Evaluation and Income Management Policy: Ten Years On

The Minister for Indigenous Affairs today announced (link here) the allocation of $10m per annum over four years for increased evaluation of Indigenous programs. Evaluation is a fraught business, particularly in Indigenous affairs. It can be highly useful, but can also be an absolute waste of money, particularly if it is used as a camouflage for taking necessary action in the face of real needs.

It is no coincidence that today’s announcement comes on the day of the scheduled release of the ANAO effectiveness audit into the Indigenous Advancement Strategy. I will post a link to that report when it becomes available. Postscript: here is the link

One area where evaluation has been at the centre of the policy debate is income management policy. While there are important issues around the policy of income management in its own right, it also offers us a window into the issues which surround evaluation approaches in Indigenous affairs.

This year marks the tenth anniversary of the introduction of Income Management into Australian social policy. Although a mainstream program, the vast majority of income managed welfare recipients in Australia are Indigenous.

I confess to being vexed regarding the impact of income management to date, its ongoing utility, its policy motivations, and its future usefulness. I previously wrote about it a year ago in this post: Imagining a Better Welfare System: Next Steps for Income Management (Link here). The points I made in that post remain highly relevant today.

So I was pleased to see the publication of a Special Issue of the Australian Journal of Social Issues (Volume 51(4)) devoted to the topic of Income Management (link here). My comments are necessarily limited to key points rather than comprising a comprehensive review. Each of the articles in the volume makes a positive contribution to providing a better understanding of income management, and they deserve to be read in their own right.

The contributors to the volume raise a host of issues, mostly persuasively argued, many following well-worn paths, and virtually all critical of the policy of income management.

The most powerful and now highly relevant article in my view is Rob Bray’s excellent summary account of the various evaluations of income management over the past ten years, (Bray: ‘Seven Years of evaluating income management - what have we learnt?’). Bray was a co-author for the largest and most rigorous of those evaluations, the 2014 evaluation of New Income Management in the NT. The findings of that evaluation were that there is no ‘consistent evidence of income management having a systematic positive impact’, and nor is there ‘evidence to indicate that income management has any effects at the community level, nor that income management, in itself, facilitates long term behavioural change’.

Bray also has a range of interesting comments on the methodological approaches and rigour other evaluations of income management to date. He points to the disjuncture between the views (both favourable and unfavourable to income management) of those subject to income management and the evidence of program success or failure. His point being that these are different things. Bray also discusses the limitations of the program evaluation approach, given that it is solely concerned with what has happened compared to the formal objectives of the program. In relation to this latter point, Bray notes previous research by Altman and Russell which argued that evaluation in Indigenous affairs has become an ‘obfuscating tool’ which is effectively divorced from the issues it seeks to assess, a view which I find difficult to refute. Certainly, it has become the case that evaluations and their findings have become the currency of political debate over income management since the NT Emergency Response.

Nevertheless, Bray goes on to point out that the disjunction between the luke-warm at best evaluation findings on income management and the ‘persistent view within government that the program should be maintained, and indeed expanded’ requires explanation. Unfortunately, while providing ample evidence of the selective interpretation by Ministers and policymakers of the outcomes of the various evaluations of income management, in my view he fails to deliver an adequate answer to his own question.

However a number of the other contributions in the volume either explicitly or implicitly advance proposed explanations.

Marston, Cowling and Bielefeld in their contribution (‘Tensions and contradictions in Australian social policy reform: compulsory income management and the NDIS’) point to the apparent inconsistencies between the provision of choice under the NDIS reforms and the imposition of conditionality under income management, essentially concluding that Australian social policy is characterised by a confused form of ‘authoritarian –liberalism’. In effect they argue that public policy is confused and internally inconsistent. They may well be right, but they also under-emphasise the substantive differences in the intent of disability support and income support.

Lovell in her account of the parliamentary debates around the introduction of the income management legislation (‘The normalisation of income management in Australia: analysis of the parliamentary debates of 2007 and 2009-10’) lays out a range of views from both the major political parties which rationalises income management as a means of addressing passive welfare, reducing welfare dependency, encouraging ‘responsible’ behaviour and ‘self-improvement’, and addressing community dysfunction. Lovell characterises these rationales as elements of a policy designed to create ‘the ideal, ‘active’ neoliberal citizen.

I found Lovell’s analysis not entirely persuasive since the drivers of policy are not necessarily entirely encompassed by the rationales put forward in the Parliament. Nevertheless, she does add texture to the justifications proffered for income management.

An interesting contribution by Bray, Gray, Hand and Katz (‘Social Worker Assessed Vulnerable Income Management’) assesses a minor element of income management, the Vulnerable Welfare Payment Recipients Measure, which involves social worker assessments of vulnerable individuals in determining whether to apply income management. There appear to be benefits for the few people on the scheme, although it may well be that the involvement of social worker support assists in driving those outcomes rather than the program itself. Implicit in the operation of this measure is the ideal or aspirational vision of what income management might become. Yet it is apparent that the bureaucracy and government perceive it to be administratively expensive and effectively dispensable in favour of a more streamlined and automated approaches.

Finally, Altman and Klein in separate articles provide more searing critiques of income management: (Altman: ‘Blind-sided by basics: three perspectives on income management in an Aboriginal community in the NT’; Klein: ‘Neoliberal subjectivities and the behavioural focus on income management’).

The implicit answer to Bray’s question advanced by each of them is that income management is the product of an ideological commitment to market and neoliberal prescriptions which on their face have all the hallmarks of seeking to reshape Indigenous people through ‘behaviour modification’, market incentives, into homo economicus.

Notwithstanding the use of the language and jargon of ‘critical theory’ overtly signalling an ideological commitment to the progressive or left end of the political spectrum, there are important points made by both Altman and Klein which deserve serious consideration and assessment by policymakers.

Altman’s perspective is intuitively more attractive; he argues that income management in itself is only part of the more general exercise of what he terms’ the overarching state project of heightened governmentality in the name of improvement’. Implicit in his argument is that the answer to Bray’s question about the persistence of the Government commitment to income management can be found in the blind commitment of policymakers to this ‘overarching project’.

My own perspective is somewhat different.

I don’t deny that there is a government and policymaker mindset about Indigenous people and communities, and it is not necessarily accurate or well informed. It tends to seek simplistic solutions to what are complex problems, and in this sense I agree with Klein’s conclusion that policy should accept and address social and political complexity much more than it does.

The critics of income management however have allocated extraordinary efforts to parsing and dissecting the policy of income management from every conceivable angle (I am not referring just to the contributors to this volume when I say this). They allocate very little attention to discussing the existence of community violence and dysfunction, indeed, it is as if it doesn’t exist. I don’t deny that governments can frame social circumstances in tendentious ways, but it seems to me at least that the existence of deep-seated disadvantage characterised by violence against women and children, of child protection issues, of substance abuse including alcohol abuse are real issues which deserve attention by governments. Yes they can be framed in ideological ways, but they shouldn’t be ignored.

Just as Klein argues for acknowledgement of social and political complexity on the ground, I would argue for greater acknowledgment of the complexity and challenges of policymaking within government. 

Thus, in terms of income management’s origins, it is impossible to deal with it without dealing with its inclusion as a key measure in the NTER. This was in my view a largely political exercise, aimed squarely at a particular segment of the wider electorate in the lead up to the 2007 election, with conservative ideological underpinnings. Nevertheless, it was also intended by the then Government (however misguidedly) to drive positive change.

The decision in rolling out income management in the NT in 2007 to over-ride the operation of the RDA was always a serious issue, and sent an extremely regressive signal particularly to Indigenous Australians. However, the imposition of income management was arguably substantially outweighed in its impact on Indigenous people by the other more confronting NTER measures: the compulsory acquisitions of community lands, the use of the uniformed ADF, the compulsory health checks and so on.

Once legislated in 2007, with Labor’s support (arguably necessary to neutralise the prospect of a race based election campaign in 2007, but extremely regrettable nevertheless) income management became embedded in the social policy foundations in Australia. Labor, hamstrung by their prior support and their lack of Senate numbers, moved incrementally to reform the program, including to make clear in legislation that the RDA was not set aside and that the scheme was a special measure, and to remove from the ambit of income management aged and disability support pensions.

The advent of so called New Income Management in 2012 placed income management on an independent footing. It was administered from outside the Indigenous affairs bureaucracy as a mainstream program.

Was there a justifiable rationale for the program? In my view, the answer then was yes. It related to the terrifyingly destructive impact of alcohol on those communities. Admittedly, income management was not the only tool available to limit alcohol abuse, but it was important to do all that was possible, using all the policy tools available. In remote Indigenous communities, where a majority of community members of working age were on welfare, the scheme operated to limit the discretionary income available across the community for purchasing alcohol by around 40 percent. Much of the expenditure on alcohol was done outside legal channels, and there were and are few reliable statistics on illegal expenditures.

Whether income management was successful or not is not my point, rather I am arguing that it was worth attempting in an effort to minimise the adverse consequences (including violence against women and children) of alcohol abuse. Welfare is inherently conditional, the ultimate sanction being the threat of ceasing payments, a very real issue across remote Australia at present. Altman is on strong ground however in his observation that there was virtually no planning, no detailed and consultative program design work which predated the roll out of income management in 2007. This merely reinforces the predominance at the time of electoral over policy considerations.

Bray notes in his article that the New Income Management evaluation he was involved in reported:

‘Lack of substantive change was also identified for other community problems, including child outcomes and excessive drinking. Further evidence of any improvement across a diverse set of NT outcomes such as child mortality, education, alcohol-related crime, and hospital admission, was absent’

These results appear determinative, although it is not clear how robust they are, nor the extent to which they disguise intra-community variations in outcomes.  However retrospective justification is not the basis for prospective policy. It seems that ten years on it is time for a root and branch review of income management policy design.

Over time, there have been a swathe of trials and pilots, all with their own program rules etc. It is time they were rationalised and amalgamated into a single universally applicable scheme. In designing a new program, serious consideration needs to be given to the evaluation work undertaken to date. While not definitive, the evaluation conclusions provide a benchmark against which any future model of income management ought to be tested.

The unwillingness of governments to date either to implement universal income management across all welfare recipients nationwide, or to go back and remove it suggests a government and bureaucracy frozen in a spotlight. The shape of such a universal scheme might be similar to or radically different from the schemes currently in operation. For example, it might be a universal voluntary scheme; or a universal scheme based on social worker identification of vulnerable people. But it is time to ditch the trials, set aside the pilots, and commit to a national approach to conditional welfare.

In relation to remote communities, the lessons of the last ten years include that the provision of adequate support services and supporting policy frameworks are crucial in driving positive outcomes. A single program cannot carry the weight of driving positive social outcomes. But more importantly, whether one agrees or disagrees with Altman’s assessment of government motivations, the insight implicit in Altman’s analysis of the state of play in Maningrida is that Indigenous choice is important, and while governments and policymakers might believe that they can use policies (even non-coercive incentives) to shape Indigenous peoples behaviours, it will ultimately be counter-productive.

The formulation of complex social policy to operate in a highly complex social and political settings exhibiting high levels of heterogeneity is no easy task. While policy critique is important and necessary, it seems to me that critics will be more influential if they go some way towards seeing the world from the perspective of policymakers. By the same token, policymakers have an absolute responsibility to seek out and incorporate local and individual perspectives notwithstanding the pressures they face to develop and implement national level policies. In the absence of one or both of these elements, policy dialog will largely be non-existent, and we will all be worse off.

Finally, in the light of today’s announcement of a renewed focus on evaluation in the Indigenous policy domain, it is worth focussing on the extent to which government has responded to the substantial resources allocated to date in evaluating income management policy.

Personal Disclosure: I worked as a ministerial policy adviser to Minister Jenny Macklin from 2008 to 2011, and as a senior officer in the Department of Families, Housing, Community Services and Indigenous Affairs  from 2011 to 2013.