Wednesday, 8 November 2017

The next phase of funding for remote Indigenous housing


You take my house when you do take the prop
That doth sustain my house. You take my life
When you do take the means whereby I live.
Merchant of Venice Act 4, scene 1


It is widely accepted common ground that housing is an important element in addressing deep-seated disadvantage.  See for example the AHURI research report released with the recent Remote Housing Review (link here). Australia’ remote Indigenous communities are the locus of perhaps the most concentrated deep-seated disadvantage in the nation.

Since the Second World War, the Commonwealth has played a crucial role in funding social housing nationally, initially through the Commonwealth State Housing Agreement (CSHA) as well as through the development of the Commonwealth Rent Assistance Program. Following the 1967 referendum, the Commonwealth gradually ramped up its own direct investment in Indigenous housing programs both in the CSHA, and in a number of Indigenous specific programs. Up until 2007, the CSHA included an element directed towards Indigenous social housing which was delivered by the states and territories, while the Commonwealth delivered an Indigenous specific housing and infrastructure program with a nation-wide footprint called CHIP (Community Housing and Infrastructure Program).

These Indigenous specific programs funded by the Commonwealth were in specific recognition of the political reality that the states had not been prepared to provide services to their Indigenous citizens, often adopting discriminatory policies of exclusion. While the overt racism of the states’ policies has largely disappeared, the institutional and structural invisibility of Indigenous interests continues to the present day. The discussion below of ‘acceptable’ levels of overcrowding is just one example. The case for ongoing Commonwealth engagement in remote housing policy remains incontrovertible.

A review of CHIP commissioned by Minister Mal Brough in 2007, Living in the Sunburnt Country, found that the program was not focussed on need and recommended greater focus on remote housing since regional and urban Indigenous citizens had access to mainstream social housing options which were not available in remote regions. For example, the virtual total absence of private rental housing options in remote communities meant that Indigenous residents had a structural incapacity to access Commonwealth Rent Assistance, the largest social housing program in the nation.

In its last year in office, the Howard Government made a number of moves to upgrade the funding for housing in remote regions, through a program known as ARIA and later renamed the Strategic Indigenous Housing and Infrastructure Program (SIHIP).

Following the election of the Rudd Labor Government, in early 2008, the Commonwealth and the States agreed to establish a National Partnership Agreement on Remote Indigenous Housing, known as NPARIH. This was part of a broader reform of Commonwealth – State financial relations. The agreement in effect combined the former CSHA component with the Commonwealth CHIP program and provided for Commonwealth funding of $5.5bn over ten years, a massive increase in previous funding levels for remote housing, with housing construction and tenancy management to be undertaken by the states and the NT. It set specific targets for each jurisdiction related to new builds, refurbishments, and Indigenous employment. Even so, it was recognised that it would not address the totality of outstanding housing needs in remote Australia.

The ten year agreement survived the change of Government in 2013 albeit with a change of name – it is now known as the National Partnership on Remote Housing, or sometimes as the Remote Housing Strategy.

The National Partnership expires in June 2018, and thus raises the question, what will replace what has been the major national investment program in remote Australia (apart from welfare and employment programs) over the last ten years? While I dealt with this topic in a previous post (link here), this post is an update following the publication of the Remote Housing Review.

To pave the way for the Government decisions which will determine the answer to that question, in November 2016, the Minister for Indigenous Affairs, Senator Scullion, announced a review into remote Indigenous housing. In his media release announcing the review, the Minister stated that ‘Overcrowding, homelessness and poor housing conditions in remote Australia remain unacceptably high’. According to the Department of Prime Minister and Cabinet (PMC) website, the Review had two tasks: an analysis of what was implemented in the past and an assessment of what should be effected in the future.

Notwithstanding that the Review suffers from a range of serious analytical flaws which I outlined in my previous post (link here), the review’s findings essentially confirm the Minister’s assessment of the unacceptable state of remote housing.

On 10-11 May 2017, the Prime Minister’s Indigenous Advisory Council was briefed on the review. In their communique dated 19 May, the Council made a number of comments on Indigenous housing and the then unreleased review of remote housing.  They noted, inter alia:
Council acknowledged the findings of the Remote Housing Review and expressed concern that despite significant reductions in overcrowding investment is required to meet unmet need and maintenance of housing stock. Council emphasised adequate housing is critical to ensure positive outcomes are maintained in health, education, employment and community safety...
A link to the Advisory Council’s communique is here

Since May there have been a number of development which relate to the next phase of funding. In August, the Minister made a number of comments in an ABC Q&A program at Garma (link here), where in response to a question about overcrowding and the impact on peoples’ lives, he foreshadowed a number of likely policy directions:
We’re deadly serious about this. We’ve invested $5.4 billion over the last decade, and I think everybody would agree we could have done a lot better. We have reduced overcrowding from 52% to 37% – it’s still in the margins, and that took a fair while to do. So, the next rollout, which we are now negotiating with the states and territories about the National Partnership on Remote Indigenous Housing, we’ll be negotiating on the basis of what the communities have asked us to negotiate on.

So, Indigenous employment is non-negotiable. Indigenous procurement is non-negotiable. And we’ll be asking the states to match those funds. Because we need a pulse. Sometimes we can just trickle along and we’ll be just catching up, just getting ahead, but we actually need a significant injection of funds. So, that’ll be the basis of our negotiation with the states. But those houses cannot be built by whitefellas getting off planes with nail bags. Those times have to go.

Prior to the most recent Estimates Hearings relating to the Indigenous Affairs portfolio, the Minister released the Remote Housing Review Report on 26 October 2017 (link here). In the lead up to that release, a government source provided The Australian’s Darwin correspondent Amos Aikman with not just the report, but the earlier May version which had been provided to the states and territories for their comment and input. That source was presumably the Minister or his Office; who else had access to both the earlier draft and the about to be released final version? While it is normal for Ministers to provide the media with advance copies of reports they are releasing, the provision in this case of both versions was clearly aimed at highlighting and undercutting the changes the review panel had made after consultation with the states, and ‘bells the cat’ on the lengths to which the Government has gone to frame the public discussion in the lead up to its decision on future funding for remote housing.

The ‘drop’ to The Australian would likely have involved at least some level of briefing as Aikman’s 26 October story (Blot on $5.4bn indigenous housing) reported on the key changes between the two versions. This related to the extent to which jurisdictions allocated NPARIH funds for non-construction activities such as land servicing, infrastructure provision, and program administration.

The final data is presented in Table 4.2 of the report. While WA spent 33.6 percent of its allocated funding from 2008-2016 on ancillary costs (the source of the 34% figure), other jurisdictions spent less. The average across the four jurisdictions remaining in the scheme was 20.8 percent. While the inevitable inference of the media story was that ancillary costs were somehow illegitimate, the report does not say this is the case, and indeed explicitly states (section 7.1.3) that there was no suggestion these costs were illegitimate, merely that the Commonwealth did not have a line of sight to them. The reality is that houses without access to power, water, sewerage, and access roads are not ‘fit for purpose’. Much of the debate around remote housing over the last decade has overlooked this crucial fact, and instead been simplistically focussed on three key metrics: numbers of new houses, refurbishments, and levels of Indigenous employment. Important as these are, the remote housing system is more than the sum of these three metrics.

Aikman’s lead paragraph also managed to pick up on a subsidiary observation by the Review Panel (‘the scheme is opaque, “complicated by multiple objectives, poor governance and constantly changing policy settings’ and has not fostered home ownership or business growth”) rather than the Review’s headline finding, namely that:
By 2018, the Strategy will have delivered over 11500 more liveable homes in remote Australia (around 4000 new houses and 7500 refurbishments).
This increase in supply is estimated to have led to a significant decrease in the proportion of overcrowded households in remote and very remote areas, falling from 52.1 percent in 2008 to 41.3 percent in 2014-15. The Panel projects this will fall further to 37.4 percent by 2018. (Executive Summary page 1).

In particular, Aikman noted that ‘At an average price of $600,000 per house, estimated from the existing program, the next ten year agreement is likely to cost around $3bn. The report recommended that federal and state governments split the cost equally, but the NT has already demanded more’.  Given the likelihood that The Australian received a detailed briefing on the report (the $600,000 figure appears to be an extrapolation of data provided in Tables 4.2 and 4.4 of the Review, and unlikely to be a figure the journalist would come up with unilaterally), there seems to be a possibility that the $3bn figure may reflect one of the options in the Minister’s bid to Cabinet for a national allocation. Of course, such a notional calculation is fundamentally flawed, as it ignores the potential need for refurbishments as well as new houses, and also the increasing need for Property and Tenancy Management costs (which include ongoing small repairs and maintenance costs)

In a follow up story on 27 October 2017 (Housing loses out as cash diverted), Aikman’s lead paragraph reported that:
States and territories are spending up to 34 percent of federal funds provided for remote indigenous housing on “ancillary costs”, but those governments also blame Canberra for poor living conditions in the bush, arguing it has not committed to ongoing payments.

The story went on to report that the 34 percent figure highlighted in both media stories related just to one jurisdiction. Nevertheless, the headlines and lead paragraphs in the two articles laid out a narrative which painted the states and the NT as mendicants not pulling their weight.

At the most recent Estimates Committee hearings (link here), there was a rather bizarre discussion around the next phase of Commonwealth involvement in funding remote housing. The Minister and his officials made no absolute commitments, but appeared to lay down a few benchmarks which will shape future Commonwealth policy.

First, the Minister and his officials argued (against accepted wisdom and the history of Commonwealth involvement since at least 1968) that funding for remote housing was primarily a responsibility of the states and territory, and that the Commonwealth’s role was in effect optional or discretionary. As Minister Scullion stated:
Yes we have seen this review and we’ve let the states know about the review. They are aware that this is a national partnership agreement under which we haven’t reached the goals we were supposed to reach, because then it was to go back to the states’ responsibility. This is overcrowding in remote areas. So negotiation still has to happen about in which states this is mostly occurring in. And it’s a matter for the Commonwealth to negotiate with the states about what percentage of the responsibility remains with the Commonwealth. It was the intention that by this stage, the Commonwealth would have no further role and the role would go back, rightly, to the states and territories in this regard. (Estimates transcript 27 October 2017:51).

This statement is factually wrong on a number of levels. There was never an ‘intention’ that NPARIH would be the end of the Commonwealth’s involvement; indeed, it was recognised at the time NPARIH was first negotiated that the targets embedded within it would meet only around half of the outstanding national need. Second, NPARIH largely met its goals, and allowing for substitution of investment between new builds and refurbishments, in fact exceeded its goals. (Refer section 4.3 of the Review report).

One way to interpret Minister Scullion’s statement is that the Minister has yet to obtain a Cabinet decision on the quantum of funding available for the renewal, but recognises that Cabinet is unlikely to allocate a further $5.5bn over ten years. Alternatively, the Minister may have been told that he won’t get a further $5.5bn, and thus he has engineered a review and a set of arguments designed to throw dust in the air to avoid criticism for his lack of capacity to deliver an ongoing program at former levels for Indigenous constituents.

Second, the Minister appeared to rule out the use of a National Partnership going forward, indicating instead that the Government was considering a series of bilateral agreements with each of the four remote jurisdictions (WA, SA, NT and Qld) rather than an overarching National Partnership. The Minister has not laid out a persuasive argument for making this shift, but what is clear is that it will allow him to set up an effective bidding war between the jurisdictions (who now happen to have Labor Governments) based on their preparedness to match Commonwealth funding. A series of bilateral agreements also opens up the possibility of a shift away from a ten year program to shorter bilateral agreements. A further risk in moving away from the National Partnership model is that it increases the scope and likelihood of reductions in investment levels through the ongoing annual budget process since COAG endorsed National Partnerships appear to have a greater (but not total) degree of quarantining from budget revisions during the life of the agreements.

Third, the Minister was quite vague on the timing for an announcement of new funding arrangements. In responses to questions in Estimates previously, the Minister had advised that a decision would be announced as part of the Mid-Year Economic and Financial Outlook (MYEFO) statement, normally brought down around December or January. However in the latest Estimates hearing, he shifted his ground, indicating that discussions with the states and territory were at a very preliminary stage, and that ‘it’s unlikely any announcement will be made prior to MYEFO, but there will certainly be an announcement prior to or during the budget process’. This suggests that the MYEFO budget process will likely be used to decide on the national funding allocation quantum, but that there will then be a subsequent negotiation or ‘bidding process’ undertaken with the four jurisdictions.

Ironically, one of the important and sensible recommendations of the Review Panel was that ‘a minimum five year rolling plan for the program should be established’ (rec.7, page 4). Yet the Commonwealth has allowed a situation to develop where the states and the NT will essentially enter the 2018-19 financial year with minimal lead time to develop a capital works program, thus virtually guaranteeing at least a one year hiatus in capital works and potentially PTM momentum. The review was critical of the National Partnership’s two year capital works cycles, yet the Commonwealth has done nothing to ensure a smooth transition between the current Remote Housing Strategy and the next phase.

So what then should the next iteration of the program look like? As with any policy issue, the answer must take into account a number of issues. In this case the core first order issues are the levels of need across remote Australia, measured by overcrowding; and the necessary investment required to maintain the asset base and maximise asset (ie house) lifespans. These first order issues go the quantum required to address the outstanding housing deficit in remote Australia.

Other issues, such as maximising local indigenous employment and/or Indigenous employment generally; deciding how to best deliver necessary repairs and maintenance; choosing and resourcing the most effective tenancy support models; design/cost trade-offs; new builds versus refurbishment trade-offs; and construction cost versus whole of asset lifespan cost trade-offs; are all important, but second order issues. The second order decisions will be crucial determinants of the effectiveness and efficiency of the resources allocated. Merely listing them provides a sense of the complex decision-making processes embedded within this program, and the overall complexity of the remote housing system. This is further complicated by the fact that two levels of government are involved, and that choices in relation to any one of the second order issues will inevitably impact on the options and choices available on the other second order issues.

Turning to the first order issues, the Review is only of limited assistance. The Panel’s assessment of the outstanding ten year need of 5500 houses is extremely problematic. The Panel itself admits that the estimate relates only to households requiring three bedrooms or more, or to citizens who are homeless. As they note in the text under Figure 3.3, this suggests ‘the overcrowding challenge is likely to be greater’.

Secondly, they note that the estimates are based on modelling of 2011 Census data and as they state in footnote 29: ‘remote housing need requires remodelling when the 2016 Census data is available’.

Third, the Panel fails to include any estimate for houses which reach the end of their effective lifespan. Data on the effective lifespan of the current asset base nationally is virtually non-existent and disappointingly, the recent Remote Housing Review provides no insights into the issue. But if average asset lifespans were 30 years, and the national remote Indigenous community housing asset base was 15,000 houses[1], then we could expect around 500 houses to reach effective ‘end of life’ each year, or 5000 over the next decade.

Fourth, the Panel’s modelling is based on reducing overcrowding from 37 percent to ‘around 25-30 percent by 2028. This level of overcrowding is still 10 percentage points higher than the rate in urban and regional Australia.’ The Panel makes no assessment of the level of outstanding need based on parity in overcrowding levels with metropolitan Australia[2], and indeed, asserts in the Executive Summary of the report that the targets specified above involve reductions in ‘levels of overcrowding to acceptable levels’ without any discussion of what makes them ‘acceptable’ (emphasis added).

Fifth, the Panel ignores entirely the figures on current demand and projected need provided to it by three jurisdictions, which total 7520 houses without including projected need in South Australia and Western Australia (see Box 3.2).

Based on the factors identified above, a more realistic assessment of outstanding need for housing across remote Australia over the coming decade to 2028 would be at least 10,000 new houses and arguably 15000 new houses.[3] Using the figure of $600,000 per house, the cost of delivering 10,000 houses would be $6bn. This would need to be complemented by a program of refurbishments aimed at extending asset lifespans wherever possible, and assuming effective PTM policies are in place, this might involve somewhere in the region of 2 to 5 thousand refurbishments over the next decade. Add say 3000 refurbs at $100k each would cost $300m or $0.3bn.

Finally, the Review Panel correctly emphasised the need for a complementary recurrent program to be funded designed to deliver PTM to the entire asset base. The review’s most insightful conclusion, based on detailed research by NOUS Consulting (link here), was that annual rental revenues covered only between 10 and 20 percent of the actual cost of maintaining the asset base. Much of this extra cost is a function of remoteness. See section 8.1.1 of the Review report also. This suggests that some $2.8bn in additional funds will be required to maintain the existing asset base over the next decade.[4]

These admittedly rough ‘ball park’ calculations suggest that meeting the outstanding housing need in remote Australia over the decade to 2028 would cost around $9bn; $6bn for ten thousand additional houses; $0.3bn for refurbishments; and $2.8bn for PTM. While this quantum is daunting in itself, the challenge is magnified by the fact that it does not include the concomitant ancillary costs of land servicing, essential services infrastructure, and access roads.

The prospects that the Commonwealth will allocate anywhere near $9bn to fund remote housing provision over the next decade appears negligible. The level of political and electoral pressure on the Commonwealth is minimal. The daily challenges of Indigenous interests in remote Australia are largely invisible in national political discourse, which has focussed in recent years on constitutional recognition. Moreover, Indigenous interests are spread thin and exercise very limited political influence. Increasingly, the Indigenous community vote spans the political spectrum, which often mean that Indigenous voices cancel each other out. Advocacy is almost non-existent. Nationally, the National Congress of First Peoples does not currently list housing on its list of priority policy issues (link here). In the NT, the Aboriginal Peak Organisations Northern Territory (APONT) has been building its advocacy capability since 2010, and in 2015 facilitated the establishment of Aboriginal Housing NT, a body dedicated to advocating on housing issues for Aboriginal citizens in the NT (link here). AHNT made a submission to the Remote Housing Review (link here) but has not to date built a public profile.

The consequences of underinvesting in remote housing over the coming decade will be devastating on a number of levels. For taxpayers, the value of the investments of the past decade under NPARIH, along with whatever reduced amount is allocated, will in effect be substantially and prematurely depreciated as housing asset lifespans are shortened. The economic impost on taxpayers of meeting the costs of poor health, education and social outcomes will expand. The local economic multipliers which spin off capital works investments will not arise, thus further undermining the prospects for economic development in remote communities. And most importantly, the life opportunities of tens of thousands remote citizens will be irretrievably constrained. The evidence emerging form the latest census is that income levels are worsening in remote Australia, and a pull back in remote housing investment will only exacerbate this trend further (link here).

In short the economic, social and human cost of not investing in addressing the deep-seated disadvantage which derives from poor and overcrowded housing will be substantial, with the most intense impacts falling unnecessarily on remote Indigenous citizens.

Apart from renewing the National Partnership with a significant level of funding, what policy options exist for the Commonwealth (and the states)?

I would propose four possible policy initiatives, though I am sure there are others.

First, the lack of progress in opening up opportunities for private sector investment in remote community housing is a glaring policy failure by the Commonwealth. Lack of progress in advancing land reforms, which to be effective must not lessen the rights of Indigenous traditional owners or native title holders, has been one major impediment. Another has been a lack of innovative policy vision within governments. A third has been the impact of transaction costs and other hidden disincentives which hold back investment even in communities or locations which have made progress in establishing more flexible institutional frameworks (eg the township leases in a number of NT communities). And fourth is a lack of access to capital. The focus by governments on moving straight to higher levels of home ownership is in my view misdirected. Governments would do better to concentrate on building up a rental housing market in remote Australia. It is currently virtually non-existent.

One way to kick start such a shift would be to establish a government owned corporation with commercial powers to borrow in the private sector (and which might joint venture with Indigenous corporations) to build, own and rent out housing within remote communities. Such an initiative would address a number of pressing issues: the shortage of housing assets; the lack of staff housing in remote communities (which acts as a disincentive to attract and retain both locally engaged and external staff); open up new sources of private sector capital for investment in remote locations; and would begin to expand access to the Commonwealth Rent Assistance program to remote Indigenous residents who to date have been structurally denied access.

Second, policymakers could focus on supporting the introduction and expansion of three or four community housing providers across remote Australia to complement the operations of state housing authorities in remote regions. Notwithstanding the fact that the NPARIH reforms introduced institution arrangements which for the first time established formal accountabilities for the owners of social housing, the experience to date has been disappointing with state housing authorities yet to deliver on their responsibilities. The introduction of competing providers may be necessary to demonstrate that poor housing is not an inevitable and endemic feature of remote regions.

Third, given the propensity of the states and the NT to underfund PTM in remote regions, the Commonwealth should move beyond rhetoric and take a much more direct role in either delivering PTM services itself (perhaps by injecting Commonwealth officials into the relevant parts of state and territory housing authorities) or in establishing a new statutory office to monitor, oversight and regulate the provision of PTM services in remote communities. Such a policy initiative might need only exist for ten or fifteen years until new expectations were set and embedded into bureaucratic practice, but would ensure that the substantial capital investments to date, and hopefully the value of the considerable investments into the future by the Commonwealth will be protected from dissipation and accelerated wear and tear arising from poor PTM which shortens asset lifespans.

Fourth, one innovative element of NPARIH which actually worked to improve state and territory performance, and which was discontinued after the change of Government, was the capacity for funding allocations to be varied both up and down in response to state performance in delivering against key metrics. While the Review Panel (section 4.7) argues that the competitive bids process (as this financial incentive system was known) led to ‘undesirable and unintended consequences’, it is salutary to reflect on the fact that the best performing state in terms of value for money, South Australia, was penalised by the Commonwealth for poor performance in the early years of the program. The bottom line si that states and territories will respond to financial incentives, and given the importance of PTM for protecting the Commonwealth’s investment to date, there would be value in giving serious consideration to re-instituting the system with a focus on PTM delivery.

To sum up, the prospects for the renewal of the current levels of Commonwealth investment in remote Indigenous housing do not appear positive. The Minister and his Department appear to be basing their policy development work on a Review Report which is fundamentally flawed, and in particular, which substantially under-estimates the outstanding housing need across remote Australia.

The costs of not addressing this need effectively will be significant and will fall on taxpayers in the future, and importantly, substantially constrain the material life opportunities of thousands of remote citizens over the course of their (shortened) lives. The asymmetry between the short term political benefits of reduced investment and the associated long term social, economic and health costs is driving a decision-making dynamic which is not in the public interest, and certainly not in the nation’s long term interest.

How this issue is resolved will be a test; a test of the institutional capacity of the Australian Public Service to drive good policy, a test of the calibre of our political leaders, and ultimately a test of our national character, and whether we are the land of the fair go.



Disclosure: In the interests of full transparency, I should indicate that I worked as an adviser to Minister Jenny Macklin from 2008 to 2011, and as a senior public servant with responsibility for NPARIH from 2011 to 2013.




[1] The Sunburnt Country Review in 2007 states there were 21,000 houses managed by Indigenous Community Housing Organisations across Australia, but doesn’t split this between remote and regional areas. NPARIH has built 4000 houses nationally in remote regions, and refurbished 7500, so there are at least 11500 remote houses. The figure of 15000 is my estimate. Achieving effective asset lifespans of 30 years assumes effective PTM policies. In many locations, the effective asset lifespans are likely to be shorter.
[2] I am not in a position to make a detailed analysis of how many additional houses would be required to reduce overcrowding by an additional ten percent. However, NPARIH’s extra 4000 houses reduced overcrowding by 15 percent, and the Review estimates that a further 5500 houses would reduce overcrowding by a further ten percent. It seems likely that at least 5000 houses would be required to reduce overcrowding by a further ten percent.
[3] This figure is calculated based on the Review’s base figure of 5500 new houses to meet existing needs plus population growth, the estimate of 5000 new houses to replace assets which reach the end of lifespan, and the approximately 5000 additional houses required to reduce overcrowding to metropolitan levels.
[4] Table 5.5 of the Review report suggests that average weekly rents are around $90. If we adopt the conservative assumption that rents cover 20 percent of maintenance costs, then average annual maintenance costs will be $23,400 per annum, of which only $4680 is rent. If there are 15,000 houses nationally, then the additional funds required exceed $280m per annum, or $2.8bn over the next decade.

No comments:

Post a Comment