Wednesday, 27 January 2016

Imagining a Better Welfare System: Next Steps for Income Management


On 20 January, the Assistant Minister to the Prime Minister, Alan Tudge, published an op ed in the Australian (‘Communal Despair Begins with Grog Abuse’) outlining the current state of play on the proposed new Welfare Debit Card, the latest version of income management for a small cohort of the most disadvantaged welfare recipients across the nation.

The welfare debit card, named ‘the Healthy Welfare Card’, was proposed by Andrew Forrest in his review of Indigenous training and employment, Creating Parity, in 2014, and involves 80 percent of a welfare recipient’s income being loaded onto a debit card  which cannot be used to purchase alcohol, gambling or to access cash.

The Government’s proposed welfare debit card is not Indigenous specific and will apply to all welfare recipients within the trial site locations. Nevertheless, the two locations chosen to date (Kununurra and Ceduna) both involve a majority of Indigenous welfare recipients. The reality is that income management of welfare payments emerged from and retains strong roots within the Indigenous policy domain based in large part on the inability of governments over the past decade to make inroads on reducing the high levels of personal and community disadvantage amongst Indigenous citizens in some locations.

Tudge appears to be pursuing two political objectives. First, to maintain the momentum in implementing the new income management scheme following the passage of legislation late last year with Labor Party support authorising three trials of the welfare debit card for up to ten thousand recipients; and second, to lay the foundations for broader (perhaps universal) implementation of the debit card in the Government’s next term.

In this latter context, the Australian’s Sarah Martin, writing the same day as Tudge’s op ed appeared (‘National cashless welfare card plan’), and presumably based on briefing from the Assistant Minister, began her column as follows: “A cashless welfare card aimed at stemming alcohol abuse would be rolled out across the country under a welfare reform the Turnbull Government is considering taking to the election”

Tudge’s comments elicited an article in The Australian on 25 January in response by the Social Justice Commissioner, Mick Gooda, critical of income management generally, and the new iteration of income management being advanced by the Government (‘The welfare card is not the solution to alcohol abuse’). In particular, Gooda asserts that the welfare card ‘will disproportionately affect Aboriginal and Torres Strait Islander people, and claw back our hard-won rights and freedoms’. He also questions the card’s efficacy in reducing alcohol harm, citing a lack of evidence for such claims. And Gooda begins his op ed by noting that ‘old wounds are being re-opened. Many of our people are being forced to revisit the past trauma of income management and stolen wages’.

This latter point is particularly salient in terms of considering how to proceed with the new iteration of income management. The implementation of a policy which engenders strong opposition at a deep emotional level amongst its recipients will always be highly problematic. Assistant Minister Tudge appears to be cognisant of these challenges as he has stated he is keen to find a third trial site in a major regional centre, where around 70 percent of welfare recipients are non-Indigenous.

How then should we assess the merits of the policy of income management, and in particular the Government’s particular approach to taking it forward? Not only is it complex policy in its own terms, but it has taken on highly symbolic ideological status, for and against, in both the right and left of the political spectrum. This accentuates its polarising impact within the ongoing policy debate on welfare policy, particularly in relation to Indigenous welfare issues. So for example, those on the left emphasise the alleged infringements of rights embedded within the policy, whereas conservatives emphasise the importance of mutual obligation and linking taxpayers contributions to welfare to societally sanctioned outcomes.

It is worth contextualising income management policy so as to outline the sequential development of variant models over the past decade, and to place it in the longer arc of the history of public policy development.

As mentioned above, the roots of income management are to be found in the indigenous policy domain. The current proposals are the fourth iteration of income management to be rolled out over the past decade, and importantly all three predecessor versions continue to operate. I note in passing that there are a number of sub-categories to each of the three income management schemes discussed below.

In Cape York, the welfare reform agenda funded over the last decade involved a system where a Family Responsibilities Commission composed largely of community members has the power to take decisions to effectively income manage welfare recipients who fail to ensure their children attend school, breach the law, or fail to meet their tenancy obligations and thus are contributing to social dysfunction. This system was established at the request of the local communities and the Cape York Institute, and is virtually entirely Indigenous in its coverage.

The second scheme is the so called New Income Management program utilising the NT Basics Card put in place in 2010 in a revision of the original income management program introduced as part of the NT intervention in 2007, which quarantines 50 percent of a welfare recipient’s income. This scheme applies to all welfare recipients across the NT (but not aged or disability pensions), but is almost wholly Indigenous in its coverage. The original program was protected from legal challenge by a statutory exclusion of the Racial Discrimination Act, however the New Income management program has been designed as an explicitly non-discriminatory scheme.  

The third scheme is Place Based income management which was rolled out by Labor in a number of locations across Australia, both Indigenous and non-Indigenous.

The three existing schemes have a limited geographical focus (although the NT scheme applies across the NT), and are either explicitly or implicitly conceptualised as trials within the bureaucracy and government.

Turning to the longer arc of history, income management can be seen to be the product of previous modes of ‘administering’ Indigenous people’s lives. As is comprehensively documented in Tim Rowse’s insight-laden history of the origins of Indigenous citizenship in Central Australia (‘White Flour, White Power’), the entire colonial project in remote Australia was largely based on the provision of rations to Aboriginal people both through state sponsored ration depots/settlements and state subsidised pastoral enterprises.

Rowse explores the administrative rules which linked the provision of in-kind rations to Aboriginal people’s readiness and capability to work, the pressures to move from in-kind rations to more fungible cash payments, and the differing perceptions of both Aboriginal and settlers regarding the basis for the respective roles of the parties to the transactions involved.

At a more abstract level, Rowse argues that rationing policies were embedded in the policy of assimilation, and that the development of internal contradictions in the administration of rationing were based in large part on the reluctance of Aboriginal groups to leave their country to seek employment. Rowse sets out in detail how these internal contradictions in the administration of rationing ultimately led to and facilitated the eventual acceptance of remote Aboriginal community members as citizens, and contributed to the development of the Whitlam Government’s policy narrative of self-determination as a successor to paternalism and assimilation.

One of Rowse’s themes is to recognise the continuities inherent in policy shift from assimilation to self-determination. For present purposes, it is enough to note that these continuities continue into the present day and that access to welfare and the subsequent adoption of income management (albeit in a somewhat stop/start fashion) is a direct policy descendant of the ration depots of the nineteenth century. In particular, the internal contradictions remain, as does the likelihood that Aboriginal and Torres Strait Islander peoples will engage with and interpret the policy on their own terms and in their own ways.

I focus on the long road to our present policy position to make the point that policy development is iterative and the options available are constrained not only by political imagination and political determination, but also by the largely invisible yet powerful projection of incremental inertia, which exercises an extraordinary influence on the imagined parameters of bureaucratic and political feasibility. In other words, where we are heading is often largely determined by where we are coming from.

There have been a number of policy evaluations of income management over the past five years. While I don’t have the capacity to closely review them here, it is fair to say that while the conclusions of the more recent evaluations are positive, the conclusions are far from unequivocal. Moreover, the most comprehensive and significant review was the evaluation of New Income Management in the NT finalised in 2014 by a team of researchers from the Social Policy Research Centre at UNSW, the Institute of Family Studies and the ANU. This comprehensive and independent evaluation found that the current New Income Management program in the NT has not met its objectives. Interestingly, the DSS website (see hyperlink above) referencing the various evaluations omits to summarise the results of the NT evaluation.

What does emerge from a cursory scan of the various evaluations undertaken to date is that they are highly sensitive to the design parameters and the formal objectives for each scheme, and that as one might expect, each scheme has its strengths and weaknesses, and amongst those being income managed, there is a diversity of views about the utility of the schemes. This suggests that there is a need for high levels of clarity regarding the policy objectives of income management schemes if we are to be able to use evaluations as determinative guides to policy development. Unfortunately, the complexity of the schemes in place, and the innate amorphousness of the policy and political drivers for the schemes make this a virtually unattainable aspiration.

My own assessment of the current policy settings for income management is an amalgam of the concerns and aspirations of both the left and the right. In an ideal world, I would support the provision of maximum autonomy for welfare recipients in the use of their funds. Such a view has the attraction of being philosophically straightforward, and like the free market theorists who advocate pure consumer sovereignty, it asserts that no one is better able to assess an individual’s interests than the individual herself. I also believe that we have to give serious weight to the multiple considerations canvassed in the evaluation of the NT scheme. In short, we cannot merely assert or assume that income management is or will be an unqualified success, and to the extent that it is pursued as an element of our welfare policy settings, we need to be prepared to reassess and redesign to ensure that it delivers net benefits.

It is clear however that income management has established a foothold in our current welfare policy framework which is based on something more than a broadly accepted evidence base and a series of technically narrow policy evaluations. This is reinforced when one notes the comment in the NT evaluation that the Department of Social Security requested that the evaluation not assess the cost effectiveness of the program, ostensibly because the requisite data was not available (page 3, summary report). One of the most salient factors in assessing Government programs is cost effectiveness; so clearly other forces are at work here.

Apart from the fact that the evaluations of income management as implemented so far in Australia are equivocal at best, and that the program exists for better or for worse, I consider that there exists a moral imperative for flexible policy responses in relation to the activities of governments in their engagement with Indigenous communities, and to a lesser extent, in their engagement with disadvantaged citizens generally.

In particular, if Governments are funnelling substantial resources into remote communities comprised of citizens with poor education, limited financial literacy, poor health, limited civic governance skills, and with strong culturally-based social capital, but weak social capital in areas involved in engagement with the broader society, then in my view government has a responsibility to understand the implications of those resource flows for the communities, and if necessary, act to meliorate any adverse consequences. It so happens that the source of much of those resource flows is the welfare system. Accordingly I see a rationale for Government intervention, especially in circumstances where the resources flowing (in this case largely through welfare payments) are contributing to social dysfunctional and family violence which impacts vulnerable community members such as he aged, women and children.

This rationale for Government action is not necessarily an argument for income management, as there may be other tools to achieve the objective I see as legitimate. However, income management is one of the potential tools available to government, and in my view should not be dismissed out of hand.

In particular, when faced with situations where welfare payments are subsidising dysfunctional behaviours, often involving violence against women and children, I can see little justification for maintaining a philosophically pure position based on the assumption that individuals know best. Moreover, while in many locations a preponderance of welfare recipients are not engaged in or subject to dysfunctional activities, I would prefer to see constraints on their access to fungible resources in order to minimise risks to women and children from the minority who do engage in dysfunctional behaviour. This point is strengthened if it is accepted that dysfunctional behaviour is likely to impact beyond the family and household and have community wide adverse impacts.

Once it is accepted that there may be a justification for income management of welfare in some circumstances, the next step is to determine the detailed specification of those circumstances. This leads on to the detailed design of an effective income management scheme including the key design parameters which should be in place. I would add though that the design work should be broadly based, and extend to the design of the support structures which are an essential adjunct to successful income management (see below).

Key challenges facing policy designers are to ensure the scheme assists those most in need (who are largely those who are most disadvantaged) without adversely affecting those welfare recipients who don’t require income management; how to ensure the scheme does not breach the Racial Discrimination Act; how to deal with welfare recipients who are especially vulnerable and under duress; and how to minimise the administrative costs of the scheme and keep them within affordable limits while simultaneously ensuring that individual welfare recipients’ circumstances are monitored and adjustments made as necessary. These policy design issues are all substantial challenges for policymakers. Failure to overcome any one of these challenges engenders the development of the sorts of internal contradictions which ultimately undermined rationing regimes and will similarly undermine income management.

Assistant Minister Tudge deserves significant credit for progressing the next phase of policy development for income management in a consultative and methodical way, particularly in relation to its impacts on Indigenous people. The work of the Forrest Review and aspects of the design report for the Empowered Communities project, led by experienced Indigenous leaders such as Noel Pearson and Ian Trust provide a considered and robust foundation for the next steps. Nevertheless, in my view, there remains a need for further work by the Commonwealth Government to clarify its policy objectives for income management, and to establish a transparent and independent process for ensuring that issues which emerge in the implementation of income management schemes are addressed.

Moreover, there needs to be recognition that income management can never be an end in itself and must be seen as a tool to achieve better social outcomes. It will rarely be able to undertake the heavy lifting required in terms of welfare reform on its own, and will always require substantial ancillary support services. Indeed, there is an argument that income management should eventually be seen as the ancillary support for other more proactive social policy initiatives.

For example, the NT evaluation noted that one of the reasons why many income managed recipients in the NT have a positive attitude to income management is that it provides a low cost and accessible banking service. This suggests that there is a more fundamental need amongst disadvantaged citizens, particularly in remote Indigenous communities, for access to improved financial services, for financial literacy, and for culturally-informed banking services. In the US, the federal government has a number of schemes in place to incentivise and even require banks to provide accessible services to disadvantaged citizens, and there are strong grounds for believing that there is merit in the Australian Government moving in this direction too.

It follows that I am broadly supportive of the high level policy directions being pursued by the Government, and am particularly pleased to see a commitment to a non-discriminatory policy framework which is a major step forward from the 2007 NT Emergency Response framework. However there appears to be scope for a much more adventurous reconceptualisation of the purpose of income management, and for a much stronger focus on delivering the ancillary support services which will reinforce the potential positive impact of income management, and potentially lead to the a new phase where it is no longer required.

Notwithstanding my support, I do have six critical but I hope constructive comments regarding the approach being adopted by Assistant Minister Tudge:

·         While incrementalism is a fact of life in public policy, it seems particularly short sighted to continually develop new income management schemes while leaving their predecessor ‘trials’ in place. The result is a chaotic assemblage of similar yet different income management models, without a coherent policy rationale articulated by government. Government needs to simplify the number of different income management designs in operation.



·         It is time for the Government to move beyond ‘trials’ with limited geographic focus and develop a universal model, that is , one that potentially applies to all welfare recipients. The next term of government will see the completion of a decade since the contentious and poorly implemented initiation of the NT income management scheme.



·         While on its face Assistant Minister Tudge’s public statements foreshadow a move to develop and implement a universal income management scheme, it seems likely that part of his motivation in spruiking this possibility as a pre-election policy commitment is to build support and momentum within the Government for such an outcome. The implementation of a universal nation-wide scheme would involve considerable costs to the budget (in the hundreds of millions, potentially even in excess of a billion dollars), and would inevitably be strongly opposed by Finance and Treasury. The Cabinet endorsed budget rules require all new expenditure to be offset, yet Assistant Minister Tudge controls no other major programs from which to source offsets. Either the Prime Minister or the Minister for Social Security Christian Porter would need to come on board were a wide ranging expansion of income management to have any prospects of support within the Cabinet.



·         Alternatively, a more cynical interpretation of the Assistant Minister’s public statements would see the spruiking of what is a mere possibility without cabinet endorsement as just kicking the policy can down the road.



·         The focus on alcohol by Assistant Minister Tudge raises the question whether the Government is using the prospect of universal income management at some yet to be specified time in the future as a means of reducing broader political pressure to take robust action on the supply of alcohol to remote Indigenous communities and towns. The policy evidence is clear that a simple measure such as introducing a volumetric tax on alcohol would make significant inroads on alcohol abuse. Unfortunately, both the Government and the Opposition, at both state and federal levels, are in thrall to the substantial electoral funding from the alcohol industry, and sensible reforms such as this appear not to be on the table.



·         Finally, while the Government focusses an extraordinary amount of attention on what is in essence a thin sliver of our welfare system, broader public scrutiny and debate on the effectiveness and efficacy of our welfare system overall is curtailed.

Income management is a crucial social policy issue for Australia. It has particular resonance in the Indigenous policy domain, and continues to have a polarising impact notwithstanding the apparent bipartisan support for the policy. There appears to be a strong bipartisan consensus that any income management policy ought not to be racially discriminatory and this is driving a focus on broadening the base beyond discrete Indigenous communities who arguably stand to benefit most from the successful implementation of a well-designed and appropriately supported policy.

Nevertheless, income management raises a tangled web of issues, ranging from different views on its philosophical foundations, through to more concrete issues such as the design of effective scheme, determining the appropriate support mechanisms which are required to complement an effective scheme, and identifying how to minimise the ongoing duplication of competing models in place, and how to fund the apparently desired shift to more universal scheme in the face of what appear to be insurmountable budget constraints.

Charles Lindblom, an American political scientist was best known for his explanation of public policy development as ‘muddling through’. Income management of welfare can potentially contribute to the continued safety and protection of many disadvantaged citizens, particularly women and children, yet after a decade of ‘muddling through’, we should not be surprised to find that the current state of play is a public policy muddle.

The challenge for Assistant Minister Tudge, and indeed the Government and Opposition together, is to learn from the models of income management implemented over the past decade, learn from the evaluations undertaken, and move beyond the policy foundations established to date to a more flexible, coherent, non-discriminatory and national policy model for income management. Such a model would be located within a matrix of complementary government services and initiatives which operate flexibly to ensure that welfare recipients are encouraged and incentivised to build the skills and capacities they need to move beyond welfare.

Ultimately, we should aim to prioritise and establish much more robust and innovative nationally implemented policy frameworks for reduced alcohol and substance abuse, for reduced family violence, for increased financial literacy and improved banking services which allow our national welfare policy frameworks to move beyond income management.

This is a task of policy and strategic imagination, breaking free of the notion that our pathways forward are constrained by where we have come from. It truly is a policy objective worthy of our aspirations.