran and Elvin argued that ‘bottom-up’ policy adaptation can hold strengths insofar as it nurtures feedback between social policy designers and the lived experiences of those at the grassroots level.50 This can be helpful in democratising the delivery of social policy, empowering social workers and other professionals to redefine policy goals in a way that suits local circumstances, and lessening the overall distance between high-level policy objectives and ground-level policy experiences. However, local-level decision making can also shift the ways and means of social policy in a manner that is less transparent and accountable, which can lead to discrepancies and inequities across different jurisdictions. The increased reliance on not-for-profit organisations to deliver social policy in Australia can also serve to deflect political risk and responsibility away from governments.51

Debates and non-agenda issues

Social policy is often highly contested. Debates regularly invoke questions about who is deserving of different forms of welfare, and the overall size of Australia’s welfare funding envelope. There is also widespread debate about the nature and strategy behind welfare in Australia, including disagreement concerning the ‘social engineering’ objectives of some social policies in line with behavioural economics. This is particularly apparent in unemployment policies, which embed productivist assumptions about the pre-eminence of paid work over all other forms of work, such as informal care and unpaid domestic labour. An ethics of care is marginalised, while the paid work ethic is eulogised in these debates.

Who are the ‘deserving’ beneficiaries of social policy?

Social policy is an avenue through which goods and services, both tangible and intangible, can be provided to some members of society, while simultaneously being denied to others. Decisions about the distribution of government-funded welfare resources frequently require governments to draw boundaries around identities and social groups, labelling some as deserving and others as less so. These boundaries are sometimes arbitrary and are often fraught.

The concept of legal citizenship provides an example, as it opens the door for individuals to access a range of social resources that are otherwise unavailable to non-citizens. Whether this is always fair or just is debateable. For example, before being recognised as citizens in the 1967 constitutional referendum, Indigenous Australians were largely denied basic rights on the basis of their non-citizen status. This included being denied access to many forms of social welfare that were enjoyed by settler Australians. Instead, Indigenous Australians were (and arguably still are) subject to domestic policies and practices that problematised them as requiring heavy modification and intervention to conform to the standards of settler society.

Who is deemed to be deserving of social welfare thus depends on how social policy ‘problems’ are framed, and which individuals or groups are problematised as a result of that framing. Bacchi’s ‘What’s the problem?’ approach recognises that policy ‘problems’ are not objective truths, but are instead socially and discursively constructed and reproduced.52 According to Bacchi’s approach, social policy responds to ‘problems’ that, just as they have been socially constructed, can also be questioned, contested and disrupted.

With regard to gender equality, Bacchi and Eveline stated, ‘policies do not simply “deal with” the “problem” of “gender inequality”. Rather, policies create different impressions of what the “problem” of “gender equality” entails.’53 Bacchi later discussed policy responses intended to address the pay gap between men and women, focusing on one response that provided additional training to women.54 Bacchi argued that the response placed the blame for the gap on women’s shoulders, implying that it was women’s lack of training that had caused the pay gap. This framing, however, ignores other fundamental structural and historical issues that also play a critical role.

Discourses around ‘welfare dependency’ also provide a pertinent example of how framing can directly impact social policy responses. For instance, recent discourse tends to frame welfare as being innately problematic, with dependency on the state perceived as a moral bad, while dependency on markets is celebrated as a marker of success and independence. Welfare ‘poison’ is now perceived as a core contributor to long-term social disadvantage, rather than a potential solution.55 Therefore, the policy ‘problem’ shifts from claims that there is not enough welfare to pull people out of poverty, to claims that there is too much welfare for people to pull themselves out of poverty, thereby causing the ‘poverty trap’, where incentives to remain on welfare outweigh incentives to move into paid work. This refocusing of the issue shifts discussion away from historical, social and structural causes of poverty to the individual themselves, and leads to responses that focus on overcoming perceived individual deficits such as laziness, lack of skills and moral hazard.

The reframing of welfare in Australia has prompted increased calls for the ‘activation’ of welfare recipients by engaging them in welfare-to-work or ‘workfare’ programs. Since the late 1980s and early 1990s, a series of reforms to the provision of unemployment support have resulted in an increased focus on ‘activating’ and upskilling the unemployed. This has involved, for example, the introduction of activation measures from 1986 (e.g. a requirement to register with the then Commonwealth Employment Service) and an ‘activity test’56 in 1989 as a condition of social security payments.57 Since then, active participation requirements have continued to be strengthened through a range of incremental reforms to social security and employment services.

This relatively recent history sits in contrast to the approaches of past Australian governments, particularly between Federation and the Second World War, which moved towards a focus on the demand (rather than supply) side of the unemployment ‘equation’. In particular, historical policies primarily sought to boost the availability of jobs through mechanisms like job guarantees and full employment, and ensure suitable work conditions. Policies since the late 1980s have, instead, recast the unemployed as the core ‘problem’ and site of possible intervention. Australia’s current remote-employment program, the Community Development Program, provides one example of how this sort of framing can influence social policies and have significant implications for participants who are subject to strict and coercive program rules.

Case example: Community Development Program

The Community Development Program (CDP) currently operates in 60 remote regions across Australia, which include more than 1,000 separate communities.58 The program supports ‘job seekers in remote Australia to build skills, address barriers to employment and contribute to their communities through a range of activities’.59 According to the minister for indigenous affairs, Nigel Scullion, the program also aims to ‘put an end to sit-down welfare’ and transition unemployed, remote-living (mainly) Indigenous Australians into employment.60 By July 2018, 32,000 individuals were participating in the program, about 80 per cent of whom identified as Indigenous.61

Under the CDP, participants are required to attend frequent appointments with their service provider, actively look for jobs, and undertake up to 25 hours of work-for-the-dole activities per week. If participants do not comply with program rules, they can have their welfare income suspended until they re-engage.

The requirements under the CDP are more intensive and punitive than for unemployed persons living in urban parts of Australia who operate in the urban-equivalent program, JobActive.62 At least partially as a result of this, substantially more financial penalties have been applied under the CDP than under JobActive (despite JobActive having over 20 times more participants) and also under previous remote programs.63 However, noncompliance with CDP rules can result from a range of factors, including low English literacy (e.g. participants not being able to communicate with program staff, who rarely speak Indigenous languages), and cultural/family commitments.64 For at least some participants, noncompliance may also result from poor health and wellbeing, which may not be properly assessed or for which participants may not be able to provide adequate supporting documentation, due to poor access to medical facilities in remote communities.65 Nevertheless, the high rate of financial penalties under the CDP has led to reductions in income for some of Australia’s most socially disadvantaged and poorest populations.66

The CDP seeks to address the perceived issue of welfare dependency by ‘activating’ and upskilling individuals, thereby implying that inactivity and poor skills are the causes of unemployment. However, there are obvious silences in this framing of the issue. For instance, remote economies where the CDP operates are generally very weak, with relatively few job opportunities available. Thus, the demand for jobs regularly outstrips supply, leading to entrenched high unemployment. However, this is not appropriately acknowledged in the design of the CDP. As Jordan and Altman argued, ‘if one acknowledges the major structural barriers to employment opportunity, it is difficult to accept that withholding welfare payments unless recipients display the “correct” behaviours (judged according to mainstream Australian norms) will be sufficient to lead to a job’.67

The CDP also does very little to address other barriers to employment, including the multifaceted circumstances of disadvantage experienced by many CDP participants, like poorer health, standards of living, and access to basic social services. Much of this intergenerational disadvantage is the result of settler violence and racist colonial policies. Thus, withholding payments for noncompliance with the CDP punishes individuals for circumstances that are, in many cases, caused by broader socio-political and historical issues.

At the same time as remote-employment policies are becoming more coercive, employment rates in remote Australia have remained largely stagnant and the gap between Indigenous and non-Indigenous employment in remote areas has widened.68 Despite claims about the importance of evidence-based or informed policy, there is little to no robust evidence available on the public record which demonstrates that the CDP, or similar previous programs, have worked to improve employment outcomes for remote participants, especially those experiencing complex employment barriers.69 This suggests that factors other than evidence are driving the design of social policies in this area, and indicates a need to reconsider the framing of the remote-employment policy ‘issue’.

Social policy as a behavioural tool

Social policy can be used to enable and empower, but also to govern, coerce and control. Social policies inevitably embed normative assumptions about suitable or desirable ways of living and behaving, which can serve to restrain each individual’s power over their own lives and identities. They can also explicitly contain behavioural objectives, seeking to influence the ways that social policy ‘subjects’ view and interact with the world, including through behavioural economics and ‘nudge’ interventions.70 For instance, social policies often seek to influence how individuals address their health, spend their time, grow their wealth, and more.71

Thaler and Sunstein discuss ‘nudge’ interventions as being grounded in a libertarian-paternalist framework, which recognises the critical importance of personal liberty, but which also acknowledges the potential benefits of ‘soft’ paternalism in influencing behaviour without restraining individual choice. ‘Choice architects’ – those responsible for devising nudge interventions – seek to subtly manipulate the context within which choices are made so as to encourage, but not to require, certain choices over others. Thaler and Sunstein provide the example of placing fruit at eye-level in school cafeterias to encourage students to choose healthy food options. While this policy does not restrain their ability to choose other options, it nevertheless subtly influences the likelihood that their choices will be healthier than if the fruit was placed elsewhere.

At the other end of the scale are policies driven by hard paternalism. These policies tend to limit individual freedom and choice, instead coercing individuals to conform to particular standards of behaviour or ways of being. Extending Thaler and Sunstein’s example, a hard paternalist approach to improving school students’ diets might involve regulating the food options available in school cafeterias to exclude unhealthy foods, thereby restricting choice and removing individual discretion.

Over the past decade, there has been an intensified interest in translating behavioural economics theory into public policy development.72 Bonoli has referred to this as the ‘active social policy paradigm’: one in which governments routinely use social policy instruments to pursue the health and wellbeing of their populations.73 In Australia, the Department of the Prime Minister and Cabinet now has a dedicated behavioural economics team, ‘BETA’, established to enhance the ability of the Australian public service to, among other things, ‘apply behavioural insights to public policy and administration’.74 The team has worked on a range of policy projects to date, including in relation to tax compliance, influencing consumers’ energy choices, and ensuring compliance with labour laws.75 Though there are arguments for and against the use of behavioural economics, behavioural objectives are apparent in a range of social policies in Australia. Income management (discussed in the below case study) provides one example.

Case example: income management in Australia

Compulsory income management involves ‘quarantining’ proportions of an individual’s welfare income and diverting the quarantined amount to a ‘BasicsCard’, a type of debit card where funds cannot be converted to cash, nor used to purchase certain items that are deemed (by the state) to be morally hazardous, including alcohol, tobacco, pornography and/or gambling services.

Having been first introduced in 2007 under the Northern Territory Emergency Response, income management now operates in discrete jurisdictions across Australia, including across the Northern Territory, and in parts of Western Australia, Queensland, New South Wales, South Australia and Victoria. By March 2018, there were around 25,270 participants across Australia (though this excludes participants in the areas added since this time).76

These schemes operate differently, quarantining between 50 and 80 per cent of an individual’s welfare income, and taking different approaches to the administration of income management. For example, individuals who live in Bankstown (New South Wales) can have 50 per cent of their welfare income managed for at least 12 months if they are referred by a social or child protection worker, are less than 25 years of age and considered to be a ‘vulnerable welfare payment recipient’,77 and/or if they volunteer.78

Alternatively, income management under the Cape York Welfare Reform initiative79 is delivered as one of multiple options (including referral to support services) available to the Family Responsibilities Commission: a statutory authority that undertakes restorative conferencing with welfare recipients who breach certain social triggers.80 Conferencing is undertaken by Indigenous community Elders (sitting as ‘local commissioners’) and the decision about whether to apply income management, including what percentage and for how long, is also made by the Elders. This model was ‘opted into’ by the participating communities as part of the broader Cape York Welfare Reform initiative.

In Ceduna and the East Kimberley, trials of the cashless welfare card (a different form of income management) are ongoing, with recent trials implemented in the Goldfields and East Kimberley regions of Western Australia, as well as the most recent trial being implemented in Bundaberg and Hervey Bay in Queensland from January 2019. Under the cashless welfare card, individuals are not referred. Instead, up to 80 per cent of an individual’s welfare income is automatically redirected to a debit card, which can only be used to purchase approved items (i.e. not alcohol, gambling products or for withdrawing cash).

Although the models differ in their design and administration, they all involve behavioural objectives, which seek to coerce certain behaviours by restricting or removing individual choice over expenditure – a form of hard paternalism. In this way, welfare income is used as a lever for behavioural compliance with selected social norms. Welfare conditionality of this nature is also supported by broader discourses around welfare dependency and, thus, it is hoped that by increasing conditionality, individuals will ultimately be incentivised away from long spells on welfare.

This social policy focus raises challenging questions around the roles of different players in social policy design and implementation. For instance, income management in all areas (aside from Cape York) has been conceived of and designed by the Commonwealth government. However, it co-opts state-level bureaucracy in its implementation by requiring child safety workers, for instance, to make client referrals to the scheme. Its operation therefore depends on the co-operation and compliance of state government level public servants. The case study of compulsory income management, like so many social policy examples illustrates the complexity of federalism, particularly the degree to which co-operative or competitive federalism is at play in the design and delivery of social policies.

The Cape York trial, which has been ongoing since 2008 and was designed and implemented through a partnership between four Indigenous communities (with a fifth community added later on), an Indigenous not-for-profit organisation, and the Queensland and Commonwealth governments, also raises questions around clashes between Indigenous self-determination and hard paternalism. Design and governance of the trial has empowered some Indigenous groups and communities in the policy-making process, while other voices have been marginalised. Nevertheless, this case study presents a conundrum for libertarians: to what extent can or should Indigenous ‘communities’ (however defined) be free to choose and design social policy for themselves, even if these social policies are paternalistic?


Social policy is a way of describing the actions and configuration of governments, the family, markets and civil society in meeting the wellbeing of citizens and residents. The formal and informal rules and regulations governing access to goods and services create forms of inclusion and sites of exclusion. Invariably, the design and implementation of social policies reflects and embeds value judgements about the good life and the good society. As discussed throughout this chapter, the means and ends of social policy are often controversial and highly contested. Political debates frequently revolve around the overall size of the welfare funding envelope, the shape of the welfare system, the forms of delivery and the identities of ‘deserving’ and ‘undeserving’ beneficiaries. Although Australia has a long history of welfare conditionality, there has been an intensification of this in recent decades. This has coincided with changing discourses around the role and impacts of welfare, including an increased focus on the perceived toxic impacts of long-term ‘welfare dependency’. Where welfare was once viewed as an acceptable safety net or fallback measure for alleviating poverty, it is now regularly described as a core contributing factor to moral decay and decline.

The framing and reframing of welfare in Australian society has had, and continues to have, a direct influence on how social policies are constructed, implemented and monitored. Social policies themselves both reflect and are active in the construction of these dominant narratives. They can be used as tools to unite, enable and empower, but also to divide, govern and coerce. The history and case studies contained in this chapter provide examples of this variability. Debates around these aspects of social policy will likely continue into the future, as Australia rapidly defines and redefines its national identity in both domestic and international contexts. Being alert to the changing contexts within which social policy is discussed, designed and implemented is an important first step in being able to deconstruct and question social policy objectives – critical factors in ensuring robust democratic debate among students and scholars of social policy, but also among practitioners, beneficiaries and the wider public.


Althaus, Catherine, Peter Bridgman and Glyn Davis (2018). The Australian policy handbook, 6th edn. Crows Nest, NSW: Allen & Unwin.

Australian Bureau of Statistics (ABS) (2017). Inequality stable since 2013–14. Canberra: ABS.

—— (2005). Latest household income and wealth statistics. Canberra: ABS.

Australian Government (2018). Guides to social policy law: social security guide, indicators of vulnerability. Canberra: Australian Government.

Australian Institute of Health and Welfare (AIHW) (2017). Australia’s welfare 2017. Canberra: AIHW.

—— (2016). 25 years of health expenditure in Australia, 1989–2013/14. Canberra: AIHW.

Australian National Audit Office (ANAO) (2017). Design and implementation of the Community Development Program. Canberra: ANAO.

Bacchi, Carol (2017). Deploying a post-structural analytic strategy: political implications. Paper presented at the 4th Contemporary Drug Problems Conference: Making alcohol and other drug realities, Helsinki, 23–25 August.

—— (2009). Analysing policy: what’s the problem represented to be? Frenchs Forest, NSW: Pearson Education.

Bacchi, Carol, and Joan Eveline (2010). Approaches to gender mainstreaming: what’s the problem represented to be? In Carol Bacchi and Joan Eveline, eds. Mainstreaming politics: gendering practices and feminist theory, 111–38. Adelaide: University of Adelaide Press.

Bonoli, Giuliano (2013). The origins of active social policy: labour market and childcare policies in a comparative perspective. Oxford: Oxford University Press.

Bryson, Lois (1992). Welfare and the state: who benefits? London: Palgrave Macmillan.

Castles, Francis (1985). The working class and welfare: reflections on the political development of the welfare state in Australia and New Zealand 1890–1980. Wellington, NZ: Allen & Unwin.

Colebatch, Hal, ed. (2006). Beyond the policy cycle: the policy process in Australia. Crows Nest, NSW: Allen & Unwin.

Deeming, Christopher (2016). Rethinking social policy and society. Social Policy and Society 15(2): 159–75. DOI: 10.1017/S1474746415000147

Department of Human Services (DHS) (2018). Income management: New South Wales. Canberra: Australian Government.

Department of the Prime Minister and Cabinet (DPM&C) (2018a). Community Development Program: where the CDP operates. Canberra: Australian Government.

—— (2018b). About BETA.

—— (2018c). Behavioural insights for public policy: case studies from around Australia. Canberra: Australian Government.

—— (2018d). Closing the gap: prime minister’s report 2018. Canberra: Australian Government.

Department of Social Services (DSS) (2018). Income management and cashless debit card summary data, 31 March 2018. Canberra: Australian Government.

Dickey, Brian (1980). No charity there: a short history of social welfare in Australia. Melbourne: Nelson.

Dollery, Brian, and Stewart Webster (1995). Unemployment benefits and unemployment in Australia, 1950–89. The Australian Journal of Social Issues 30(1): 1–9. DOI: 10.1002/j.1839-4655.1995.tb01022.x

Edwards, Meredith (2017). Can good policy rescue politics? Paper presented at the Peter Karmel Forum on public policy, Canberra, 19 April.

Ewig, Christina, and Gastón Palmucci (2012). Inequality and the politics of social policy implementation: gender, age and Chile’s 2004 health reforms. World Development 40(12): 2490–504. DOI: 10.1016/j.worlddev.2012.05.033

Fowkes, Lisa (2016). Impact of CDP on income support of participants. In Kirrily Jordan and Lisa Fowkes, eds. Job creation and income support in remote Indigenous Australia: moving forward with a better system, 16–23. Canberra: Centre for Aboriginal Economic and Political Research, Australian National University.

Gregory, Bob (2004). Where to now? Welfare and labour market regulation in Australia. Australian Bulletin of Labour 30(1): 33–45.

Gupta, Nabanita, and Marianne Simonsen (2016). Academic performance and type of early childhood care. Economics of Education Review 53: 217–29. DOI: 10.1016/j.econedurev.2016.03.013

Head, Brian (2009). Evidence-based policy: principles and requirements. In Gary Banks, ed. Roundtable proceedings: strengthening evidence-based policy in the Australian federation, 13–26. Canberra: Australian Productivity Commission.

Howlett, Michael, and M. Ramesh (2003). Studying public policy: policy cycles and policy subsystems. Oxford: Oxford University Press.

Johnson, Carol (2011). Gillard, Rudd and Labor tradition. Australian Journal of Politics and History 57(4): 562–79. DOI: 10.1111/j.1467-8497.2011.01614.x

Jordan, Kirrily (2016). Better than welfare? Work and livelihoods for Indigenous Australians after CDEP. CAEPR Research Monograph No. 36. Canberra: Centre for Aboriginal Economic and Political Research, Australian National University.

Jordan, Kirrily, and Jon Altman (2016). From welfare to work, or work to welfare? In Kirrily Jordan, ed. Better than welfare? Work and livelihoods for Indigenous Australians after CDEP, 1–29. CAEPR Research Monograph No. 36. Canberra: Centre for Aboriginal Economic and Political Research, Australian National University.

Klapdor, Michael, and Don Arthur (2015). Welfare what does it cost? Canberra: Parliamentary Library of Australia.

Kral, Inge (2016). Only just surviving under CDP: the Ngaanyatjarra Lands case study. In Kirrily Jordan and Lisa Fowkes, eds. Job creation and income support in remote Indigenous Australia: moving forward with a better system, 19–23. Canberra: Centre for Aboriginal Economic and Political Research, Australian National University.

Lamy, Cynthia (2012). Poverty is a knot and preschool is an untangler. In Robert Pianta and Susan Sheridan, eds. Handbook of early childhood education, 158–74. New York: The Guilford Press.

Lasswell, Harold (1951). The policy orientation. In Daniel Lerner and Harold Lasswell, eds. The policy sciences: recent developments in scope and method. Redwood City, CA: Stanford University Press.

Lipsky, Michael (2010). Street-level bureaucracy. New York: Russell Sage Foundation.

Lloyd, Peter (2017). The evolution of tariff protection and wage protection in the late colonies and early federation. Economic Papers 36(4): 459–76. DOI: 10.1111/1759-3441.12183

Lowe, Toby, and Rob Wilson (2015). Playing the game of outcomes-based performance management. Is gamesmanship inevitable? Evidence from theory and practice. Social Policy and Administration 51(7): 981–1001. DOI: 10.1111/spol.12205

Markham, Francis, and Nicholas Biddle (2018). Income, poverty and inequality. Canberra: Centre for Aboriginal Economic and Political Research, Australian National University.

Marston, Greg, Catherine McDonald and Lois Bryson (2014). The Australian welfare state: who benefits now? South Yarra, Vic.: Palgrave Macmillan.

Marston, Greg, and Rob Watts (2003). Tampering with the evidence: a critical appraisal of evidence-based policy making. The Drawing Board: An Australian Review of Public Affairs 3(3): 143–63.

McDougall, Derek (2015). Edward Gough Whitlam, 1916–2014: an assessment of his political significance. The Round Table 104(1): 31–40. DOI: 10.1080/00358533.2015.1005360

Moran, Mark, and Ruth Elvin (2009). Coping with complexity: adaptive governance in desert Australia. GeoJournal 74: 415–28. DOI: 10.1007/s10708-008-9240-y

Nevo, Isaac, and Vered Slomin-Nevo (2011). The myth of evidence-based practice: towards evidence-informed practice. British Journal of Social Work 41: 1176–197. DOI: 10.1093/bjsw/bcq149

Nutley, Sandra, Huw Davis and Isabel Walter (2002). Evidence-based policy and practice: cross-sector lessons from the UK. London: UK Centre for Evidence-Based Policy and Practice.

O’Connor, Brendon (2001). The intellectual origins of ‘welfare dependency’. Australian Journal of Social Issues 36(3): 221–36. DOI: 10.1002/j.1839-4655.2001.tb01100.x

Oliver, Adam (2013). From nudging to budging: using behavioural economics to inform public sector policy. Journal of Social Policy 42(4): 685–700. DOI: 10.1017/S0047279413000299

Organisation for Economic Co-operation and Development (OECD) (2016). Social expenditure update 2016. Paris: OECD Directorate on Employment Labour and Social Affairs.

Pearson, Noel (2009). Up from the mission: selected writings. Carlton, Vic: Black Inc. Books.

Regan, Sue (2014). Australia’s welfare system: a review of reviews, 19412013. Canberra: ANU.

Ryan, Neal (2005). A decade of social policy under John Howard: social policy in Australia. Policy and Politics 33(3): 451–60. DOI: 10.1332/0305573054325738

Sanders, Will (2013). Changing agendas in Australian Indigenous policy: federalism, competing principles and generational dynamics. Australian Journal of Public Administration 72(2): 156–70. DOI: 10.1111/1467-8500.12014

Scullion, Nigel (2014). Minister Scullion: more opportunities for jobseekers in remote communities. Canberra: Australian Government.

Senate Standing Committee on Finance and Public Administration (SSCFPA) (2017). The appropriateness and effectiveness of the objectives, design, implementation and evaluation of the Community Development Program (CDP). Canberra: Parliament of Australia.

Shaver, Sheila (1987). Design for a welfare state: the joint parliamentary committee on social security. Australian Historical Studies 22(88): 411–31. DOI: 10.1080/10314618708595759

Smyth, Paul (2012). Securing the social conditions for individual autonomy in Australia. Australian Social Work 65(1): 1–5. DOI: 10.1080/0312407X.2012.659165

—— (2011). The British social policy legacy in Australia. In James Midgley and David Piachaud, eds. Colonialism and welfare: social policy and the British imperial legacy,175–88. Cheltenham, UK: Edward Elgar Publishing.

Staines, Zoe (2018). Ground-level impacts of remote-employment policy: social disadvantage under the Community Development Program. Journal of Australian Political Economy (82): 107–32.

Thaler, Richard, and Cass Sunstein (2009). Nudge: improving decisions about health, wealth and happiness. Camberwell, Vic.: Penguin Group Australia.

Titmuss, Richard (1974). Social policy: an introduction. In Brian Abel-Smith and Kay Titmuss, eds. Social policy: an introduction. London: Allen & Unwin.

van Hooren, Franca, Alexandra Kaasch and Peter Starke (2014). The shock routine: economic crisis and the nature of social policy responses. Journal of European Public Policy 21(4): 605–23. DOI: 10.1080/13501763.2014.887757

Venn, Danielle, and Nicholas Biddle (2018). Employment outcomes: 2016 Census Paper no. 5. Canberra: Centre for Aboriginal Economic and Political Research, Australian National University.

What Works Centre for Local Economic Growth (2018). Maryland Scientific Methods Scale.

Watts, Rob (1999). Warfare and the Australian welfare state. Journal of Australian Studies 23(60): 84–95. DOI: 10.1080/14443059909387452

—— (1987). The foundations of the national welfare state. Sydney: Allen & Unwin.

Wright, Sharon, Greg Marston and Catherine McDonald (2011). The role of non-profit organizations in the mixed economy of welfare-to-work in the UK and Australia. Social Policy and Administration 45(3): 299–318. DOI: 10.1111/j.1467-9515.2011.00770.x

Woolcock, Michael (2013). Using case studies to explore the external validity of ‘complex’ development interventions. Evaluation 19(3): 229–48. DOI: 10.1177/1356389013495210

About the authors

Greg Marston is professor and head of School of Social Sciences at the University of Queensland. He has been researching critical social policy and the politics of policy making for nearly two decades. He is active in basic income, employment and poverty debates and is undertaking research into the role of welfare states in transitioning countries to a low-carbon future.


Zoe Staines is a research fellow at the School of Social Science at the University of Queensland. She has previously held research and policy positions in the Queensland government, academia and the not-for-profit sector, and holds degrees, including a PhD, in criminology. Her research interests are social policy, criminology, governance and social justice.

1 Marston, McDonald and Bryson 2014.

2 Gupta and Simonsen 2016; Lamy 2012.

3 Titmuss 1974, 4.

4 Smyth 2012, 2.

5 Smyth 2011.

6 Dickey 1980.

7 Lloyd 2017.

8 Watts 1999; Watts 1987.

9 Shaver 1987, 411.

10 Shaver 1987.

11 Marston, McDonald and Bryson 2014; Watts 1999, 92.

12 Dollery and Webster 1995.

13 Smyth 2011.

14 Chifley in Smyth 2012.

15 Bryson 1992; Castles 1985.

16 Regan 2014.

17 Gregory 2004.

18 Smyth 2011.

19 McDougall 2015.

20 McDougall 2015. Medibank was later weakened under the Fraser government (1975–83) before being revitalised as Medicare under Hawke (1983–91).

21 Sanders 2013.

22 Marston, McDonald and Bryson 2014; van Hooren, Kaasch and Starke 2014.

23 Deeming 2016.

24 For example, see Johnson 2011; Ryan 2005.

25 AIHW 2017.

26 AIHW 2016.

27 AIHW 2017.

28 AIHW 2017.

29 Klapdor and Arthur 2015.

30 OECD 2016.

31 AIHW 2017.

32 ABS 2017; ABS 2005.

33 Markham and Biddle 2018.

34 Marston, McDonald and Bryson 2014.

35 Althaus, Bridgman and Davis 2018; Lasswell 1951.

36 Althaus, Bridgman and Davis 2018; Edwards 2017.

37 For example, see Colebatch 2006; Howlett and Ramesh 2003.

38 Nutley, Davis and Walter 2002.

39 Head 2009.

40 Marston and Watts 2003.

41 That is, studies that use either randomly or non-randomly sampled/allocated control groups to isolate the effects of the variable(s) being studied.

42 For example, detailed qualitative case studies; see also, the Maryland Scientific Methods Scale in What Works Centre for Local Economic Growth 2018.

43 Marston and Watts 2003, 145.

44 Woolcock 2013.

45 Lowe and Wilson 2015.

46 Colebatch 2006.

47 For example, see Nevo and Slomin-Nevo 2011.

48 Ewig and Palmucci 2012, 2491.

49 Lipsky 2010; Moran and Elvin 2009.

50 Moran and Elvin 2009.

51 Wright, Marston and McDonald 2011.

52 Bacchi 2009.

53 Bacchi and Eveline 2010, 112.

54 Bacchi 2017.

55 Pearson 2009; O’Connor 2001.

56 This required recipients of unemployment benefits to undertake job-search and job-preparation activities.

57 Deeming 2016.

58 DPM&C 2018a.

59 SSCFPA 2017, 6.

60 Scullion 2014.

61 DPM&C 2018a.

62 Jordan 2016.

63 ANAO 2017; Fowkes 2016.

64 SSCFPA 2017; Staines 2018.

65 SSCFPA 2017.

66 Kral 2016.

67 Jordan and Altman 2016, 10.

68 DPM&C 2018d; Venn and Biddle 2018.

69 Staines 2018.

70 Thaler and Sunstein 2009.

71 Deeming 2016.

72 Oliver 2013.

73 Bonoli 2013; Deeming 2016.

74 DPM&C 2018b.

75 DPM&C 2018b. See also DPM&C 2018c for case studies of BETA projects.

76 DSS 2018.

77 That is, if they are in receipt of certain categories of welfare, live within a specified income management area, experience financial hardship, fail to undertake ‘reasonable self-care’ and more.

78 Australian Government 2018; DHS 2018.

79 This initiative has been implemented in the Far-North Queensland communities of Aurukun, Coen, Hope Vale, Mossman Gorge and most recently Doomadgee.

80 For example, not sending children to school or being convicted of a crime.