Five Ideas for the War Cabinet

The ‘war cabinet’ to address Indigenous disadvantage sounds gimmicky on first blush, but may actually play a useful role if it allows the federal government to rely more on the evidence and less on interest groups and the median voter. In my view, the most promising reforms are all around education, so if I had the chance to make a pitch to the cabinet, here are five ideas I’d propose.* 

  1. School attendance rates are appalling, and as Woody Allen says “90% of life is just showing up”. So pay Indigenous children to attend school.
  2. Literacy and numeracy gaps are large, and part of the difference may be teacher quality. So the federal government should promise bonuses of up to $50,000 to teachers who can get large improvements in performance in Indigenous schools. Teaching disadvantaged kids is the most important job in Australia – so why does no-one doing it earn a six-figure salary?
  3. Indigenous people are overrepresented in Australia’s jails, which do little more than warehousing. Since many are now private, why not rewrite the contracts, making payment conditional on post-release recidivism and earnings? Let’s create incentives for those who run jails to do more education, and less clock-watching.
  4. A major impediment to children attending school is drunkenness in communities. But a ban is a drastic measure. Let’s allow communities to set their own tax rates on alcohol, and keep the revenue (remember, a ban is effectively a tax rate equivalent to the cost of petrol to the nearest no-ban town).
  5. As many Indigenous policies as possible (including those above) should be subjected to rigorous randomised trials. Those that fail should be discarded without sentiment, and those that succeed should be expanded. We know from the headline indicators that many Indigenous policies haven’t worked; it’s time to start sorting out the wheat from the chaff.

* Anyone who has read Imagining Australia, a book I coauthored in 2004 with Macgregor Duncan, David Madden, and Peter Tynan, will recognise numbers 3, 4, and 5.

This entry was posted in Indigenous Policy. Bookmark the permalink.

23 Responses to Five Ideas for the War Cabinet

  1. Some interesting points here, but would the first suggestion equate to a simple cash transfer, or could it might it perhaps be put in mandatory savings accounts that can only be accessed as a certain age is reached? Should it supplement the Abstudy program or replace it? I’m always suspicious of the potential downsides of straightforward cash transfers.

    Has any research been done regarding the negative/positive effect of children attending who otherwise might on those who would attend anyway? That is a fear that I have regarding any mechanism that ‘encourages’ potential absentees into attendance.

    Further, has your own research broken down the outcomes for those students ‘compelled’ to attend school an additional year who otherwise would have left against those who would have stayed anyway? It’s all very interesting stuff, and I hope the discussion continues.

  2. Pingback: Oh please, oh please, oh please at John Barrdear

  3. Kris,

    Andrew’s first suggestion doesn’t need to be (indeed, probably shouldn’t be) a simple cash transfer. It can (and should) be conditional on proof of attendance at school.

    Here is an article from the latest edition of The Economist speaking on the success of a similarly motivated scheme in Brazil and the gradual adoption of the idea around the world: Happy Families (7 Feb 2008).

  4. Thanks for that link. Really interesting stuff!

  5. Michael W says:

    Hear hear! – on the randomised trials in particular. They never seem to occur in social policy. I think one of the big reasons for failure in social policy (as opposed to, say, public health policy) over the last 100 years is the absence of a scientific outlook.
    Kevin Rudd said it would be “hard, very hard”. So much harder if we just flounder around in the dark with no knowledge and no learning about which measures really work.
    There are of course ethical issues about control groups, or placebo treatment groups, who do not gain the benefit of the successful intervention. But they gain no benefit either if they experience an unsuccessful program without knowing it. Trials can be stopped and treatments revised if evidence emerges sufficiently strongly to show one treatment is better.

  6. I doubt (3) will work. Given that the poor record of adult education and training on poorly-schooled individuals I doubt that much can realistically be done on a population that is likely to have an even weaker prospects than the average for this group. It also encourage ex-inmates to seek revenge on their jailers.

    Given the atrocious levels of violence against Indigenous women, jails are doing service enough in keeping the perpertrators in places where they can do no harm to others.

  7. Spiros says:

    “places where they can do no harm to others.”

    Apart from fellow inmates (bash, mutilate, rape, murder).

    Re 1: You also want them to pay attention when they get there. They should be paid for completing school years (which requires showing up), not just showing up.

    Re 4: It’s hard to imagine a system more open to abuse and corruption.

  8. Steve W says:

    Re. 1: I doubt it is possible to make meaningful inroads (at least in remote communities where the attendance is lowest) without tackling public health. I saw a study a few years ago which claimed a prevalence for serious ear infections resulting in hearing impairment in over 50 per cent of children in remote indigenous communities, with this infection rate being largely driven by inadequate housing. If children can’t hear the teacher they are much less likely to attend school, and much less likely to learn if they do so. Thus moving to a situation where each nuclear family has a house, and where such housing has been designed such that it can be maintained locally would seem to be a necessary (but not sufficient) condition for improving educational attainment.

    Re 5: I’m a big supporter of proper evaluation of government programmes, but most of the ideas listed above aren’t really amenable to randomised assignment which only really works at the individual level (except perhaps recidivism programmes), so you would need to go for the second best solution of a matched comparison approach.

  9. alao says:

    I really like your five ideas Andrew. I hope you can present them to the 2020 conference.

  10. Spiros – OK, do less harm to others. As prisons are still single-sex and adult, their most vulnerable victims will be protected.

  11. Pepper says:

    I would have an additional suggestion of the magic bullet type: ready access to high quality internet at remote communities with a view to seducing the children to a wider world of possibilities.

  12. JM says:

    #4 is pretty poor analysis:

    “A major impediment to children attending school is drunkenness in communities. ”

    Agree, but lack of schools is more “major”. Provide the schools.

    “But a ban is a drastic measure. ”

    Strawman argument alert. Such a ban is already in place in many of the remote communities and has been for many years. One of the major complaints made against the “intervention” of last year was the effective removal of these bans.

    “Let’s allow communities to set their own tax rates on alcohol, and keep the revenue (remember, a ban is effectively a tax rate equivalent to the cost of petrol to the nearest no-ban town). ”

    Too narrow, and contradicts your first point. If local drunkeness “impedes” school attendence then local sobriety removes the impediment. Getting alcohol from a remote location, bringing it back and disrupting local life has a cost beyond that of the petrol for the journey.

    That’s why the communities implemented the ban in the first place.

    The next problem is that a tax that returned the full cost to the community would be overturned by democratic processes within the community demanding its reduction. The drinkers would simply negotiate the tax rate down. I’m quite sure that the “marginal benefit of the extra vote” would always outweigh the marginal damage caused by the extra drinker.

    After all, that’s the way it works in white society where prohibition has proved such a disaster in the past.

  13. Patrick says:

    I agree with commenters – number 4 is hard to believe in. Maybe it should be trialled on a randomised basis?

    All the others are fantastic ideas!

  14. Peter says:

    To be applied with any validity, randomized trials in any sociological domain would have to take acount of micro-cultural, micro-geographical and micro-historical differences, such as the specific history of black-white interactions in a particular area over the relevant time period (3 or 4 generations beforehand), since such causal factors without any doubt at all impact social policy outcomes. To allow for such causal factors in a trial design, the overall sample size required would then need to be extremely large — perhaps larger than the total aboriginal population — in order to achieve statistically-meaningful sub-sample sizes.

  15. Bruce Bradbury says:

    Peter, I don’t think social complexity per se is an obstacle to randomised design. After all, our understanding of human biology is still very incomplete, but can make some progress with randomised drug trials. (Similarly the ethical issues are often more acute in medical rather than social experiments).

    In addition to random assignment, the key requirements are that the intervention to be trialled has to be reasonably simple (we need to be able to clearly state that one group has, and another hasn’t, received the intervention) and it has to be an intervention which has the potential to have a sizeable impact (otherwise any effect will be swamped by the ‘noise’ introduced by the types of factors that you talk about).

    So, while the micro variations that you talk about are important to bear in mind, if the intervention is large enough then the evaluation will not need an overly large sample.

  16. Sinclair Davidson says:

    I think #4 constitutes an excise tax which is unconstitutional at the state level. Allowing the community to keep the money may be unconstitutional at the Federal level.

  17. Bob Durnan says:

    Andrew may be correct in his assertion that “the most promising reforms are all around education”, and Woody is undoubtedly onto a key insight into the nature of the universe. However, unless adequate policing, other safety measures, drug and alcohol reduction and violence reduction occur at the same time, then it is a forlorn hope that education initiatives will assist much anybody other than a few isolated individuals.

    School attendance rates are generally appalling, but it is interesting that attendance has already improved considerably in a number of the communities affected by the full measures of Brough’s NT Emergency Intervention. This is after the introduction of welfare quarantining (aka ‘Income Management’, or ‘IM’) under which 50% of welfare income is protected from being spent on grog, drugs & gambling.

    I believe that when the Emergency Taskforce introduces the next tranche of these reforms, which will actually identify those responsible for the care of the kids not attending school regularly &, where there is no good reason for the non-attendance, increase the level of IM, thus further restricting the cash available for grog, drugs gambling etc (this will purportedly happen quite soon in the first communities that went on IM last year), then we may well see quite high levels of school attendance, without having to introduce a special benefit for Indigenous kids that would quickly feed another insurrection by the Hansonites and their dog-whistling suitors on the coalition benches & elsewhere.

    Andrew should be acknowledging the work of Noel Pearson & Marcia Langton in relation to his suggestion of “$50,000 to teachers who can get large improvements in performance in Indigenous schools”. But for all I know he may be one of their advisors.

    His point about gaols is well made, and was pursued by Peter Toyne when he was NT A-G, but Peter could not obtain the support of his NT Cabinet colleagues, who notoriously preferred to put their excess cash into upgrading the Darwin collosseums and boulevardes for petrol heads and turf warriors rather than doing the electorally un-sexy work on educating & training Indigenous prisoners of high-school & tertiary education ages while they are still healthy and not yet damaged by many years of addictions. It is true that the lack of effective pre- & primary schooling experienced by most inmates means this will not be easy, but it is one of the few relatively productive avenues available for effectively improving the education & training of those who have missed out on proper educations. It may well be relatively cost-efficient, compared to most of the alternative solutions, but this could be measured by random audits.

    A number of the comments have identified that the weakest part of Andrew’s proposals is his idea that “communities [could] set their own tax rates on alcohol, and keep the revenue”. This is basically what Bjelke-P., Killoran and co did when they introduced ‘wet canteens’ onto some Qld reserves in the late eighties, and it has been identified as the key factor in the corruption and deterioration of those communities. Once you do this you introduce a classic conflict of interest into the situation: to improve their chances of ever escaping poverty, the power brokers in the community need to maximise sales of alcohol to pay for the badly needed services, jobs, sports facilities etc. Then you quickly manufacture addicts out of the bored welfare recipients & lowly paid poorly skilled workers. Then you have more trouble delivering education, training & work programs into the violence ridden communities. And so it goes on & around. Don’t ever suggest that one again please Andrew. It would make much more sense to have any alcohol outlets owned & mmanaged by a sparate body, governed by sober people, with any profits going into improving services to prevent and treat substance abuse.

    As for subjecting Indigenous policies to rigorous randomised trials, wherever this is feasible and sensible, just do it. It’s basically in line with what Pearson, Dodson, Langton, Yu, Pearce et al called for in their Nth Qld summit meeting a couple of years back.

  18. ChrisPer says:

    Knoiw how to raise attendance?

    Get the welfare to remove the kids if they are truants.

    Thats why their grandparents are literate, and this generation are not.

  19. Justin says:

    Do you ever get tired of dreaming up ideas for engineering other people’s lives?

    Here’s a good social engineering idea: instead of your childish expectation that government will make the money for your schemes appear with a magic wand, why don’t you and everyone else who agrees with your schemes raise the money voluntarily?

    Don’t tell me, lemme guess: because you own everyone else as well as the Aborigines?

  20. Patrick says:

    Sinclair, I’m pretty sure that there would not be much Constitutional problem with N 4, since the Territory doesn’t really have s90/92 issues about excises and other discrimination between states, nor does the Cth. Also, it wouldn’t discriminate because it would tax the alcohol of all States equally (albeit differently in each area). So even a State could do it.

  21. Sinclair Davidson says:

    Maybe. I’m thinking of these two clauses from the Commonwealth Constitution.

    s81. All revenues or moneys raised or received by the Executive Government of the Commonwealth shall form one Consolidated Revenue Fund, to be appropriated for the purposes of the Commonwealth in the manner and subject to the charges and liabilities imposed by this Constitution.

    s86. On the establishment of the Commonwealth, the collection and control of duties of customs and of excise, and the control of the payment of bounties, shall pass to the Executive Government of the Commonwealth.

  22. Patrick says:

    raised or received by the Executive Government of the Commonwealth I don’t think this is so broad. It would be raised and received by the community. If there was a problem it would suffice to call it a fee or rent.

    shall pass to the Executive Government of the Commonwealth who is essentially omnicompetent as to which of its powers it may delegate.

    Also as above I don’t believe that this could be an excise because it is not related to production nor customs because it is not, except tangentially, related to State boundaries.

  23. Pingback: Club Troppo » Missing Link Daily

Comments are closed.