Skip to main content

Housing Affordability: report to the Premier

The NSW Government's housing affordability package of measures was developed with the oversight of former Reserve Bank of Australia governor Glenn Stevens.

On this page

By Glenn Stevens AC

Download the Housing Affordability Report to the Premier (PDF, 484.41 KB)


Housing affordability seems to be a perennial topic of discussion, particularly in Sydney. In many respects this is hardly surprising. The city and its environs are inhabited by an increasingly affluent and growing population, one in which the production and consumption of services and ‘experiences’, including housing services, form a growing part of our economic life. Serious economic hardship – like widespread high unemployment – is a rather distant memory for many. Also a distant memory is the housing hardship faced by earlier generations who, in the aftermath of wars and depression, lived in a much less affluent community, where land was plentiful and cheap but building materials and finance were often much scarcer than they are today.

Our purchasing power over many of the things we want has increased greatly. The cost of producing manufactured goods, for example, has declined to levels unimaginable 50 years ago. Today motor vehicles, television sets and so on are a fraction of the cost, relative to the average income, that they once were. They are virtually disposable items. The reason is the increase in productivity and efficiency of production, brought about by technological change and international trade. Many of the services we consume, particularly those delivered using modern information technology, are likewise much cheaper. Some can be produced or replicated at marginal costs of virtually zero.

Some other goods, in contrast, cannot be so easily reproduced. In Sydney, there is only so much harbour-front land, or other land within a short commuting distance to the CBD. While Australia has a long coast line, not much of it is within easy commuting distance to centres of commerce and economic activity, at least with current transportation networks. As incomes have risen and preferences about housing have changed, the prices commanded by land in the most desirable locations have increased a great deal. As a community, we can increase the efficiency of use of such scarce land through ‘densification’. We certainly cannot all live in traditional detached-bungalow, low-density housing with views and expect to walk to work.

Of course, this is not a phenomenon seen only in Australia. In fact, it is seen around the world in large, otherwise successful cities. Likewise, the advent of sustained very low rates of interest, which unquestionably works to raise all asset values, including property, is a global phenomenon. If anything Australia has stood out in the period since the international financial crisis for interest rates which, though low by our historical standards, were rather higher than those on offer in most of the developed world. And over recent years the same interest rates have been in operation in Perth, where prices first rose and then fell, in response to real economic factors – such as the mining boom and its aftermath. So while financial or other common factors are doubtless at work in all housing markets, idiosyncratic real factors – local factors – also have a big influence.

The extent of increase in the cost of shelter in Sydney has been quite prominent in the international context. Making comparisons across countries is hazardous because of differences in data, differences in the way populations are distributed across nations and so on. Some of the more colourful claims about housing affordability in Australia relative to elsewhere are, in my judgement, exaggerated.

That said, it is hard to avoid the sense that, for a country that ostensibly has so much space and relatively low population for its physical size, we seem to have ended up with the cost of shelter being quite high. For Sydney in particular, the problems seem more acute than in other Australian cities. Prices have risen further than elsewhere in Australia.

A host of explanations have been offered, and often they focus on one simple story. The reality is almost certainly more complex and we should be sceptical of simple stories and the simple solutions that are usually offered with them. A range of factors are likely to have contributed to the situation we face. If so, a range of responses are likely to be needed to address the problems. And even good responses are likely to take quite a long time to have their full effect.

On the demand side of the market, it should be noted that, in recent years, the NSW economy has been outperforming most other regions of the country, helping incomes to rise and employment to grow faster than elsewhere, and attracting new migrants, particularly as the mining investment boom tailed off. This came after a fairly lengthy period of underperformance in NSW following the 2000 Olympics - some of it is ‘catch-up’. So in a way, some of what we see is the sorts of problems associated with a growth economy. That is to say, prices are high in large part because so many people could afford them. In addition, Sydney is highly internationally visible, and perhaps attracts a disproportionate share of foreign investment in Australian residential property.

Rising incomes will result in higher prices for a more or less fixed supply of existing homes in the most desirable areas. Shifts in preferences may put a premium on proximity. An example is the often-cited desire of ‘the younger generation’ to live in or near the CBD. Presuming that this is in fact the case, such a change will necessarily cause relative price changes in the property market. That is the market’s way of signalling that we can’t all live in the most desirable areas, that proximity has costs as well as benefits and that it takes time to adjust the structure and location of the stock of housing, since the feasible rate of construction is small compared with the existing stock of dwellings.

As real as these problems are, they are preferable to the problems of stagnation. Still, over time, we need the supply side of the market to respond to the community’s needs, and to the growth in population and income. It is highly desirable that the supply side of the market meet the demand for shelter for the bulk of people efficiently and at minimum cost.

Most observers agree that the supply side of the market in general has struggled to keep up with demand, probably for most of the past decade. To be sure, construction rates (a flow) are presently at record highs. But there is a stock issue: the previous period of underbuilding beginning around the mid 2000s leaves a cumulative shortfall in the number of dwellings, estimated by the Department of Planning and Environment (DPE) to be as large as 100,000 dwellings. Now such estimates are based on observed population growth and assumptions about household formation rates and so on, and are, even if implicitly, estimating a notional demand that people would have if they were not constrained by prices. In practice demand in the private market is rationed by higher prices, the market clears at any point in time, and everyone lives somewhere, even if it is in accommodation less well-located, or more crowded, or of a lesser standard, than they might wish.1 This is what price changes are supposed to do in a market system.

But it seems reasonable to think that the community’s well-being would have been significantly improved had the supply side been more responsive. Certainly, if our objective is housing being “affordable” in an environment of growth in population and income, we need to have the market clearing at lower prices for dwellings. This means we need to have the supply side able to respond to demand in a more elastic way. The only alternative would be to find other, non-price, ways of rationing demand. Tempting as these might be at times, they are likely to have serious unintended consequences. Even if they did not, suppressing demand at any given price level is surely inferior to meeting genuine demand through higher productivity.

Some things on the supply side are not amenable to change. Sydney already has a fairly large footprint. It is bordered by sea and national parks, and has some other terrain challenges. A casual comparison of the Sydney environs with those of Melbourne suggests Melbourne could expand almost without limit in most directions while the potential for expanding Sydney’s footprint is, although not exhausted, more limited.

Other things can change but slowly. Importantly, Sydney has less than adequate transport infrastructure, a critical weakness. The set of reasonable trade-offs that people can make in terms of housing cost and commuting distance are seriously limited by slow transport, congestion and so on. The Government is on the right track in addressing this, but realistically it will take many years of investment to rectify the problems.

Other things ought to be more amenable to change more quickly. If, as some suggest, things like regulation, zoning, the approvals process and so on are making the supply side less than flexible, these are things that are in our power to change.2 More than one developer has pointed out that the cost of supplying a given size of dwelling is considerably more expensive in Sydney than in Melbourne or Brisbane due mostly to high land prices, and regulatory costs. The amounts involved are material: they talk of additional costs of the order of $150,000 per dwelling. If changes to regulations and charges could alleviate this, the price of shelter could, in time, be materially lower.

To achieve that, however, political leaders will have to explain to communities why population growth is likely to occur, why we need to accommodate it, and what that involves. This is a fundamental point. NSW needs to plan for growth. It needs to think about how the modern economy, strongly-service oriented and driven by innovation, which in turn relies on collaboration and connectedness, can be accommodated. It needs to think about how the population that will be attracted by the opportunities afforded by such an economy will be housed, moved around, cared for, educated and so on. Looking at the present situation on housing and infrastructure, and notwithstanding the very welcome steps the government has been taking in recent years, it is very hard to avoid the conclusion that, for a long time, we didn’t quite have that plan. Our community’s collective unwillingness to think ahead and to build adequate infrastructure meant that when growth occurred, it caused more congestion, cost escalation and inconvenience than it really needed to. This in turn creates community resistance to further growth. This is a situation that developed over a few decades. While investment is under way now to try to catch up, realistically a problem that took 20 or 30 years to come to a head won’t quickly be solved. While we catch up on past lost ground, population continues to grow.

So if government is serious about tackling the issues at the heart of the ‘affordability’ problem, and not just responding to symptoms, it needs a plan for growth. It also needs to articulate to a sometimes-sceptical populace – those who are already here - what we need to do to accommodate more people, why growth without a plan is a not a good outcome but also why an even worse outcome would be stagnation. After all, houses tend to be quite affordable in locations that are declining - because people don’t want to live there.

DPE has research suggesting that many more medium-density homes – up to 600,000 - would be feasible for developers to supply today if planning controls were not a constraint. Of course there have to be some controls, and this is a difficult balancing act for state and local governments. Among the interests that must be considered are those of:

  • some existing residents, who value amenity from hitherto quiet suburban neighbourhoods of detached dwellings;
  • other existing residents who would be keen to realise the potential uplifted value of the land on which they presently live if re-development were allowed; and
  • potential new residents – both from natural population increase and interstate and international immigration.

Local councils are elected by the first two groups; the question is whether anyone at local level represents the third group, apart from developers who of course have their own interests. My judgement is that political leadership at State and indeed Federal level is critical if the interests of the third group are to be given adequate weight.

The role of the Greater Sydney Commission would also appear to be central. The GSC has a well-articulated high level vision for the ‘three cities’ future of the Sydney environs, and it works to develop housing supply targets for the various districts. But delivery of these targets must, in large part, go through local council processes, and there appears to be no sanction on councils if they fall short of the target. And in fairness, some local councils may not be adequately resourced to process high volumes of development applications. Difficulties in coordinating provision of necessary infrastructure, on which approvals are contingent, cannot all be sheeted home to local councils either. So there may be benefits in the state government considering not only how to expedite zoning and approvals and reduce the cost of regulation, but also contemplating how to give more teeth to the GSCs targets, and facilitating more effective infrastructure coordination.

How realistic it is to think that all this can be done is hard for me to assess. It appears that the history of reform in this area is not a happy one. Perhaps the main thing that has changed is the greater sense of urgency, after more years of rising prices. While it is a political judgement for the government how aggressive to be in seeking reform, inaction is costly and becoming more so, while it is far from clear that the politics of reform will get easier.

To repeat the point about narratives, whatever the Government ultimately decides to do would need to be part of a conversation about how some more medium-density development will bring improved amenity on a range of new dimensions – local vibrancy, diversity of commercial and retail activity, etc. People need to understand how changed neighbourhoods can be better than unchanged ones.

What I Agreed To Do

After the Premier’s invitation and following a meeting with her, I agreed to:

  • Engage with the relevant officials to understand and evaluate their work and policy proposals, as an independent and outside ‘additional pair of eyes’.
  • Give the Premier my assessment of the issues and proposals developed by the officials, in writing.

It is important to be clear what was, and was not, in scope. The work that I took part in or observed was consciously not aimed at the problems of the genuinely homeless, nor was it aimed at social housing. These are important issues, but I am not in a position to offer an opinion on whether there is a case for government to do more in these spaces. The focus has been on the Sydney region, not regional areas of NSW. That does not mean affordability issues are unimportant in those areas. There will be some spillovers from expensive Sydney housing to the regions. And, at the margin at least, some regional centres could be part of the solution, insofar as better transport connections would make it more feasible and attractive for people to live outside Sydney but still be part of the connected community centred on Sydney.

But to be tractable the focus has been on Sydney. The focus of the work in which I participated was, in a nutshell, how we might make it easier for people of moderate means to own a dwelling in which they would live in the Sydney region. Even this raises some very fundamental questions of economic strategy, investment, regulation and political leadership.

My Process

I met with officials from the Departments of Premier and Cabinet, Planning and Environment, and Treasury. Officials from the Premier’s Office attended some of these meetings. I met with the Minister for Planning and Environment, the Minister for Education (previously the Planning Minister) and the Treasurer. I participated in meetings that officials held with other interested groups. In addition I held meetings with other interested parties including (but not only) developers, in some cases at their instigation and in other cases at mine.

Observations arising from the process

I found the quality of people with whom I engaged to be high. Most of the officials were very experienced, seemed on top of their brief, and had nuanced understandings of the complex issues. I found their assistance to be invaluable.

Meetings were efficiently conducted and in a spirit of genuine discussion, probing and debate. This reflects well on those involved and I think the government is well served by its officials. Officials are principled. They are also realistic: they cannot be unaware that some policy ideas will be politically unpalatable or even unacceptable. But it is important that the government always ensures that it hears all the policy ideas, including the unpalatable ones.

The private sector were generally constructive and seized of the opportunity to make a difference. Of course they have their own agenda, but I felt that they entered into the spirit of the discussion.

Key themes that emerged from the meetings

I am aware that various meetings were occurring to which I was not a party. That is entirely appropriate but I can only comment on what I saw. From the meetings I attended, the themes could be summarised as follows.

First, the size of the task is very large. The GSC growth targets envisage, roughly speaking, about 700,000 new dwellings in the Sydney region for an additional 1 ¾ million residents, over twenty years. There are about 1.7 million dwellings in the Sydney region now as a result of accumulated investment over the period since 1788. The plan is to add, in 20 years, 40 per cent in net terms to that stock. Such an outcome would require current rates of commencements, which are at a high, to be maintained, or exceeded, on average, over two decades, in a notoriously cyclical industry – and without costs escalating significantly. This is, to say the least, an ambitious goal.

Looked at through another lens, the required addition to the dwelling stock is of the order of 2 ¼ per cent per year. Such a rate of growth has been achieved or exceeded at times in the past. For example, over 200 years the average rate of growth of the stock of dwellings was probably over 3 per cent per year. But that was starting from a very low base and, for most of our history, occurred in a less complex environment and with cheap land. The challenges of achieving the expansion that the planners say is needed, including all the associated infrastructure, is not insurmountable, but is very large.

It may be that these projections/ aspirations are too high. Any long run projections are extremely sensitive to assumptions – about household formation, population, preferences and so on.3 Were there to be a material slowing in economic growth in NSW, we would probably find that population flow and demand for new housing would lessen for a time. The issues of affordability and congestion could well go down the list of concerns of the public, replaced by worries about joblessness, mortgage defaults and so on, not to mention pressure on the State budget. But while such slowdowns inevitably occur from time to time, it is still important to plan to grow. Arguably a plan for growth is even more critical at such junctures.

Second, the issues and processes around housing supply are complex. Facts are more difficult to find and assess than is desirable. As an outsider, I found it difficult to understand all the moving parts: who has land where, what is involved in developing it, which bodies need to be approached to approve a proposal, or evaluate a study on traffic, environmental concerns and so on. This is inevitable in an engagement of this nature but it occurs to me that, while the large developers have the skills and resources to navigate the system, smaller players would struggle. That cannot be helping speed up supply.

Third, land prices in Sydney are very high. A study commissioned for the government by Urbis provides the following comparison of land acquisition costs per square metre for development.

Land Acquisition Costs by Density and Distance from CBD

Density Distance from the CBD Land acquisition cost per square metre
Greenfield (Kellyville) 37.7 km $1,294
Medium density (Mascot) 8.4 km $3,928
High density (Mascot) 8.4 km $2,828
Greenfield (Wollert) 44.2 km $218
Medium density (Brunswick) 5.7 km $1,964
High density (Brunswick) 5.7 km $680

Source: Calculations by Department of Planning and Environment based on Urbis Pty Ltd Costs of Development in NSW: Draft Report

Other data I have seen from one national developer tell a similar story. As Urbis also points out, since many other costs – professional service fees, stamp duty and so on – essentially key off final sale values, these high land prices are multiplied up through the cost structure.

To be sure, Sydney has geographical challenges that some other capitals do not. But it is worth asking the question why land has to be so expensive. Are there artificial constraints to land supply that may be exacerbating this problem?

The costs are compounded by unwillingness to contemplate smaller lot sizes, in contrast to some other cities. If land is genuinely scarce, then we need to be prepared to use it more efficiently.

These factors go most of the way to explaining why the cost of dwellings in Sydney is so much higher than elsewhere; according to Urbis research, there is limited evidence for construction costs being materially higher in Sydney than elsewhere.

Fourth, the length of time taken for rezoning, and development approval is much too long. This lowers the short to medium term elasticity of supply and raises long run costs. Developers’ capital tied up over many years of owning land but not being able to start construction is costly (and would be more so if the cost of capital starts to normalise with global interest rates heading upwards). The complexity of the process also inhibits competition since smaller developers will have fewer resources to learn how to work within the process than larger developers. In part the slowness of the process reflects resource constraints within both the state Department of Planning and Environment and local councils: the sheer volume of work to be done overwhelms the best efforts of the relevant officials.

But it also, according to many participants, reflects the nature of the process, with many development approvals being subject to bespoke processes and delays due to the difficulties coordinating the requirements of various agencies. New South Wales has a small proportion of approvals granted under ‘code assessment’ or ‘complying development’ compared with other States. Doubtless proponents of the current arrangements will defend them on the basis of ensuring quality, but if we have inadequate supply it is worth asking what evidence exists that higher use of code assessment in other states has produced a notably inferior housing stock. This is something that I am not able to assess here, but would be worth further study.

Freeing up the supply side will require addressing these sorts of issues by making zoning happen faster, DA processes more easily coordinated and less detailed. If the rules for development are transparent and clear and well understood, it ought to be possible for ‘yes’ or ‘no’ answers to be provided more quickly. I am also persuaded that there is a case for the elected representatives setting the boundaries but delegating the decisions on development applications to appropriately-qualified officials.

Fifth, the GSC has set out targets for dwelling supply at the district level. But as noted above there seems to be little sanction on councils that do not meet their share of the target. Hence it is worth canvassing options for these targets to get more traction.

Sixth, measures that seek to assist first home buyers could be more targeted. As a general observation, I personally do not favour these measures and I could not recommend extending them.

This is less because of the common objection that such measures will inflate prices – while that will be the direction of the effect, I am unconvinced that the effect is all that large.4 The point is rather that the Government may be able to achieve a lot more by using those funds differently. After all, giving people money so that they can afford high prices not only does nothing to lower prices, it is almost a counsel of despair. Whereas spending $200 million more per year on resourcing government and council agencies to expedite re-zoning and development approvals and facilitating infrastructure investment would, in the long run, probably do much more to achieve the most effective thing we can do for first home buyers: lowering the cost of supply.

But even if, for political reasons, the Government feels obligated to persist with these measures for first home buyers, it is important to note that at present about half of the grants made by the government do not go to first homebuyers: they go to those buying their second or subsequent home, if it is a new dwelling. There may have been a case for this to assist the building industry during the financial crisis, but the builders don’t need that demand stimulus now: what they need is faster processes for converting zoning and development applications into construction and an ability to respond to those segments of the market that will accept smaller, lower cost dwellings. In addition, we should be considering the question of how many first home owner grants go to those whose first home is an investment property, and whether we want to subsidise that investment strategy.

Seventh, and while on the topic of investment, a theme that emerged from the various meetings is that the structure of the rental stock in Australia is unusual in that there is little in the way of institutional ownership: rental dwellings are largely in the hands of individual investors. There are some calls for greater investment by private institutions in rental dwellings - ‘build to rent’ - and so on.
It would be worth the Government and/or the Commonwealth Government consulting tax experts on this matter, but in my judgement it is very likely that the structure of ownership reflects at least in part tax incentives, namely negative gearing and capital gains concessions to individuals. Individual investors are prepared to accept very low rental yields, because of the tax concessions. Institutional investors are unlikely to find those yields attractive. It would be a bad mistake to offer them a separate round of tax concessions or subsidies in order to secure their involvement. It would be better to lessen the generosity of the concessions to individuals.

Eighth, the role of foreign purchasers was raised less often than I would have expected. This could be a result of the particular groups with whom I met, or it could be that foreign purchases are only important in a fairly limited set of locations, whereas ‘affordability’ is an issue right across the housing market. In the discussion about whether, and by how much, we might seek to levy taxes of some kind on non-resident holders of property, we need to be clear about our objective. If we want to significantly lessen these purchases, we may need to be fairly draconian, but we may find it difficult to calibrate the application of measures. For example we would not want to provoke a simultaneous sell off by a large number of foreign holders. If on the other hand we want simply to raise revenue then, to a point, a relatively inelastic demand on the part of foreign buyers is a good opportunity. But any decisions here will be difficult because, so far as I can judge, we don’t have very good information about the extent of holdings of foreigners. It would be worth examining the experience of other jurisdictions to assess how effective these sorts of measures have been, in raising revenue or dampening foreign demand.

Finally, the issue of the caps on council ‘section 94’ charges was raised several times. Here there is an obvious tension. On the one hand artificial price restraints, sooner or later, bind in ways that were not anticipated and cause distortions. This argues for removing the caps lest they result in worthwhile development not proceeding. On the other hand, we should be asking whether the associated infrastructure could not be delivered at a lower cost: is the infrastructure being unnecessarily ‘gold plated’? This argues for leaving the caps in place for the moment, and pressing developers and councils on the issue of what the minimum cost for infrastructure really is. It would be worth benchmarking these costs against those in other jurisdictions, so that a decision to raise or remove caps, or leave them in place, could be well informed.

Evaluation of the proposed actions

During the second half of April and early May, officials from DPC, DPE and Treasury worked through several iterations of possible measures aimed at helping housing affordability. I was able to observe the evolution of the proposals through a sequence of meetings, though I was not able to take part in all the meetings.

The proposed measures on which I offer an opinion below are attached as Annex I.

My comments are, unavoidably, fairly high level. This is for two reasons. First, it is not possible for me to assess at a detailed level the technical feasibility of many of the measures, nor to determine whether or not their costings are accurate. I have not seen and offer no opinion about the quality of modelling work which may have been done. I, like the government, can only rely on the quality of the work of the officials. My judgement, based on the discussions I have had, is that there are grounds for reasonable confidence in the officials’ capability to design measures effectively, provided that the design parameters are not unduly constrained.

Second, the details have been shifting as officials have interacted with the government over recent weeks. Some aspects of the proposed measures have only been available to me since late last week. This is unavoidable, but it also means that I must be circumspect about endorsing the details of particular measures.

With those two caveats, the proposed measures generally are aimed at addressing the right issues, as I see them. They are arranged under four main headings. I consider them in that order.

1.More Supply.

The thrust of my analysis above is that the supply side of the housing market is too inelastic. That is, when demand increases, the quantity of housing available is very slow to respond, so that the market clears in the short run, and even the medium run, by prices rising. As dwelling prices rise, development and re-development becomes more feasible on land which developers already own. Supply then accelerates. But this is a slow process and it would be preferable for development to be more feasible at lower levels of dwelling prices.

One thing that we clearly want to do is to accelerate the process by which new supply can come on stream. Unavoidably, a considerable degree of planning is required in producing housing and of course for other potential uses of land. But these processes are very slow – for example it can take 5 years for a Local Environmental Plan to be finalised.

Given that, measures which incentivise the GSC and local councils to expedite their planning processes are to be welcomed. Since councils are in many cases resource-constrained, being prepared to support some additional resourcing, in a targeted way and subject to hard evidence of delivery schedules being met, makes sense. Making some more resources available in the relevant areas of DPE to support a reduction by half in the time taken for re-zonings to be processed also makes sense.

To try to de-politicise the process of granting Development Approvals, it is suggested that Independent Hearing and Assessment Panels be used more widely and indeed be compulsory in many instances. It is also suggested that for certain classes of proposed dwellings there be clear standards and less scope for discretionary decisions to deny development approvals in cases where the proposal clearly meets the standards. Such an idea will be resisted by some on the grounds that quality may suffer, which is certainly a risk. But given all the evidence, we have to ask whether NSW errs too far on the side of bespoke processes and discretionary decisions, in comparison with other jurisdictions in Australia. Hence I support the thrust of these proposals.

Proposals to facilitate the resolution of impediments between Councils, Government and builders are only briefly mentioned. This is one of those ideas which always sounds appealing, but which often have little practical effect. In this case, though, improved coordination could, by dramatically shortening delays in the supply response, make a tangible difference to affordability. I support it, but suggest that the proposal be made as specific as possible. The coordinating body or individuals need to be given some genuine authority and resources to expedite matters in the supply chain.

Local councils may face incentives unduly to protect what they judge to be the interests of certain existing residents, and to give less emphasis on the interests of potential new residents who do not as yet pay rates or take part in elections. Since the State Government can, and has, adopted a wider perspective, the ability in extremis for a minister to appoint another entity to, in effect, assume the role of the relevant planning authority is a useful tool. It would be my hope (and I imagine the Minister’s) that, given the other measures and incentives being established, the use of such power would be rare. But I support it being in the government’s arsenal. The Greater Sydney Commission already has certain powers of this nature. It may be worth careful consideration how that power and the minister’s may overlap.

As noted above, our attitudes to land use need to be consistent with the degree of shortage of land. It doesn’t make sense to persist with restrictions on lot sizes dating from earlier times when land was plentiful and cheap, and which stop the market offering combinations of house and land space which people are prepared to choose. Hence we have little choice but to support measures that will allow smaller lot sizes, thus making feasible the supply of terraces, town houses and the like. Regarding parking space regulations, I personally would be wary of allowing large apartment buildings without any off-street parking. But it does make sense to allow some flexibility in how parking is provided, including potentially by unbundling parking spaces from particular apartments, considering in some cases above-ground parking and by taking account of proximity to good public transport.


A number of measures are proposed to re-purpose existing infrastructure funding, and to lift the cap on developer contributions, aimed at speeding up delivery of housing. It is also proposed to close a scheme which exposes the government potentially to very high costs in matching developer infrastructure contributions in some areas.

It is difficult to assess these measures, since there are many moving parts. What is clear is that it is very important to ensure that the right infrastructure is built, and that stakeholders see value for money, when either developers or government are required to stump up for infrastructure. Hence the proposal to ensure that IPART, or some other body, has a clear mandate to ensure value for money is very important. I think it is imperative that this, or something like it, accompany any decision to lift caps.

Regarding subsidising council borrowing, I am not opposed to government entities borrowing for the purposes of worthwhile capital investment, though I suspect that the case is stronger at state and federal levels than at local level. I would proceed with caution here. We would want to be confident in councils’ decision-making processes and their skill sets in managing a debt portfolio before encouraging too much borrowing. That may well require further study. But I do not have an in-principle objection; I simply counsel caution.

3.Support for First Home Buyers

As noted above I have not been a big fan of these measures. But if the argument about whether society should help those seeking an initial foot on the ownership ladder is over, then the question is how to do so most effectively. The real value of the support has declined as dwelling prices have risen, which suggests that thresholds could be reconsidered. Done carefully, the extent of help could perhaps be moved up, carefully, to recognise the price rises that have already occurred, without this measure itself materially contributing to further upward pressure on prices. I note that over successive versions of the draft proposals I have seen, the generosity of what is proposed seems to be increasing.

The more that additional generosity to some is combined with a tightening up in some other area – such as removing the grant given to anyone buying a new home – such that the changes are more or less budget neutral, the more easily it can be defended against the criticism that it is self-defeating by inflating prices. Such an approach would also mean that the measures more closely target the group the government wants to support more effectively. But I also note once again, for the record, that the government might expect to achieve much more for affordability in the long run by spending this money in other ways, that would lead to lower cost supply of new housing.

The latest proposals refer to shared equity arrangements within the Community Housing Provider system. I am unable to comment on the details of this system. My main high level comment is that while shared equity arrangements have a certain appeal in principle, it will be important that the rights and obligations of all the parties are well understood. If, some years hence, prices have risen further, we do not want the beneficiaries of the scheme to claim that they didn’t understand that the capital gain must be shared with someone else. And if prices were to fall, how that is borne between the two equity providers will need to be very clear as well. It is not hard to imagine loud complaints if and when it comes to sharing losses. I would keep such schemes small until more experience, through the whole range of the cycle in the market, is gained.

Under the heading of ‘addressing community concerns’, a number of detailed measures are proposed, some of them taking a slightly tougher line regarding foreign purchasers than in draft material I had seen, and there are some specific proposals which I had not earlier seen. My general comments on the issues of taxing foreign purchasers/holders were made above. To reiterate, we should be clear what our objectives really are; it will be difficult to calibrate these measures, partly but not only because our information is incomplete; and understanding clearly the experiences of other jurisdictions in the use of such tools would be instructive. I am unable to offer much guidance about most of the specifics here.

4.Targets and Accountability

Having a couple of high level, simple targets has the advantage of showing the government is serious. It also serves as a frame within which the narrative to the public can be continued. Not only that, well-chosen targets can focus the energies of key decision makers and players in a productive way.

Of course, once articulated, targets almost certainly will become a rod for the backs of the government. That is the intention, in fact. People will want to know whether they will be achieved. And a target or set of targets not carefully chosen can result in key actors losing sight of other variables which are important in the delivery of good policy but not articulated in the numerical target.

The message from that is not to avoid any target, but to choose carefully. I think that having a goal of “x per cent of DAs being approved within y days” makes a good deal of sense because it focuses on one of the core problems. It is a matter of choosing values for ‘x’ and ‘y’ that are something of a stretch, but not impossible.

Regarding something like a yearly target for dwelling completions, it is important to remember that construction is and always will be a very cyclical industry. That being the case it would make sense to frame an objective here not as an annual number in any particular year but as an average over several years. It should be something that, if achieved, will keep us on track to meet the 700,000-plus dwellings over 20 years discussed earlier in this report.

I think it is important that the long run projections of demand, and any associated annual or moving average target be reviewed periodically. The frequency should not be annual because such a process would be hostage to the vicissitudes of the cycle. But it should be often enough that the projections and targets are seen to remain relevant as circumstances evolve, as they will. I would think that at intervals certainly not exceeding five years all the parameters should be reviewed.

Suggestions for further work

Apart from the suggestion immediately above about reviewing projections that motivate much of this whole discussion, which may already be in train, I have only one other: it would be useful to know the answer to the question whether the greater use of ‘code assessment/complying development’ in other jurisdictions has a material effect on the quality of dwellings on offer. This information may already be available somewhere, but I have not seen it.

Concluding Observations

It has recently been suggested by some commentators that the housing market is peaking. It is impossible to know whether that is so, but in some respects it would make sense: if housing really has become ‘unaffordable’ for many, demand will slow, unless credit becomes progressively more available on easier terms to compensate. But credit, as measured, has not picked up much at all and the authorities are voicing concerns about some aspects of the credit landscape. Banks are also being more cautious. Non-bank lenders are attempting to fill the gap, but probably cannot be expected to do so fully – nor do we want them to. And, of course, dwelling supply has finally started to respond to the surge in demand.

Should a slowing in prices become clearer over the months ahead, the Government will naturally be expected to consider whether its response should be altered. One area for caution might be demand side measures like taxing foreigners; if foreign purchasers are slowing down anyway, we may not want to push them down further.

But more generally, if the market is slowing, that affords a breathing space to allow long-run supply reforms to be brought in. That won’t make a major difference to the current cycle, but would allow the next upswing to be handled with more supply elasticity. It would also give Sydney the chance to remain on track for the long run goal which, as noted earlier, is a demanding one. At the risk of being unduly repetitive, the most important thing that the political leadership can do is carry the careful, balanced and realistic conversation about growth and how to accommodate it, to the public.

On a final note, I thank the Premier for the invitation to complete this engagement. I also thank those who have met with me, and especially Simon Draper, his team, and the teams at Treasury and DPE, for their courtesy and patience in answering my many questions.

ANNEX I: Proposed Measures Provided by Department of Premier and Cabinet

The list of proposed measures of which I was aware as of Friday 19 May 2017 is reproduced below for reference.


More supply

More housing supply in the right areas

i.) The Greater Sydney Commission (GSC) will be required to ensure their final District Plans contain housing supply targets for each local government area (LGA) for periods of 5 and 10 years, sufficient to cater for expected population growth and address the estimated 100,000 undersupply in Greater Sydney over the last decade (noting this undersupply will need to be progressively addressed over the medium to long term).

ii.) The GSC will only approve changes to a Local Environmental Plan (LEP) if the GSC considers the LEP gives effect to the relevant District Plan targets on housing supply and diversity, and the proposed residential zones are considered to be feasible to develop.

iii.) The Government will require 10 priority local councils in Greater Sydney (as identified and recommended by the GSC) update their Local Environmental Plans (LEPs) to reflect the final District Plans within two years of the GSC finalising the District Plans.

iv.) The GSC will allocate up to $2.5 million per priority council to assist and encourage them to update their LEPs. Payments will also be available to up to five other councils that volunteer to accelerate updating their LEPs

v.) All other councils in Greater Sydney will be required to update their LEPs with appropriate housing targets within three years of the release of final District Plans.

vi.) Where any council fails to meet its obligations in updating a LEP, the Minister may repeal the existing LEP and the GSC may assume the role of the relevant planning authority to update the LEP in line with the District Plan.

vii.) The Department of Planning and Environment (DPE) will require that local housing strategies protect local character as well as supporting growth and housing targets.

viii.) The Minister for Planning and Environment will accelerate rezoning of around 30,000 additional dwellings in existing priority precincts or new partial precincts.

ix.) The Minister for Planning will require Landcom to take an active role in the market to support housing affordability.

Faster planning approvals

x.) Mandatory independent panels to determine all local development applications (other than smaller DAs delegated to Council staff) will be introduced for all councils in Greater Sydney and specific regional local government areas. The process of appointing members to independent panels will be updated to ensure members have professional qualifications and no conflicts of interest

xi.) GSC District Commissioners will be removed as members of Sydney Planning Panels, which will reinforce the delineation between the GSC’s role in strategic direction setting, development applications and rezoning approvals.

xii.) The Minister for Planning will reduce the complexity of Development Applications (DA) by introducing baseline standards (e.g. height or setbacks) for the scale of residential development allowed in priority precincts and priority growth areas in Greater Sydney, which will be based on studies of suitable building sizes for each precinct and will retain Councils’ responsibility for maintaining the character of their local areas by approving the final design of buildings as part of a DA.

xiii.) The Minister for Planning will release of proposed amendments for a Greenfield Housing Code to simplify development standards for one and two storey dwellings being built under the complying development approval pathway.

Smaller houses at lower prices

xiv.) The Minister for Planning will:

a)require Councils to create smaller minimum lot sizes when updating their LEPs in line with District Plans, and that minimum lot sizes may vary across Sydney LGAs depending on character and land value.

b)finalise a Medium Density Housing Code to allow well-designed dual occupancies, town houses, manor homes and terraces to be complying development.

c)facilitate smarter and compact apartments (for example, with less floor space)in well-designed buildings that complement their neighbourhood, with car parking not linked to apartment titles and greater flexibility in parking requirements near transport hubs.

Accelerating infrastructure to support growing communities

More State Infrastructure

xv.) Allocate $1.6 billion for state infrastructure to accelerate housing in priority areas, funded by a $600 million additional reservation from Restart NSW Fund and reallocation of $1 billion between 2018 and 2021/22 from reprioritisation of the existing State capital program towards projects that will maximise housing supply.

Reform to developer infrastructure contributions

xvi.) Expanding Special Infrastructure Contribution (SIC) levy to 10 new areas. Combined with existing areas, this will provide at least $850 million in infrastructure funding over the next 5 years.

xvii.) Providing $309 million over 3 years to fund local infrastructure in high growth council areas that are currently, or about to be, supported by the Local Infrastructure Growth Scheme (LIGS).

xviii.) For current and future LIGS areas, increase the cap on local infrastructure (section 94)contributions by $5000 per year to provide a strong signal about the cost of servicing development, and ensure developers contribute appropriately, and then remove it completely from 1 July 2020.

xix.) For all other areas, immediately remove the cap on local infrastructure contributions to help councils fund local infrastructure directly.

xx.) Put downward pressure on local infrastructure costs by maintaining IPART’s role in assessing local contribution plans where the per dwelling contribution exceeds $30,000 for greenfield areas or $20,000 for urban renewal areas, and releasing new Guidelines to assist IPART and councils

Support local councils to deliver more infrastructure

xxi.) Support up to $500 million of loans from commercial lenders or through the T-Corp ‘Fit for the Future’ loan scheme, by providing up to $31 million to deliver a 50% discount on councils’ interest costs to bring forward the delivery of local infrastructure. To receive the subsidies councils must:

  • meet ‘Fit for the Future’ loan scheme requirements, and
  • will use the loan for infrastructure to enable new housing supply, and
  • have LEPs that contain sufficient feasible residential zoning to meet targets set by District or Regional Plans.
Better coordination of infrastructure, zoning and assessment

xxii.) Provide greater governance, coordination and support to local councils by establishing a team of case managers and an Office of Housing Coordinator within the Department of Planning and Environment to resolve impediments between councils, the State Government and developers. DPE will be allocated $40 million from 2017/18 to 2020/21 to implement.

Giving first home buyers a chance

Streamlined assistance for first home buyers

xxiii.) Abolish Stamp Duty for all first home buyers of both new and existing dwellings up to $650,000, with a stamp duty discount for first home buyers purchasing both new and existing homes up to $800,000.

xxiv.) Limit the $10,000 First Home Owners Grant to new properties worth up to $600,000 and closing the $5000 New Home Grant Scheme. Recipients of first home buyer assistance will be required to live in the home for 12 months from the date of purchase, rather than six months.

xxv.) Abolish insurance duty on lenders’ mortgage insurance (LMI) from 1 July 2017 to reduce the cost of insurance premiums of home buyers.

Address community concerns about investors and partly fund the housing strategy

xxvi.) Increase the foreign investor surcharge from 4% to 8% on Stamp Duty and 0.75% to 2% on Land Tax, with foreign developers to be exempt.

xxvii.) Remove concessions for investors by abolishing the 12 month off-the-plan transfer duty concession for residential purchases by investors from 1 July 2017 consistent with rules for foreign investors

Keeping a focus on results

Clear targets

xxviii.) Adopting a new Premier’s Priority on housing affordability, which comprises a headline target of a xxxxxx completions per financial year statewide and supported by the following supplementary measures:

a)the existing measure on faster housing approvals (90% of approvals determined in 40 days); and

b)a new measure on accelerated land rezoning.

Accountability for results

xxix.) The Premier will establish a ‘Housing Affordability Taskforce’ (HAT) reporting to the Cabinet Infrastructure Committee to drive implementation of the strategy and achievement of the Premier’s Priority.

1 This analysis does not consider of the genuinely homeless, though there are many reasons for genuine homelessness, with cost of housing not the only one – and perhaps not the main one.

2 It has also been suggested that the process of council amalgamations and the reviews being undertaken by the Greater Sydney Commission has slowed the process of approval at council level in the past year. If so, this will show up in a slowdown in new supply over the coming couple of years.

3 It is important that these assumptions and projections be periodically re-visited, accompanied by an assessment of how housing supply is tracking relative to the targets. Changes to the projections, and progress relative to targets, should not be seen as a big drama; it is simply good practice.

4 A common criticism of these measures is that, by allowing first home buyers to pay more for a dwelling, they really amount to a gift to the vendors. This is logically correct, but the important question is the size of the quantitative effect on prices. About 9,000 grants per year are made to first home buyers with a similar number of stamp duty concessions to first home buyers. The cost of this is of the order of $200 million. This compares with turnover in the property market, and purchases of vacant land, totalling about $150 billion. At 0.1 per cent of the total value of transactions, it is something of a stretch to say that the grants and concessions to first home buyers could have inflated prices by very much at an aggregate level. That said, the effects may be more discernible in the segment of the market that first home buyers may be seeking to access.

Top of page