Mr Gary Banks AO

7 July 2009, National Library of Australia theatrette, Parkes Place, Canberra

Mr Gary Banks

Chairman

Productivity Commission

Lecture Three in Reconciliation Australia’s Closing the Gap Conversations Series

Click here to download a copy of the lecture

Click here to download an audio file of the lecture (12MB)

Lecture Transcript

Introduction

I feel greatly honoured to have been invited by Reconciliation Australia to contribute to this important series of Closing the Gap ‘conversations’. Reconciliation Australia has done much to stir national consciousness of the pressing need to remedy Indigenous disadvantage — not only for the wellbeing of the first Australians, but also that of Australian society generally. It has achieved this through a commonsense forward-looking approach, based on ‘partnerships for success’ that involve organisations and people from all parts of the Australian community.

My own organisation, the Productivity Commission, is pleased to count itself among those ‘partners’, including in its role as Secretariat to the Steering Committee whose report I will talk about today. This Report — Overcoming Indigenous Disadvantage (OID) — benefitted from Reconciliation Australia’s cooperation in a number of ways, not least the many beautiful photos that adorn the pages of the Overview, and the case studies drawn from its Indigenous Governance Awards (in another partnership with BHP Billiton). I have also personally benefitted greatly from my involvement as a judge on those awards, learning much about good governance in challenging settings from the organisations participating.

The OID Report was publicly released late last week in conjunction with a COAG meeting in Darwin that had a particular focus on Indigenous policy. Many of you will have seen the Report by now — or at least its more accessible Overview — and read about its contents in the press. So I will not be offering a ‘report briefing’ today, but rather, a narrative around the report, picking up findings and implications that seem to me of particular significance. In so doing, I will often be speaking in my own right or as Chairman of the Productivity Commission, not just as Chair of the Steering Committee responsible for the Report.

The Prime Minister’s widely reported statement that he was ‘devastated’ by what this Report reveals is a sentiment that would be shared by many Australians. There are at least two reasons for this. One is the wide gulf between the circumstances of Indigenous and other Australians, documented in page after page of the Report. The second is the limited progress we have made in bridging it.

It is now seven years since governments made a commitment to work together in new ways to tackle the root causes of disadvantage. In an important break with the past, in 2002 they agreed to commission a ‘regular report’ to monitor national outcomes in a systematic way — and thereby hold themselves accountable. My expectation when presenting the first OID report in 2003, was that many of the disparities evident at that time would have begun to narrow by now. Six years and three reports later, that has been clearly achieved for only about 20 per cent of the indicators. In 10 per cent of them things have actually gotten worse.

There are, however, a number of caveats that need to be made. Perhaps the most important is that for half the indicators in the Report the data are still not good enough for us to know whether there has been any progress. (If we look at the extent of progress among only those indicators for which trend data are available, the ‘improvement’ ratio doubles to 40 per cent.) It is also the case that much of the data only goes to 2006, and therefore cannot shed light on outcomes since more recent initiatives. So, despite considerable improvements in data since this series of reports commenced, some major deficiencies still prevent the Report from completely fulfilling the role COAG intended for it.

Role of the Report

The Report has been purposefully designed — following consultations with governments and Indigenous people around the country — to focus on areas that really matter in overcoming Indigenous disadvantage. Evolving from earlier ground-breaking work by the Council for Aboriginal Reconciliation and the Ministerial Council (MCATSIA), the Report is structured around a strategic framework of outcome indicators, based on a ‘preventative’ model. This framework contains not only high level goals and related high level indicators, but also a second tier of indicators in ‘strategic areas’ where, in many cases, prior improvements in outcomes are needed and where policy actions could be expected to have an effect in shorter timeframes.

The framework also recognises that disadvantage has multiple dimensions and multiple causes, many of which interact with each other. For example, educational outcomes for Indigenous children depend not only on the effectiveness of available schooling services, but also on students’ health, and, importantly, the circumstances of their homes and communities.

The OID Report has evolved over time, to ensure that it retains its relevance and meaningfulness to Indigenous people and governments. In this fourth edition, there have been some significant amendments — the higher level indicators now include the six new COAG targets, and the lower tier has been modified to align it more closely with the COAG Working Group on Indigenous Reform ‘building blocks’. Extra indicators have been added, but none have been lost. As we have done after previous reports, we will be conducting consultations with Indigenous people to get their feedback on the changes to the Report and their ideas on how it may be further improved.

At the time of the last report, the question of whether government accountability should be elevated further by them signing up to specific targets and timeframes remained contentious. That COAG has since agreed to what it justifiably describes as ‘ambitious’ targets is to be commended. However there is a commensurate need for data of sufficient quality to enable progress in the targeted areas to be properly assessed. Unfortunately, among the indicators mentioned earlier as not yet having adequate trend data are four of the six COAG targets.

These include what is arguably the most important, and certainly most symbolic, indicator of Indigenous disadvantage, life expectancy. This year’s Report contains good and bad news for this indicator. The good news is that the gap in life expectancy between Indigenous and other Australians is smaller than we thought. At 10 years for women and 12 years for men, it is still clearly unacceptable and justifies COAG’s efforts to eliminate it within a generation, but that task now seems more tractable than under the previous estimates of 17 – 20 years. (The new estimates also mean that international comparisons now look less invidious than they did in 2007 when I was fielding questions on this matter at the OECD’s international conference on ‘Measuring the Progress of Societies’ in Istanbul.)

The bad news, as will have already become apparent, is that there is no news as to whether the life expectancy gap has actually narrowed. The lower estimates simply reflect more accurate mortality data and better methodologies. I understand from our statistical agencies that we can at least be more confident that we now have a robust basis for measuring progress in the future.

Progress at the ‘headline’ level?

The reality is that we cannot expect rapid progress in an indicator such as life expectancy, representing as it does the culmination of many influences that have accumulated over the lifecycle. That said, the average life expectancy for Indigenous people is reduced by much higher mortality rates at the very start of life. It is unqualified good news, therefore, that there has been a significant improvement in mortality rates for the crucial period from birth to 12 months. Indeed, this is one of the few higher level indicators where the gap has actually narrowed — although infant mortality is still 2 to 3 times higher than for non-Indigenous infants.

Among the other top indicators, outcomes have generally been significantly better in ‘economic’ than ‘social’ areas. For example, there was improvement in Indigenous employment rates and in average incomes in the period to 2006. This is probably best interpreted as Indigenous people sharing in the benefits of buoyant economic times, as the gap with the rest of the Australian community did not get any smaller in that period. It may also prove challenging to hold this ground in the more difficult economic times ahead.

Among the social indicators, there has been a further rise in substantiated cases of child abuse and neglect. Substantiations for Indigenous children have risen — from 4 times to 6 times the rate for non-Indigenous Australians in the eight years to 2007-08. This is concerning, and has been picked up in much media reporting since the Report was released. However, few have acknowledged the qualification we emphasised, which is that the higher rate partly stems from improved reporting and follow-up in response to stricter government requirements for notification. As the Minister, Jenny Macklin, remarked in Darwin, this at least needs to be seen as a positive feature that is bringing the reality of child abuse ‘out of the shadows’.

The story is somewhat different for another key social indicator, imprisonment rates, where there has also been a sharp rise throughout this decade — especially for women — one that is less likely to be distorted by variations in the accuracy of reporting. However, the most recent data, for 2008, provide cause for hope that imprisonment rates may have peaked. This was the first year that rates did not increase. Indeed, there was a small absolute decline — which looks more significant when compared to what would have been projected based on the previous trend. It is also relevant that this reduction coincided with a period of stricter policing in many communities.

However, no such reversal is apparent in the available data on juvenile detentions. Although the most recent data is for 2007, the rate for that year is 30 per cent higher than at the beginning of the decade — and 28 times greater than for non-Indigenous juveniles.

Of the four higher level indicators relating to educational performance or participation, the only discernable improvement is in Year 12 attainment, which rose by 5 percentage points between 2001 and 2006, but is still well below rates for other students. Given the importance of education, it is particularly concerning that there has been no improvement in learning outcomes at school, with performance for Indigenous students dropping sharply in successive class years relative to other students.

In sum, the overall picture for the high level indicators can be described as ‘mixed’ at best. For the six important COAG targets there has been an improvement in absolute terms in three, but only one (infant mortality) involves a reduction in the ‘gap’.

It is also true, as noted previously, that most of the indicators at this level are the product of many influences over many years and will be slow to turn around. For this reason, indicators in the second tier of indicators have been chosen to provide a more strategic focus on areas that can have positive flow-on effects, and that are more amenable to policy-induced improvement over shorter timeframes. Again, I will not try to be comprehensive and will abstract from detail, given that the Report already provides both.

Getting started in life

There are a range of indicators in the Report that relate to key aspects of early child development, particularly health and learning outcomes. As the first strategic area for action, this has been strongly supported. Too many Indigenous peoples’ lives are compromised when they have barely begun.

Three of these indicators have a direct bearing on child mortality rates. (As noted, these rates have shown some improvement but there is still some way to go.) One is birthweight, with twice as many Indigenous as non-Indigenous mothers having a low birthweight baby in 2004-06. Maternal health and ‘lifestyle’ are key contributors, suggesting that personal responsibility has an important part to play, instead of just expecting that government services can do it all. (For example, 50 per cent of Indigenous women report that they have smoked during pregnancy.) Nevertheless, government programs in some jurisdictions have reportedly had considerable success in getting more women to attend antenatal care, and to do so earlier in their pregnancies.

A need for such services for Indigenous mothers in particular is heightened by the much greater incidence of teenage pregnancy, which in itself is a significant risk factor. It has been suggested that the Baby Bonus has perversely contributed to this in some communities, although this is not evident in aggregated trend data.

Healthy lives

The old saying ‘prevention is better than cure’ would seem particularly apt for Indigenous health policy. For example:

• Indigenous people were several times more likely than others to die of avoidable causes in 2006, with only slight improvements since the 1990s.
• They were hospitalised for potentially preventable diseases at over six times the rate for non Indigenous people, and the rate is increasing. Complications from diabetes play a major part.

In this area, as in others, there are no easy answers. But is seems well established that more accessible and appropriate primary health care would get us a long way. It makes possible early detection, treatment and monitoring that can prevent a small problem from becoming a large and possibly fatal one. There are some great examples to build on. One of those with which I am personally familiar from the Indigenous Governance Awards, is Wuchopperen Health Service in Cairns. It demonstrates what a well-run Indigenous organisation, with caring and highly professional staff, can achieve in terms of better health awareness, higher treatment rates and improved health and social outcomes for the local community. I understand that the WHS model has been extended to other centres and there should be scope to build further on such success.

Learning for life

Education and training have been made a strategic area in their own right in this edition of the Report, so as to align the framework with COAG’s ‘building blocks’. This reflects the widely acknowledged importance of education to overcoming many aspects of disadvantage. Reflecting on this fact in his speech as ‘Australian of the Year’ for 2009, Mick Dodson said:

I’d like to see every Australian child next Australia Day geared up for the start of the 2010 school year…That every child deserves a good education, and that a country as prosperous as this one should be able to provide it, are things all of us agree on. We’ve been agreed on it for a very long time – and yet we still can’t do it.

Attendance at school is fundamental to better outcomes. For many years, it has been observed that rates of enrolment and attendance at school have been low for Indigenous children. But there have been no nationally consistent data available to verify or monitor this. So it is a step forward that this year’s Report has been able to include attendance data for the first time. These data confirm that attendance rates for Indigenous students are indeed lower than for other students, with the differences widening at high-level years. The differences, while significant — particularly in those jurisdictions with more remote communities — are not as great as some had anticipated. However, the data only relate to students who are enrolled; other data indicate that a larger proportion of Indigenous children are not even enrolled in school.

Lack of engagement of Indigenous young people with school is also apparent from the sharp decline in retention rates from Year 10 for cohorts of Indigenous students who had enrolled in Year 7/8. By Year 12, only 43 per cent of those students were still in school. The picture has not improved over time (and helps put in perspective the slight improvement in Year 12 completion rates).

A study conducted by DEWR in 2006 at Halls Creek, found that key factors in low school attendance were lack of parental pressure, bullying at school and perceptions about teacher performance. More generally, it is evident that achieving better educational outcomes for Indigenous students will require actions both at home and within Indigenous communities, as well as within the school environment itself.

The most important determinant of a school’s performance, as Chris Sarra has emphasised, is the calibre of its leadership and the quality of its teaching. It is therefore pleasing to see teacher quality receiving due recognition by COAG, and an indicator directed at this has been added to the Report’s framework. Unlike the challenges of improving students’ home environments, this is an area that governments should actually be able to address directly. But that means confronting the causes of a more general malaise in public education and the teaching workforce, which extends beyond the ambit of Indigenous education.

The most difficult schools and students within the public system should at least be receiving a fair share of the more able teachers. But this is obviously not the way the system currently works. How to transform it toward this end is a topic worthy of a speech in its own right and I will not presume to have the answers today.

However, I would like to draw attention to one part of the answer — the need to train and recruit more Indigenous teachers and teaching ‘assistants’. I first became aware of their potential during a trip to South East Gippsland (Victoria) when consulting on the first OID Report. On that trip, I met a young Koori woman who had turned her life around and was now doing the same for other Koori kids, as an assistant at the local high school. I asked her what she found the most challenging part of her job. Expecting her to tell me about the problems of dealing with difficult kids, she instead surprised me by saying that her biggest initial challenge was educating the other teachers at her school.

Indigenous staff in schools with significant numbers of Indigenous students provide a reassuring presence for those students, and they will often be better placed to offer support and guidance when it is most needed. But the teaching assistant in Gippsland convinced me that they can also help guide and support other teachers in what is a very challenging area to do well in.

Efforts to upgrade the teaching workforce in schools receiving Indigenous students are essential, but are unlikely on their own to close the education gaps. Commensurate efforts need to be made within Indigenous communities themselves to encourage families and children to value school. Programs that bridge school and home environments can play an important complementary role. A number of innovative examples of such programs have emerged in recent years and we have showcased them as ‘Things that work’ case studies in the OID Report. They include incentive-based programs such as the Clontarf Football Program in WA and the NT, and the ‘No School No Pool’ program in some remote communities. Such initiatives have generally been developed in cooperation with local communities. They have not only led to significantly improved school attendance and performance, but also better health and societal outcomes. Whether other programs that rely more on sticks than carrots can be as successful, such as the linking of welfare payments to school attendance, remains to be seen.

Home and community

The environment at home and within communities has major impacts on Indigenous peoples’ lives and prospects — providing protective ‘resilience’ that help Indigenous people cope with adversity, but also considerable strains and tensions. From the many indicators under these headings, I will touch on just two, both of which have pervasive effects.

The housing issue

The first, which has been much in the press recently, is housing. As is now well known, overcrowding can give rise to a variety of poor social and economic outcomes. The latest data available, from the 2006 census, reveals some progress compared to five years earlier, though Indigenous people were still five times as likely as other Australians to be living in an overcrowded dwelling. Unsurprisingly, the problem is more severe in remote regions, particularly in the Northern Territory, and often involves rented community-owned dwellings.

Indigenous housing has therefore rightly been elevated as a policy objective, with some $5.5 billion in spending to be devoted to it over the next 10 years. There has been criticism in recent days that the construction of new homes with this money has not been happening fast enough. If delay is caused by undue redtape, that criticism is warranted. But if not, I have some sympathy with the Minister’s observation that it is desirable to ‘hasten slowly’ in this area, given a need to avoid the mistakes of the past.

The unpleasant fact, not always publicly acknowledged, is that deficiencies in housing are partly due to a significant proportion of the existing housing stock having been rendered uninhabitable by the occupants. This can be at least partly attributed to such houses not being valued by those who live in them, and a lack of incentive or obligation to look after them. So if any delays in construction are a consequence of better consultation with communities about their housing needs and design preferences, or through trying to achieve a greater sense of ‘buy-in’ than has occurred in the past, then that should be accepted as part of the investment needed to get better outcomes in this crucial area.

Speaking of costs, it is obviously important from the perspectives of both Indigenous people and taxpayers that housing needs are met in a cost-effective manner. The high unit cost of Indigenous housing, especially in more remote areas, is astonishing and seems difficult to explain solely in terms of transport costs. I suspect that part of the high price tag is due to building standards that may not properly reflect the tradeoffs between benefits and costs. It may also reflect scarcity of skilled tradespeople and lack of an adequately competitive market. If so, these are further reasons why it may be wise to take the time to achieve greater value for money, with the goal of providing a larger number of houses from scarce public funds.

Alcohol and drug issues

The second issue in this strategic area relates to alcohol and drug use, which have been blighting many communities. Indigenous people are increasingly expressing a view that, regardless of deeper causes, the problems require strict and urgent remedies. Noel Pearson has been the most vocal Indigenous leader on this issue, and perhaps also the most confronting.

Substance abuse ….. is a learned behaviour and the factors necessary for the growth of substance abuse in communities are firstly, the availability of the drug, secondly money to acquire the drug, thirdly time to use the drug, fourthly the example of drug use in the immediate environment and lastly a permissive social ideology in relation to the use of the drug. (Pearson 2003)

The relative severity of these problems do not show up in the Report’s aggregate statistics about usage, no doubt for the same reason that gambling within the wider community is heavily under-reported in the Household Expenditure Survey. People do not like to admit to ‘bad’ behaviour. A more revealing indicator that does not rely on self-reporting is homicide data. This shows that alcohol was implicated in 70 per cent of cases involving Indigenous people, compared to 23 per cent for other homicides. Moreover, hospitalisations related directly to alcohol occur at much higher rates for Indigenous people across most relevant conditions (for example, they are seven times higher than for non-Indigenous people for disorders arising from ‘acute intoxication’).

A number of approaches to these problems have been followed in different communities and some are documented in this Report. They include the initiative to restrict sales of alcohol to low strength bear in Fitzroy Crossing (Western Australia), the Groote Eylandt Management Plan in the Northern Territory, and the Alcohol Management Plans in Cape York. The key differences among the various programs relate to the severity of the controls and the degree to which they are ‘owned’ or accepted by the local communities concerned. The most wide-ranging initiative currently is under the NTER, which involves the banning of grog except in prescribed canteens, together with an income management regime. This has some similarities to what is happening in Cape York, with the important difference that the four communities in the Cape York trial were engaged in its development. Monitoring and evaluating outcomes from these different approaches will be crucial to finding a sustainable way forward that gets the balance right.

Participating in the economy

In all societies, increased participation in the economy has been a key to reducing disadvantage, yielding both material and social benefits. It is therefore heartening that we have seen increases in Indigenous employment and income levels in recent years. But we are far from seeing the gap close with the rest of the community, especially when allowing for the contribution of CDEP to Indigenous employment.

There are various contributors to the employment gap, but a key one has been a welfare system that has undermined work incentives. Indigenous people are 2 to 5 times more reliant on different forms of welfare support than other Australians. In 2004-05, almost one half of Indigenous people of working age relied on government payments as their main source of income, compared to (an already high) 17 per cent for the rest of the community.

It is not generally known that CDEP came into being following concerns by Aboriginal elders themselves about the corrosive effects on their communities of ‘sit-down money’. It was a sound idea, especially in remote areas with limited job opportunities. But the program acquired a life of its own — growing in coverage beyond where it was most needed and becoming a dead-end for many of those on it. After some initial problems, the current reforms, which include provision for more training and mobility assistance, seem to be on the right track.

It is hard to overstate the challenges in this area, especially where people have been unemployed or underemployed for a long time. Training is an important part of the answer, but needs to be relevant and linked to real work prospects. Some of the more successful programs, such as the one initiated by Dick Estens at Moree (New South Wales), have involved partnerships with potential employers and effective systems to motivate and monitor the progress of the people concerned.

Self-employment and private enterprise initiatives are a major feature of the Australian economy, but Indigenous people appear to be greatly underrepresented. One contributor to this has been an inability to make effective private use of communal land. It is therefore welcome that initiatives are underway to address this.

Governance and leadership

Leadership is very important in Indigenous communities, as it is elsewhere in Australia. Where communities have initiated significant reforms, it has often been through the drive and standing of individual men and, increasingly, women. I have met many impressive leaders and potential leaders in the years since we started on this exercise. They include talented members of a younger generation, providing real hope for the future. But many Indigenous leaders face a Herculean task dealing with problems that have accumulated over decades, as well as the rivalries and divergent interests that are evident within many communities, the members of which may have quite disparate regional and ethnic origins.

Effective leadership — getting the important things done — can be greatly facilitated by good governance arrangements. As noted by Stephen Cornell in the first ‘Conversation’ in this series:

….investments in education or health or housing or dozens of other things are unlikely to pay off without a capable governance system in place that can translate plans into action, priorities into concrete strategies, commitments into behaviour…..Governance is not one of a number of silos standing out there in the Indigenous world. It is a foundation of effective action across the board.

Good governance is easier to define than to achieve in practice. And it is very difficult to capture in indicators. Drawing on international research and a program of work by Reconciliation Australia, the Report sets out ingredients that are integral to effective Indigenous governance and examines a number of examples that may have wider potential. In the last two reports, we have showcased case studies taken from shortlisted applicants for the Indigenous Governance Awards.

Involvement on the judging panel for those awards opened my eyes to how many well-run and innovative Indigenous organisations there are in this country — a good news story that rarely gets a run in media reporting of Indigenous affairs. Indeed, I stand by the observation that I made in an article coinciding with last year’s awards, that the best of these organisations outclass many mainstream organisations or enterprises in Australia. In addition to the standard requirements for good governance, they exhibit a profound connection with community that makes the best of them truly remarkable.

There nevertheless remains a pressing need to improve Indigenous governance arrangements, especially at the community level, and the IGA and OID exercises are pointing to models that could potentially be adopted or introduced more widely.

While good governance has been lacking in many Indigenous communities, it has also been lacking within government itself. This is partly a legacy of divided jurisdictional responsibilities, and partly due to ‘silo-based’ approaches to service delivery and policy development within individual administrations.

The result has been a staggering lack of coordination in service delivery, inadequate policy development and program evaluation, and a surfeit of redtape — all of which have contributed to poor outcomes and a lack of capacity to take corrective action when things go wrong. COAG’s commitment to forge new, nationally coordinated approaches is therefore of major significance. These will presumably have learnt from initial misadventures with the earlier COAG trials, most of which fell considerably short of their objectives. One lesson is the importance of collecting baseline information and ongoing performance reporting for evaluation.

Key recent initiatives by COAG include the ‘National Framework of Principles for Government Service Deliver to Indigenous Australians’ (2004); the new reporting of national expenditure on services to Indigenous people (allowing us for the first time to discover how much is being spent and where) and, most recently, the ‘National Partnership on Remote Service Delivery’ (2009). The last of these targets 26 Indigenous communities in remote regions, adopting processes that integrate service planning and delivery across governments, with a single government ‘interface’ for each community. This has been complemented by the appointment of a new ‘Coordinator General for Remote Indigenous Services’, responsible for implementating major reform initiatives in housing, infrastructure and employment programs. While promising, the effectiveness of these new arrangements will only be revealed in time. If successful, hopefully they can be drawn on more widely.

Concluding remarks

At the broadest level, there are two sorts of challenges for government in overcoming Indigenous disadvantage. The first and most obvious challenge is developing and implementing effective policies and programs. The second and perhaps more fundamental one, is to generate the information needed to assess which policies are working and how much progress has been achieved. The two challenges are clearly closely related: good policy development depends on good information.

It is very important therefore that this has now been formally recognised by all governments, with COAG establishing a reporting framework to generate policy-relevant information at three levels: high level monitoring of programs, expenditures linked to outcomes and program effectiveness.

The OID report has an important role at the top level, not only in monitoring outcomes for Indigenous people, but also in its focus on areas where policy needs to make most difference. To fully meet its potential in these respects, the Report must have access to robust data. As the Prime Minister observed in the revised terms of reference for the exercise, ‘without high quality data, it is impossible to understand where we are headed’.

Despite considerable advances over the past 7 years, we still do not have trend data for about half of the 50 indicators in the report. Moreover, the data for many indicators do not support reliable comparisons across jurisdictions or geographic areas. Better data on outcomes for Indigenous people in remote regions relative to urban centres is particularly important, as Helen Hughes has emphasised. The OID Report includes it wherever possible, but there is as yet little data being collected to support the sort of breakdowns she favours.

Of particular note are deficiencies in hospitalisation and deaths data, where the extent of Indigenous identification has been improving but remains inadequate for reliable comparisons across geographic areas. Rising levels of Indigenous identification over time — a positive development in itself — have made it particularly difficult to interpret some time series.

There are also deficiencies in availability of data specifically for Torrens Strait Islander people. Given their relatively small numbers, they tend to get swamped statistically in administrative collections. Nevertheless, it would be useful to have more detailed reporting for Torres Strait Islanders at least in those parts of Queensland where there are higher population densities.

While outcomes data in the OID Report provide important information, for policy-making purposes there is no substitute for detailed evidence-based assessments and reviews of specific programs. One advantage of our Federation is that it has generated many different policy and program innovations. However, with some exceptions, Australia has squandered the opportunity to learn systematically from these diverse experiences in order to indentify those that could make a difference if applied nationally. In a small way, the OID Report has sought to redress this, by including mini case studies of ‘things that work’ (or appear to be working) in areas targeted by the framework often at the level of particular communities or regions.

The report identifies four factors that are common to many of the ‘things that work’:

• Cooperative approaches between Indigenous people and government, often involving non-profit and private sectors as well. (The Cape York Welfare Trial is illustrative of the power of this.)
• Community involvement in program design and decision-making — a ‘bottom-up’ contribution, rather than just relying on ‘top-down’ direction. (There are many instances of governments designing programs that have resulted in unintended perverse consequences through lack of community input.)
• Ongoing government support — human, financial and physical. We have often seen, even between editions of the OID Report, promising programs that have initially been very successful lose momentum for want of sustained government support.
• Good governance — as noted earlier this cannot be taken for granted, but must be nurtured and supported. It is needed in both Indigenous communities and organisations, and within government itself.

In an important development, COAG has agreed to establish a national clearing house on best practice and success factors. It is designed to provide a central repository of rigorously assessed evidence, giving governments and the wider community ready access to quality information about ‘what works’. It is also intended to improve the coordination of research and identify future research and evaluation priorities.

It if lives up to this, the clearinghouse could be a very powerful tool for policy learning. However, it will be useful only to the extent that governments really do get serious about generating the data needed to evaluate their programs.

Finally, I am sometimes asked by those who may confuse the Productivity Commission’s normal policy advisory function with its secretariat role in this COAG exercise, why the Report does not include specific policy findings or recommendations? The answer is that it is not that kind of report — it serves a different function — and its coverage is not the Commission’s call anyway. That said, if I can put my Productivity Commission hat back on to conclude, I believe that the Commission could assist governments in this important and sensitive policy area, as we have done in other important and sensitive policy areas (like paid parental leave, gambling, public housing, the not-for-profit sector….). Areas that could benefit from the sort of rigorous independent and consultative reporting that we are set up to provide include employment programs, housing policies and programs — indeed public evaluation of key programs generally — as well as perhaps extending our reviews of regulatory burdens on business to include the redtape that is diverting scarce human resources from more urgent and important work in Indigenous organisations.