Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Standing Committee on Tax and Revenue
Housing affordability and supply in Australia

ELLIS, Dr Luci, Assistant Governor, Economic, Reserve Bank of Australia [by video link]

JONES, Dr Bradley, Head of Economic Analysis, Reserve Bank of Australia [by video link]

CHAIR: I now declare the hearing open again and welcome witnesses from the Reserve Bank of Australia. Thank you, Luci and Brad, for appearing again. This hearing is a legal proceeding of the parliament. The giving of false or misleading evidence is a serious matter and may be regarded as a contempt of parliament. The evidence given today will be recorded by Hansard and attracts parliamentary privilege. It is important to note that although your evidence is protected by parliamentary privilege this protection cannot be enforced outside Australia. I now invite you to make an opening statement before we proceed to discussion, or are you happy to go straight to discussion, Dr Ellis?

Dr Ellis : Thank you, Chair. I do have an opening statement. Thank you for the opportunity to appear again before this inquiry into a very important topic. Shelter is a fundamental human need, so, naturally, housing affordability is a keenly felt issue. Its importance is highlighted by the number of parliamentary and other inquiries on housing over the past 20 years. Our submission to this inquiry covers a range of aspects of this issue, and in my opening remarks I'll confine myself to some observations about the longer-run context for housing markets and then outline some noteworthy features of the current situation as Australia emerges from the COVID-19 pandemic.

The longer-run context is that financial deregulation in the 1980s and the global decline in inflation in the 1990s combined to reduce nominal interest rates. With lower interest rates, people could service a bigger mortgage with the same repayment. Some of this extra buying capacity financed an increase in the quality of the housing stock. But, because most of the housing stock was already in place, the main effect was to bid up housing prices. This was captured in land prices.

There were some distributional consequences to this. The price increases were greatest in the most desirable regions and neighbourhoods. First-home buyers now had to save a bigger deposit relative to their incomes and some features of the tax system, combined with the possibility of leverage, meant that buying rental properties became especially attractive. The big-picture story, though, was one of increased demand stemming from greater availability and serviceability of finance.

The pandemic and policy responses to it have shifted the balance of supply and demand in some housing markets. Over the past year or so we have again seen strong growth in housing prices. This followed periods of weakness in 2018 and early 2019 and again briefly at the very beginning of the pandemic. Properties coming onto the market are being snapped up very quickly at the moment. Data on housing finance shows that this increase in demand was initially mostly owner-occupiers, but demand from property investors has also increased more recently. Clearly, some purchasers are willing to pay these prices, although others have missed out. Indeed, strong rises in housing prices have been a feature of many advanced economies' recoveries from the pandemic. This is not surprising when interest rates are low, incomes were boosted by fiscal policies and many households are cashed up with savings accumulated during lockdowns.

Another common feature across the country is that many households are wanting larger homes than they did before the pandemic. The combination of the time confined at home during lockdowns and the likely future of more working from home has brought the quality and size of one's home sharply into view. We do not know how lasting this shift in preferences will be. For now, though, renovation activity, commencements for new detached homes and price growth of high-value homes are all at high levels. This demand response to the pandemic relates to the quality and comfort of each individual home and this is a very different margin of adjustment than the number of new homes. In Australia additions to the housing stock have run ahead of population growth for a number of years. This led to higher vacancy rates in the largest capital cities. Vacancy rates in the two largest cities rose further following the onset of the pandemic. With the international border closed, population growth dropped sharply. At the same time the HomeBuilder subsidy and similar state based grants have encouraged more home building, and will for a while yet, so it is hard to see the ingredients of an obvious overall shortage of homes in this combination.

If Australia had not been building enough homes to house its population, you would have seen rents rising across the country. Instead, in recent years the situation has been quite mixed. In the larger cities, where most of the new apartments were built, rents were quite flat. Lower migration and other shifts in population in response to the pandemic and related health measures have mostly exacerbated these trends. Rents have been quite weak in the two largest cities of late, but in some regional areas pressures on rental affordability are reportedly quite acute.

As we outlined in our submission, the supply side of the housing market has inherently limited capacity to adjust to sharp increases in demand. There are some things that could be done to make the flow of new housing supply more responsive. These include speeding up the planning systems to reduce general inefficiencies, improving transport infrastructure to increase the supply of well located land and lowering the cost of new construction. Supply-side changes of this nature would dampen the increase in prices that ensues when demand increases. But they will not undo the fact of rising demand, and they will not prevent price increases entirely. Moreover, housing market policies that add to demand will only amplify any upswing in prices. My colleague Brad Jones and I would be happy to answer any questions you may have about the bank's submission. Thank you.

CHAIR: Thank you, Dr Ellis, and thank you, Dr Jones. We have you down for 1½ hours, but we will obviously try to get you out sooner than that. The reason we asked you to come back was that in your initial evidence before the committee you raised a number of intriguing points, and I was hoping that we could explore some of them. For example, you said when you first appeared before us that you don't increase affordability by giving more money to spend on housing. Does the RBA accept then that, by lowering interest rates, that's what you did?

Dr Ellis : To the extent that low and stable inflation means lower nominal interest rates on an ongoing basis over a run of some decades now, then at some level that is our doing. I guess the response I would have is that the alternative was being a high-inflation country—with much higher inflation than our peers—and that would involve difficulties attracting capital and it would involve economic instability. There are swings and roundabouts, but I know which one of those two configurations I would prefer.

CHAIR: Fair enough. In a sense, you accept—when I say 'you' I mean the Reserve Bank—some responsibility for the housing boom.

Dr Ellis : We recognise that it is a natural consequence of moving from a high-inflation, high-nominal-interest-rate world to a low-interest-rate, low-nominal-inflation world, and in particular where you go from a highly regulated constrained supply of finance to one that is more market based. One of the consequences of that will be that people can service a bigger mortgage on the same repayment, and so with the same income they can service a bigger mortgage and they will therefore tend to bid up house prices. We've been saying this for more than 20 years. We've recognised that this is a consequence of being in a low-inflation environment.

CHAIR: Dr Ellis, you sound like me talking to my daughter. I have been saying these things to her for a number of years. You're quite correct that that's been a theme of your reports from about the early nineties through to today. But serviceability as a percentage of household disposable income has gone up from the mid-eighties, when it was about 30 per cent, to a little over 40 per cent today. Do you think that that demonstrates that housing has actually become more unaffordable in Australia?

Dr Ellis : I'd refer you to graph 5 in our submission. Serviceability has been broadly constant, looking through all the swings and roundabouts, but the serviceability on a mortgage has been broadly constant for about 25 years. Making the distinction between the eighties and subsequent years, there was of course considerable government regulation on the provision of finance, so there were artificial constraints on the provision of finance at that time. Once that particular constraint had been lifted, people were table to borrow up to a point where they could reasonably service it. But then the constraint becomes what you can reasonably service, not an artificial regulatory constraint as we had in the seventies and eighties.

CHAIR: So your thesis is that people's limitation of what they were willing to spend as a percentage of their disposable income was not a matter of personal choice; rather, it was regulatory oversight by regulators that limited how much they could borrow?

Dr Ellis : In the seventies and eighties, yes. Since the deregulation of the eighties, the fundamental difference has been that then inflation fell and nominal interest rates fell, and therefore people could service a bigger mortgage on the same repayment.

CHAIR: So—this is for my benefit—what I'm hearing you say is that, as households became used to a low-inflation environment by the mid-nineties, that gave them comfort in borrowing more for their housing, because they felt that they would be in a low-inflation, low-interest-rate environment for some time?

Dr Ellis : The relevant constraint is the repayment, not the overall size of the mortgage, and so the consequence of having lower interest rates was that for the same serviceability constraint—and I want to make an important nuance there: for the same initial serviceability constraint, that mapped into a larger mortgage and therefore that corresponded to the ability to bid a higher price.

But an important distinction here of course is that people are paying their mortgage down over 25 years or so, and so the accumulated amount of interest that they end up paying is also lower. Most people don't actually stay in the one home for 25 years. But, if you buy a home, you live in it and you pay off your mortgage, in a high-interest-rate world you end up paying more interest overall, but your income probably rose more quickly, and so the burden of those repayments declined more quickly; you inflated your debt away more quickly. In a low-inflation world, you end up paying less interest overall, cumulatively, but the burden of that repayment doesn't inflate away quite as quickly when you've got lower nominal income growth. Our former governor Ian Macfarlane gave a really good speech about this in 2003, where we showed this. The shark-fin graph—the burden of mortgage servicing is different. This is a consequence of being in a low-inflation world. There are distributional consequences and the balance sheet configuration you end up with is different.

CHAIR: But you also had massive asset inflation as well.

Dr Ellis : That is a consequence of that. The level of asset prices that corresponds to that same capacity to service is now different because you're now discounting your asset at a different interest rate.

CHAIR: I'm going to come back to that. The RBA says that's a consequence; later on, I'd like to examine that, because I don't think that that is necessarily an outcome you have to have. That is an outcome we've designed. But can I move on to another part of your submission, where you've examined high rates of population growth as a driver of house prices. During 2020-21 we saw net migration, obviously, fall to pretty close to zero, but we saw house prices skyrocket. Does that mean that we may have overstated population as an issue?

Dr Ellis : I think a really good way to triangulate an understanding of this is to look also at the rental market. I don't think it's possible to look at the housing market without looking at the rental part of that—and Brad may want to chime in with some of the detail on the facts. But what we've seen is that there has been a redistribution of housing demand. The closure of the borders and the cessation in the flow of immigration has primarily affected the two largest cities. You've seen that in vacancy rates and you've seen that in rents. There's always a net flow in both directions. As a consequence of the lockdowns, the people who were in the regions who would ordinarily move to the cities were not doing so, but there were still people moving out to the regions. So we're seeing tight rental affordability and very low vacancy rates in some of the smaller cities and regional areas but in the bigger cities we're seeing high vacancy rates and a very weak rental market, and consequently what we're seeing is this redistribution. As a general proposition, if you don't build enough new homes to house your new people you will end up with the bidding up of house prices, but it is not the only influence on demand, and so it's not the only reason house prices might rise. The combination of very low interest rates, people who are cashed up because they haven't been spending during lockdowns and people re-examining whether maybe they need a home office in their home has contributed to the most recent episode of strong house price growth. And we've seen that in every country; it is not an Australia-specific development.

CHAIR: Brad, do you want to add to that?

Dr Jones : I'd just add that when we think about periods of time as short as those you've just mentioned, say 18 months or so, the nature of the supply response in the housing market means that big movements in prices and rents are largely going to be a function of changes in demand, not supply. The rental market, I think, is a really interesting microcosm. One reason we focus on it is that it allows us to tease out developments that are not always obvious in other parts of the housing market. What we've seen on the population side is a very strong increase in demand, partly population driven, in the regional areas, and that is where we've seen advertised rents go up most notably. In Sydney and Melbourne, in particular in the apartment market, we've seen rents go down, and that is in large part because our borders were closed and the typical influx of foreign students basically dried up. So population growth over short periods of time, combined with some of the other factors Luci mentioned, has had an impact and is one of the variables that we're always monitoring.

CHAIR: To that point, I'm looking at a CoreLogic graph showing annual change in rents for the combined capital cities. What it essentially shows is that after 19 October we saw apartment rents fall by five per cent in 2020 and that so far this year, to October 2021, we've seen them increase by 4.7 per cent. Consistent with your point, we've seen house rentals increase by 9.7 per cent over 2020-21. Does that not indicate that we've got a supply issue that goes beyond population?

Dr Ellis : The first point I'd make is that you're looking at the combined capital city number.


Dr Ellis : As both Brad and I indicated, the two biggest cities are the ones that have lost population as a consequence of the closure of the borders. The rest of the country hasn't. So there's been a redistribution, and you won't see that in a combined capital city number.

CHAIR: So I should ask CoreLogic for Sydney and Melbourne?

Dr Ellis : Yes. Brad, do we actually have that? We've got the advertised rents graph—that's graph 14 in our submission—which shows you that advertised rents in Sydney have basically bottomed out. They've just started to come back up again, and similarly in Melbourne, whereas you can see that they were rising in the other states. They are the advertised rents; they're not the experienced rents. There are a lot of people who got a big rental discount last year and, at best, they've gone back to where they were in 2019, or they may have held on to that discount. If you look at the CPI measure of rents, that has also been relatively subdued in those cities.

CHAIR: I want to make it clear that my Canberra landlord didn't reduce my rent, but he's well-known for being a hard arse.

Dr Ellis : You probably also weren't in a position to make a hardship claim for loss of income.

CHAIR: That's probably true, too—or any hardship claim, what that might be. Also in the initial public hearings you said that you need to look at the housing market holistically. Do you want to give that more flavour?

Dr Ellis : It may take me a while to paint a picture of a holistic system, but I think this is something that the bank has been looking at for a very long time and that I, personally, have been looking at for a very long time. One of the things that makes the housing market both so interesting and so complex is that it connects up to the finance system, which is the mechanism I described. I was listening to APRA in their appearance, and there are a whole bunch of things around lending, lending standards and credit worthiness. There are a bunch of things around land supply, urban structure and architecture—and, of course, migration policy is relevant. There are so many touchpoints to the housing market. If you look at one little bit in isolation, you'll miss the whole picture.

CHAIR: Okay, thank you. Are you worried that the low interest rates are fuelling wealth inequality in Australia?

Dr Ellis : I think that's a legitimate concern. The way I would characterise it is that it's an intergenerational issue, that people whose parents already have homes will be—I mean, the house doesn't disappear as each household no longer exists, so there's a mechanism by which the children of people who own homes can end up relatively easily becoming homeowners. People whose parents rented are going to be in a much more difficult situation to actually get into housing themselves. So there is absolutely an intergenerational component to this, and I would agree that that is something that is worthy of public consideration.

CHAIR: Okay. You're aware that further analysis of Thomas Piketty's data has indicated that most wealth inequality globally, but internally as well in most Western nations, has been driven by whether or not people own their home.

Dr Ellis : That is certainly an important element of it. I think that the issue is whether you're looking at inequality across the whole distribution versus the top one per cent. The top one per cent story has got very little to do with housing. Again, it does tend to be a natural consequence of just the random fate of generations and the random fate of different families. Walter Scheidel's work is also very relevant here. Housing is an important element of it; ultimately though, what we're talking about is land prices. It's not so much about the structure that you've built on that land; it's the fact that all land is not created equal, and the supply of well-located land is in some sense fixed but in many senses dependent on the transport infrastructure and what constitutes 'well located'.

CHAIR: In short, if you're concerned, for all the reasons that we can explore for quite a period of time, with having a society that is equal and has equality, particularly when it comes to wealth, you need to encourage people, or you need to make it possible for people, to be able to purchase their own home?

Dr Ellis : Again, it's the intergenerational issue, isn't it? It's the fact that some people will find it easier to purchase a home than others on account of the socioeconomic position they were born into.

CHAIR: Well, equally, we shouldn't visit the sins of our parents upon our children. And that goes in reverse, too. So we should make sure that any Australian, regardless of where they're born and who they're born to, has, as far as possible, or as far as public policy allows, equal opportunity to own the home in which they live.

Dr Ellis : I can see that being an attractive principle for public policy.

CHAIR: I will grab that statement because I know how hard it is for the Reserve Bank to opine on these things. In your recent Financial Stability Review, you say that house prices could fall sharply if rates were to rise in line with market expectations. What does the RBA consider to be the impact on asset prices when tightening monetary policy?

Dr Ellis : This is something that we do think about. For very obvious reasons, we do not publish housing price forecasts, thinking through the implications. The role of the Financial Stability Review is to think about what could go wrong. It's about thinking about the extreme scenarios. Of course, nowadays, extreme scenarios turn out to be the ones that happen, although, on the output side, as we've seen, we've been pleasantly surprised about how wrong our forecasts have been, particularly on growth and the ability of the labour market to recover from the pandemic. That is one thing that we're mindful of. You referenced the conversation that you had with APRA just before. Those sorts of considerations are one of the reasons you have an interest rate buffer required in a serviceability calculation. Just to be very clear, most people do not borrow all the way up to what the banks will lend them. They don't tend to max out. Some people will, but most people will borrow a bit less than that.

CHAIR: Do you have data on that?

Dr Ellis : Yes. I don't think it's in the most recent Financial Stability Review, but that's something that the financial stability department has certainly published material on in the last couple of years.

CHAIR: Could I ask you to take that on notice?

Dr Ellis : Sure. Hopefully, someone else is taking notes, because I'm bad at taking notes.

CHAIR: The secretariat is very good at following up with witnesses, don't worry.

Dr Ellis : That would be great. Certainly, our colleagues in the financial stability department do have information. The maximum that you're allowed to borrow guides how much people will actually borrow. Most people do not borrow all the way up to the maximum, but some will. Coming back to my opening remarks, that is why, if the maximum you're allowed to borrow increases a lot because interest rates are much lower and you can service a bigger mortgage on the same repayment, that will tend to bring everybody's actual loan size up, even if they're not maxing out the minimum. That need to keep the housing sector and the household sector resilient to an increase in interest rates through the cycle is absolutely one of the reasons why you would have an interest buffer in your mortgage serviceability calculation.

CHAIR: Let me go to something directly before I hand over to the deputy chair. Australia's household sector is one of the most indebted in the world.

Dr Ellis : It has a similar level of debt to comparable countries such as Canada. I don't have the graph in front of me at the moment, but we are not an outlier. Once you allow for the fact that, in Australia, most investment properties, most rental properties, are owned by other households, so the debt that's held against them is household debt rather than corporate debt, where we are is explicable relative to other countries.

CHAIR: That goes to my point, which is that we have one of the most indebted household sectors in the world, but it's leveraged against assets, typically superannuation and housing.

Dr Ellis : It's leveraged almost entirely on housing. You may recall our submission to the Murray inquiry. In our submission to the Murray inquiry in 2014, we opposed leveraging into self-managed super. That wasn't a recommendation that was taken up. The debt is primarily against the housing stock, but an important nuance, when you're looking at cross-country comparisons, is that the household debt in Australia is leveraged against the entire housing stock, whereas in many other countries quite a lot of the rental housing is owned by pension funds, insurance—


Dr Ellis : The corporate sector. So there's a bigger asset base in some sense behind that relative to some other countries. That's just an important nuance. But, yes, our household debt is comparable to and in the same ballpark as other countries, particularly once you make allowances for who owns the housing stock.

CHAIR: A former Reserve Bank analyst writing in the Australian Financial Review said that his expectation is that house prices will fall 20 per cent from their peak. That's going to cause systemic and economic wide issues, isn't it?

Dr Ellis : I think a useful comparator to look at—and we have seen this movie before in Perth—is what happened to housing prices at the end of the mining boom in Perth and Darwin. It was painful. But, again, the housing market is holistic. As my APRA colleagues discussed with you earlier this morning, that's why people do stress tests about big falls in housing prices, to think about how people would cope with that. It's important to understand what the context of that fall in housing prices is. If people still have jobs, that's a very different context to if people don't have jobs and their incomes have fallen and interest rates are higher. So it depends a little bit on what else has happened. But I think a good example of what can happen—not necessarily what will happen but what can happen—is what happened in Perth and Darwin over the past decade.

CHAIR: Thank you, Dr Ellis. We will go to Julie.

Ms OWENS: Thank you for your submission and your evidence. It has been really interesting. I want to go back to the serviceability of loans being flat, because, again, it sounds counterintuitive. When you measure serviceability, is that just of people who have loans?

Dr Ellis : That's correct. I'm just going to read the footnote on graph 5. So the graph that we show in the submission is the percentage of median gross household income needed to service a variable rate mortgage with an 80 per cent loan to valuation ratio for a median-priced dwelling. I want to make clear that the typical first home buyer does not buy the median-priced dwelling; they buy a cheaper dwelling that that. So this is quite indicative. But, if you look at that graph, you'll see that for the last decade it's been pretty flat. It's come up a bit. House prices matter for that calculation and interest rates matter for that calculation. But, on the combination of the two, actually, mortgage serviceability is lower than it was in the previous decade. We've just shown this century's data in that graph, but my recollection of the 1990s is similar. So this is really thinking about a mortgage that's just been taken out. Obviously, if more people have got mortgages, there are more people paying that. The sum total of interest that's being paid by the household sector is reported in the national accounts at an aggregate level, and that, again, is lower than it was back in the 1980s and 1990s.

Ms OWENS: And because APRA adjusts the serviceability requirement, it's kind of a self-fulfilling prophecy, really, isn't it—that serviceability should be flat because it should be regulated to be flat?

Dr Ellis : Well, it's not regulated to be flat, but it's regulated to be resilient to the change in interest rates. The Reserve Bank has publicly welcomed the decision that APRA made to tighten that up.

Ms OWENS: Absolutely. Is there a measurement done in terms of the capacity to service the median mortgage if you are on the median wage?

Dr Ellis : One thing that we know is that a single person on the median wage is not buying the median house and borrowing the median mortgage; they're borrowing something less than that. Certainly, once you go to total household income, of course, there's a whole bunch of people who own their home outright and are not paying any mortgage at all because they're older. This is where distributional things matter a lot. But, yes, we do think about those things, but it's kind of hard. You can look at individual household experience—and in fact my colleague John Simon and a former colleague Tahlee Stone released a discussion paper about this a couple of years ago—and the accumulation of the deposit is actually the binding constraint under those circumstances. The majority of people pay more off their mortgage than they're required to by the loan contract. We've seen that happen quite a bit now. People have been socking away money in offset accounts and redraw accounts during this period. Particularly when you had lockdowns, some people were not spending as much as they ordinarily would, and a lot of that has been going into offset and redraw accounts. They're effectively paying more than they need to, and it's being paid away against their mortgage and their future servicing needs. One important context here is, if and when rates do eventually rise, a lot of people will not actually need to raise their repayment, because they're already paying more than they need to according to their loan contract. That is actually one of the desirable features of the Australian institutional framework that a lot of other countries don't have. Other countries do not have the widespread use of offset and redraw accounts, which are the most tax-effective form of precautionary saving ever invented.

Ms OWENS: You made the statement before that the median wage is not buying the median house. Is there any longitudinal study of where, in the range of house prices, the median wage was over time?

Dr Ellis : I can't think of a study like that off the top of my head. I'm happy to take that one on notice. It may be that someone else has done the analysis, tracing that through. You could use the HILDA survey for that, for example, but I don't think we've done a paper like that.

Ms OWENS: Can we talk about rental? Rental is really interesting in Australia. In Paris you can't get a lease for less than six years. Here you sometimes can't even get six months. I know that real estate agents make their money on the churn, but are there any other factors going on in Australia that prevent us from providing what a lot of people want, which is long-term stable rental in the commercial market?

Dr Ellis : Having lived in Switzerland, I know it is just a very different mindset in the rental market here. People always say they own an investment property. They don't say, 'I am a landlord,' or 'I have a little side business providing house services to my customers.' My landlord in Switzerland actually once said to me that he thought of his tenants as his clients, and I nearly fell over, because I can't imagine any landlord in Australia saying that. So there is a different mindset. It is a different rental structure. We did talk about this in our 2003 submission to the Productivity Commission inquiry on first home ownership and affordability. There does seem to be something a little bit different about the way the rental market works here. One of the things that happens is that you have a lot of control over your home, so, if you want to move back into that house, you can. You can kick the tenant out. It's like a kind of no fault. If you want to move in, you can kick them out again. There is a tendency to have quite short-term leases. That works for a lot of people.

I have spoken about this publicly before. I gave a speech in early 2017, where I showed a graph of how often people move. Renters do move more often than homeowners. For someone who is young and maybe wants to move around for different jobs, that provides some flexibility. The fact that you can rent out your own home, go and work somewhere else and rent there and then come back to your own home instead of selling and buying and then selling and buying again does add some flexibility, but absolutely there are swings and roundabouts to that. As we have said publicly before, but a long time ago, it's not an institutional configuration that gives you long-term rental and long-term rental contracts, but, again, if you think about the countries where you do have stable long-term rental, you also have relatively low rates of home ownership. I know people, friends of mine, who have said: 'I got sick of moving every year, so I finally bought a place. Even though it was maybe not the ideal place, I bought somewhere because I was just sick of the lease expiring and having to move again.'

Ms OWENS: Let's talk about the by-product of housing, because housing actually isn't the point. It's the bath, if you like. The baby is the stability and the financial certainty and all of the rest that it gives you over time. Australia is a really innovative group of people. We're really early adopters. We come up with great ideas, and even though there's this massive group of people out there who are not served by the current housing market, we don't seem to have come up with a whole stack of submarkets of land trusts and co-ops and shared equity and institutional investors. Do you have any idea why that is? Is there something that stops us from finding or creating other markets for the answer?

Dr Ellis : One of the things that I'm very mindful of and that we've spoken about publicly before is the tax treatment of investor property. While it is the same as the tax treatment of other income producing assets, because you can leverage it, the combination of negative gearing with concessional capital gains means that it is very attractive to leverage into investor property as opposed to thinking of yourself as a landlord with a business providing housing services. As a consequence, individual households do it. What happens is that the rental yield on rental properties in Australia is quite low, so the overall level of rents is low relative to prices, but, because of the tax advantages, individual households investing in rental property are willing to accept a relatively low yield for that. It's still attractive relative to other investments, but they're willing to accept a relatively low yield. But, because corporates don't negatively gear and because they're not necessarily planning to sell it later on; it's a long-term investment for them, the concessional capital gains is less relevant to them. I'm not even sure that even works in a corporate environment. My knowledge of corporate tax law is close to zero. Essentially what it means is that a household owning an individual house to rent out is willing to accept a lower yield than a corporate would.

Ahead of our 2003 submission to the Productivity Commission inquiry, Ric Battellino and I actually went on a bit of a study tour to a whole lot of other countries, and we were talking to them about the rental market. One of the questions we got asked was: how on earth does it work having all these individual landlords? This is particularly a story in the US—'How do you get the economies of scale of managing the properties?' And it's like, 'We have real estate agents for that.' They were very mindful of those sorts of costs, and a corporate landlord just doesn't seem to be able to run pockets of apartments here and a couple of houses there. The cost of managing that as a corporate landowner is quite high, so they would want a higher rental yield, and so, quite frankly, the individual property investors would outbid them because they're willing to accept a lower yield.

Ms OWENS: I want to ask a question about data. We've been hearing from a range of witnesses that there's data that just doesn't exist—where the land is, when it's coming online. There's a whole range of data that doesn't exist. Is there data that you want that you can't get?

Dr Ellis : Is there data that I want that I don't have? I think that's true for pretty much every domain. More data is always better. We're not necessarily looking at small area land supply issues. That's well outside our remit, but I can imagine for agencies and other actors in the housing finance system for whom that's really relevant about planning for infrastructure and planning for where to develop things. The population moves that we've seen because of COVID have just been completely different to what we were seeing over the previous ten years. Yes, having information I imagine would be very valuable to people.

Ms OWENS: Thank you.

Ms KEARNEY: I have a couple of questions. We've talked about local levers and concentrated very much on all of our regulatory bodies and frameworks in Australia. To what extent is the Australian housing market linked to international issues? There was, for example, an article on the ABC recently opining about the possibility of inflation rising and therefore interest rates rising in the US and how that will have an impact on banks in Australia that rely on financial institutions in the US. It might put up the cost of borrowing in Australia and have a flow-on effect there. I'm interested in your view about that and how reactive our housing market is to international or global influences like that.

You mention in your submission that other countries—again, this is a global issue—have sought to improve the balance between supply and demand by restricting demand. Are there any particularly successful countries or anyone who has done that very well, and are there lessons that we can learn here?

Dr Ellis : On your first question, I would make a distinction between the multidecade trend and the cycle. The multidecade reduction in interest rates and the move to a low-inflation world was a global phenomenon, so the overall level of inflation was something that was in our interest to follow with. The overall level of real interest rates—inflation adjusted interest rates over multidecades—is more or less driven by global trends. The governor has talked about this in some of his parliamentary testimony in the past. But there can be quite a lot of dispersion in the cycle.

If you look at the history of interest rates in the US versus here in Australia, there is quite a lot of differentiation because our macroeconomic experience through the cycle and our inflation experience are not necessarily the same. Certainly, at our current juncture, I'd refer you to the overview of the statement on monetary policy where we step through this a bit. We're not quite in the same situation as the US in terms of the inflation pressures. We're seeing some of it, but to a lesser extent. There's no necessary linkage or requirement for us to move step for step with what the Fed does. There is obviously a correlation over a longer horizon between interest rates in different economies because there are global factors at play. But inflation rates and interest rates over a shorter horizon do differ, so I wouldn't necessarily say that what happens in the US happens here.

On your other question about examples of countries that have done well constraining demand, I think the jury is out. I wouldn't want to point to any particular country and say, 'This was a good intervention.' One that we are watching to see how it plays out over the next little while is New Zealand. They went from a situation where their tax system was even more generous to investor property than ours. For a long time, they didn't even have a capital gains tax. Their negative gearing was also quite generous. They've tightened up on that a lot recently. It'll take a few years to see the washout from COVID. The population flows that they've experienced as a consequence of closing their borders have been a bit different to ours. They had a big surge of population as all the Kiwis came home, whereas we've had a reduction in our population growth because it was more about the flow of foreign residents coming to become Australian residents being tapped off.

So there are examples, but is there a case I could point to and say, 'This was a really good example where we stopped people demanding so much'? In the US, again, their housing prices have risen at least as much as, if not more than, ours over the past year, including in some of the areas that were previously considered to be quite flexible, like the cities in Texas. But they were doing very much the wrong thing in terms of encouraging people to borrow more prior to the global financial crisis. They have tightened up on that more recently, and housing prices have actually had a bit more of a lid on them. But it's hard to know how much of that was because of the fallout of what happened before versus the consequences of coming back to a more rational level of demand and availability of finance compared to where they were in 2006 and 2007.

Mr THISTLETHWAITE: Thanks, Luci and Brad, for your evidence. Just following on from Ged's question: are you a little concerned that you're starting to see increases in fixed rates here in Australia at the moment on the back of those inflation figures in the United States? Obviously, I think that that's got something to do with the cost of raising capital by the banks in international markets. However, we're starting to see that coming into the Australian market already despite the fact that the RBA said that you don't foresee domestic interest rates increasing for at least 12 months.

Dr Ellis : Thanks for the question. The fixed rates that are rising are for a longer horizon than 12 months, and it's also important to remember that the majority of mortgages in Australia are still priced off variable rates, so they're floating rate debt. The banks tend to price their fixed rate loans off swap rates and so, as swap rates have risen, yes, you've seen that in the new fixed rate loans that are available. People can make a choice. That isn't affecting the people who've already taken out a loan; it's only the new fixed rate loans. People have a choice between variable rate loans and fixed rate loans, so people make choices about which they go for depending on their comfort levels with what could happen with different paths for interest rates.

Certainly, swap rates embody the market participants' expectations about interest rates out several years, and we have a different view based on our view of the future. That's okay, but a one-year fixed rate has not really risen. It's the three-plus-year fixed rates that have risen, and that's making a statement about 2024-25.

Mr THISTLETHWAITE: But, on that point, the four- or five-year fixed rates that people were able to secure even six months ago were remarkably lower than they are at the moment. The commentators are saying that that's starting to price in inflation in the US but also rates rising more quickly than you've forecast. Do you have a view on that?

Dr Ellis : I'm not going to comment about the future path of monetary policy in this context. But, yes, obviously, market rates embody market expectations of what we're going to do.

Mr THISTLETHWAITE: On the serviceability issue, there was some commentary earlier about the potential for asset values to start to fall in Australia. When people talk about serviceability, they're often discussing the ability to make the repayments. But there's also an issue with the asset value, isn't there? If a fall in the asset value affects the loan-to-valuation ratio then a lot of people could be in a lot of trouble. The thing I find is that most Australians don't understand that a bank can ask you to pay off quite a hefty amount in one go if there's a fall in the asset value or even in some circumstances take the home back. Is that something that you think that government, APRA or you should be alerting Australians to that it's not just about paying off the loan; it's the value of the asset as well?

Dr Ellis : Thank you for the question because it does allow me to put on the record that, to my knowledge, mortgages in Australia do not have those kinds of call-in or loan-to-valuation ratio covenants that commercial mortgages would. So the loan covenants that you've just described are sometimes a feature of loans to corporates, but I am not aware of home mortgages having those features and those contract clauses. It is relevant. Thinking about the amount of equity you have in your home is important if you get into trouble and can't service your loan. If you need to sell it and you have no equity, you walk away with no cash on the table. If you do get into trouble, it's better to be able to sell and have something left over so that you can then start again and get back in at a later stage than have no equity and therefore be left with nothing. I would not regard it as being self-interested behaviour on the part of the lender to call in a loan on a household that is servicing the mortgage adequately. If someone was not servicing the loan, then there's an issue of default, and you have to take them to court to get the property. It may be that someone's writing this into the clause, but I would not regard it as being common. In fact, I'm aware of no examples where a home mortgage for your own home involves a 'call in' clause or a 'put up more equity' clause. It's not a margin loan in the way that you would have with shares. Margin loans on shares have that characteristic and have those clauses. I am not aware of mortgages having that clause.

But, in terms of your statement that households don't understand this—implicitly, they do. Most households are paying more off their mortgage than they need to. And, when rates fall, they tend to maintain the repayment. So they end up amortising their debt faster than the contract requires. The majority of those households are ahead of schedule on their mortgage. I would regard that as a positive feature of the Australian market, which is not seen overseas to the same extent, because of the role of offset and redraw accounts. It shows that households do want to have that buffer. They don't want to lose their home.

Dr Jones : It might be worth adding that the bank, in particular the financial stability department, has done a lot of work at a cross-sectional level looking at the range of Australian households that have both high levels of debt and low prepayment buffers. The way that we've tended to think about this is that, if there's a vulnerability, it's households that are very vulnerable to an income shock. That's the channel of vulnerability rather than an LVR ratio at a certain number. In the latest financial stability report, the team there have done a fair amount of work to see how that nexus of borrowers, who have a lot of debt and low prepayment buffers, has changed over the course of the pandemic. The numbers suggest that that share of what we would call the most vulnerable borrowers has actually gone down a little bit over the course of the pandemic. Part of that's because, of course, lower interest rates. If you're thinking about vulnerabilities, it's an income shock hitting those households that have low prepayment buffers and are carrying a lot of debt. That's where we think the vulnerability would be, more so than an LVR at a certain level.

Mr THISTLETHWAITE: I'll move on. I might come back to that because I'm just going to run out of time. On page 14 of your submission, you point out that the way the system works in Australia is that it's made it 'favourable to own additional properties as an investment asset'. Is that an efficient system? If the number of people that can own one property is falling, a system that incentivises people to own additional properties doesn't seem to be the most efficient way to allocate resources.

Dr Ellis : I think that is a conclusion that some way wish to draw. It is the system we have. The question is: can you get from here to there?

Mr THISTLETHWAITE: You go on to say:

Two potential objectives of reform could be to incentivise more efficient utilisation of some of the existing stock of housing…

Can you elaborate on that? What do you mean by that?

Dr Ellis : Well, there are a number of different elements of that. One of the things about the way our system works is that if own your own home at the point you retire, that's basically the thing you need to do to not be in poverty in retirement. Our retirement finance system is predicated on that. The insulation of the own home from the means test and asset test on the age pension is all bound up in that, and the way different age groups are taxed is all bound up in that.

But there is also a certain level of inefficient self-insurance. Many older people are worried that they will not be able to afford aged care if they need it. Not all of them will need it, but a lot of people are holding onto more house than they can manage because they know that they need that asset base in case they end up in aged care. As a consequence of that, you have a lot of people in homes that are perhaps a bit too big for them now that they are struggling to manage, but the alternative is downsizing. Downsizing into appropriate housing isn't always easy. When you downsize to a cheaper, smaller property, you now have assets subject to the asset test and the means test. So people may feel that it is not in their interests to self-insure for aged care in that way. That is one example.

I spoke in reference to another question about how there are swings and roundabouts to our rental market and people have to move very frequently and so forth. The fact that anyone can become a landlord does mean that if you do want to move for a job, but you don't think you are going to stay in that other place permanently, you can rent out your home, go and rent somewhere else for a while, come back and move into your old home. That is a flexibility that many other countries don't have in their system. You end up with some inefficiencies around people feeling like they can sell and buy and sell and buy because of the cost of transacting. Stamp duty is part of that, but it is not the only part. In other countries where stamp duty is lower, you generally find that the transaction cost is just as high, but it goes to the real estate agent and not to the state government.

Mr THISTLETHWAITE: Is there any evidence that more Australians are reaching retirement age while still having a mortgage?

Dr Ellis : There's some evidence that some people are using retirement lump sums to finish off the mortgage, but it's usually small amounts. The Reserve Bank has not, to my knowledge, done any specific analysis of that. It may be that some academic work has looked at that specifically. In general, there is a certain element of people paying off the last bit of mortgage with a superannuation lump sum. That does happen.

I am aware that other regulators don't regard that as being a particularly appropriate provision of finance; lenders shouldn't presume that households will do that. In terms of calculating how much you will lend, you shouldn't say, 'Don't worry; they have all this super and we can just take all their super.' That's not an appropriate way to think about it. We know that through the early 2000s, for example, part of the increase in the debt-to-income ratio was that people in their 50s were more likely to still have a mortgage, but it was usually pretty small, so there was kind of an inframarginal increase in debt.

Mr THISTLETHWAITE: Going back to that earlier point I made about the asset value and serviceability, is that a recent change with the lack of covenants in mortgage based loans such as the one that I was discussing? Does that apply to interest-only loans as well? I was under the impression that—not the most recent round of APRA tightening but the one before—did result in people holding interest-only loans having to look to pay off some lump sums because the value of the assets had fallen?

Dr Ellis : There are two things going on here. In answer to the first bit of the question, I could be wrong, but I am not aware that it has ever been a feature of the Australian mortgage market that banks could just call in a loan that was still being serviced according to the contract. I do not believe that it would be in their interests—their own financial interests—to call such a loan.

In terms of the phenomenon you're describing, when APRA tightened certain criteria on interest-only loans, the result was that the interest rates on interest-only loans in the back book also increased. So some people decided that maybe they didn't want to pay that higher interest rate after all and they wanted to refinance. It wasn't that they were required to stump up some more capital in the existing loan; it was just that in order to refinance they were now facing tighter lending conditions, and it was therefore in their interest to stump up some more capital. In order to get a principal and interest loan, they needed to qualify for it. Obviously, if the value of the property had fallen, then you wouldn't get as big a loan for that, so you'd end up having to put in more capital. But that's really the consequence of someone's decision to refinance rather than pay the higher interest rate that was now on offer. It was not the case that people who were continuing with their existing interest-only loan were being asked to stump up some more capital. It was only because, in order to refinance, you were now facing tighter lending conditions than those that prevailed when you first took out the loan.

Mr THISTLETHWAITE: Great. Thank you.

CHAIR: Thank you, Matt. Dr Ellis, earlier in your evidence you said that a low inflationary environment results in asset prices increasing faster than CPI. Have I misstated your view or the view of the bank?

Dr Ellis : The way it works is that when inflation is lower and nominal interest rates are lower you are discounting the benefits or the income flow of an asset with a lower discount rate, and so the equilibrium level of asset prices is different. It's now higher. And there's a process of transition from the old equilibrium to the new equilibrium during which asset prices will rise more quickly. But it's a move from one steady state to another. It's not a permanent change in the growth rate.

CHAIR: But wouldn't it be good if we could have the best of both worlds—low inflation and asset price inflation at the same level?

Dr Ellis : In the long run, that's actually what you would expect, but when you have rates ratchet down—the big narrative there was the big move in interest rates because of lower inflation and financial deregulation, and that process ended in about 2005. There's been another ratchet down in the equilibrium real interest rate globally, what people sometimes call R-star. There are a number of different possible reasons why that has occurred, but it's a global phenomenon. So you've had another transition because of that, and then of course, at the moment, because of COVID and the policy response to COVID, which was quite properly put in place, rates are lower again. You have periods of ratchet. If you actually look at house price growth in Australia through this period, there were a number of periods, including the very end of 2003 into 2004 and 2005, where actually housing price growth didn't outpace income growth. It's been kind of like a stair step of these equilibrium shifts, but what we're describing is an equilibrium shift to a new equilibrium level of prices because you are now discounting at a lower interest rate.

CHAIR: You mentioned New Zealand's policy response earlier. Did I miss that you also talked about some of their policies being designed to allow for much faster and greater supply?

Dr Ellis : They have had even more issues—I'm not an expert in the supply side piece of the New Zealand policy response. I know they recognise it is an issue. I think one of the reasons that they recognise it is an issue is that they face even more of a geographical constraint than us. We have 45 per cent of our population—or is it 45 per cent of our urban population? I forget which now. We've got a lot of our population concentrated in two cities, and one of those cities is basically encircled on four sides by ocean, mountains and national park, so there is inherent supply. You can't just keep going out, whereas in Melbourne you can. I'm actually in the Melbourne office at the moment, and there is a lot more of Melbourne in the south-west side moving out to Geelong than there used to be when I grew up here. So your ability to expand depends on your geography. In New Zealand, a very large fraction of their population is in one city, not two, and that city is really hemmed in. If you look at a map of Auckland, it's extremely hemmed in.

You see a similar phenomenon in other countries where there is really only one big city and it's hemmed in. For example, Santiago in Chile is hemmed in by mountains. The Andes are on one side, and there are foothills of the Andes that you can't easily build on on the other. So they've gone up a lot. They've really densified because they have no other choice. But housing in Santiago is very expensive relative to the rest of Chile. If you talk to our Chilean counterparts, they will tell you they face a huge housing affordability issue. There's no alternative to living in Santiago if you want to live in a big city and have all the jobs and amenities that are involved in living in a big city, but there's nowhere further to go. If you've ever driven around Santiago, it's just crazy.

CHAIR: I would love to have Chilean counterparts, but I don't, so I'll take your word for it. I guess my underlying and essential point is about New Zealand's response recently. I think that three years ago they announced that they were going to allow people to build three-storey buildings in Auckland, which is what you referred to, and they recently announced that policy for the entire country, but it was based on the fact that they have a region called Selwyn, and Selwyn about 11 years ago decided that it would actually get rid of most of their planning controls—or, to put it differently, have very light-touch planning controls. What they saw was that, as demand came into the market, most of that demand got channelled into increasing supply and very little into price increase, whereas in Auckland, where the planning controls stayed in place, most of the demand got channelled into higher prices because there wasn't a supply response. That's pretty consistent with market economics, isn't it?

Dr Ellis : The evidence on this really depends on what your geographic constraints are. So the question is if that's been the case in that district. There are plenty of cities that are cheap to buy in in various countries because people are moving out and there are plenty of empty properties and low demand, and that's how you get cheap property. If you have people moving into a place, so you need to build more housing to accommodate that additional population, realistically whether you can manage to do that is related to how you configure your supply. Your zoning laws and your planning permission and so on do matter for how much of that shows up as additional housing as opposed to additional prices. But the lesson I draw from looking at more than one district in New Zealand and across the world is that geographical constrains are more fundamental. You just said they can now build three-storey buildings. Sydney has much more liberal laws than allowing you to build three storeys, and where there are restrictions quite a lot of them are about safety. There are issues about spacing between different houses for fire reasons. There are issues about what materials you use, again for fire safety and insulation reasons. There are other issues: 'Yes, you can build a three- or four-storey home, but only if it doesn't overlook your neighbour's garden.' That, I think, is a reasonable accommodation for us all to live in society. So what you've described as the relaxation in Auckland still sounds very restrictive relative to what we have in the cities in Australia.

Another good example, I think, that would be more comparable is that back in the nineties in Victoria they instituted dual occupancy, so, as long as you could meet the Building Code, you could basically put two dwellings on a suburban block. I can tell you that, in the part of Melbourne that I grew up in, most of the houses that were there when I was growing up are not there anymore. There are now two or even three smaller homes. So there's been a considerable densification of housing in those areas. It is not obvious that that made a big difference to the level of housing prices. Melbourne is cheaper than Sydney, and that was a big liberalisation, but it's not obvious that it made a huge difference, because there were common demand factors as well.

Dr Jones : The New Zealand piece is interesting because it connects back to a point we tried to convey in the submission, which is that we really need to be looking at both supply and demand here in terms of solutions. My understanding, at least, of what New Zealand has tried to do is, in addition to the partial relaxation of some planning laws that you referred to, they've set up a new infrastructure fund to help accelerate new housing supply. On top of that, they're basically phasing out their interest deductibility as it pertains to investment properties for established dwellings, but they're keeping it for new dwellings. They're looking at this quite holistically, I think, and that's something that will be of interest for some policymakers here. It's too early to say how this all plays out, of course.

CHAIR: I'm just going to pick you up on that. One of my concerns about the RBA's submission is that you talk about demand and supply. Just to be clear: when we talk about reducing demand, we are talking about people who are trying to buy a home to live in.

Dr Ellis : I think what we are pointing out is the difficult trade-off. As we said in our 2015 submission, there are no examples internationally of large falls in nominal housing prices that have occurred other than through a significant reduction in capacity to pay, such as a recession and high unemployment. You can do it, but it's not something you may want to do. There's no example in Australia or internationally where supply expansion on its own generated house-price declines of a similar order of magnitude to the kinds of cycles in house prices that we've seen in Australia in recent years. Yes, you could achieve a big decline in housing prices, but you may not wish to do so, precisely because of the trade-off that you're referring to. I think we're in furious agreement here.

CHAIR: I get enough economics to know that price is a function of the interaction between supply and demand. We have a choice here between expanding supply, reducing demand or doing a bit of both. The implication of reducing demand—and I think we need to be clear about this—is that we're telling some Australians that we don't want them to own a home because that's how we are going to moderate prices.

Dr Ellis : The way we characterise it in our various submissions is really more a cautionary note to say, 'You don't improve affordability by giving people even more money to then bid up prices.' In thinking about things like the various subsidies that we've seen recently, some of those did add to demand. That is one of the factors that has been relevant. The Reserve Bank is not in the business of saying, 'Wouldn't it be nice if demand were lower?' That is not what we're saying. We've been quite clear. We're saying, 'It would be nice if everybody could afford a really big house and not have to pay very much for it, but we don't think that is feasible, given the availability of finance, the geographic constraints, the cost of construction and the fact that well-located land is actually in limited supply.' You can build a really cheap house somewhere you don't want to live, but, in terms of thinking about where the jobs are and where the amenities are, in the end—one of the things I want to round off on, because I'm conscious we're coming up to time—

CHAIR: I was going to ask whether you were pressed. Could we extend the hearing for—

Dr Ellis : I have a four o'clock flight, so I'm somewhat limited. Tullamarine is a reasonable distance from the Melbourne office, and there are no later flights, so I can't afford to miss it.

CHAIR: In that case, we'll extend it for another 15, and I'll start to interrupt more often. My apologies for that up front.

Dr Ellis : I think the point is that the supply we are talking about is the supply of well-located land.

CHAIR: Yes, and I'm not talking about that. The Reserve Bank keeps complaining, 'Wouldn't it be nice if everyone lived in a mansion by the harbour.' I don't think that is what this committee or parliament is talking about. Indeed, not everyone wants to live in a big house. Every second weekend when I mow the lawns I think, 'Wouldn't it be nice to live in an apartment?' These are choices that we make, but these are choices that are best made by us personally, not by some centrally planned authority. Wouldn't that be fair?

Dr Ellis : I would concur—and I have stated publicly in past speeches—that it would be good to have a range of housing options for different people's life stages and needs. When I say the supply of well-located land, I don't necessarily mean a mansion by the harbour; I mean homes that are close to jobs and homes that are close to the amenities of a big city. You can go and—

CHAIR: You mentioned some cities in Texas in the United States that have very light planning laws and are not centrally planned. They are actually organically designed cities that are much, much better in ensuring people live closer to where they work and having well-located land than all the centrally planned cities throughout the world. Wouldn't that be fair?

Dr Ellis : I would make a distinction between cities that can grow on all sides because they're inland and cities that cannot, because they have coasts. In many respects, the restrictions with building codes and so on are not necessarily because people are trying to design a particular kind of city or enforce a particular kind of building. Some of these places—

CHAIR: We've had evidence from the Planning Institute of Australia—it's their evidence, and they are fully entitled to it—which is that they don't just want to tell you where to build; they want to tell you what to build and then they want to tell you what it looks like and how it works inside. Indeed, they went as far as saying that Australians have too many rooms in their houses and we need to design smaller houses.

Dr Ellis : I think that there are certainly people who would benefit from being able to access a smaller home, because a smaller home that is cheaper to build would be more affordable. If we're continually building big five-bedroom places, that's not going to be accessible to people who are on low incomes.

CHAIR: But we're doing that currently, under some of the most centrally planned housing markets in the world. Why don't we try doing what we do in every other marketplace, which is let suppliers meet buyers, and they can decide? And planners will get it wrong. They'll get it absolutely wrong. Docklands is an example of where they got it wrong. Doesn't that make sense?

Dr Ellis : I'll make a couple of points in relation to that. House prices in the cities of Texas have actually risen in the last year or so at least as quickly as all of the coastal cities in the US. So it's not—

CHAIR: I remind you of your earlier point: we shouldn't complain about longer-term price rises with—

Dr Ellis : Correct. So, yes, absolutely, if we were living in inland cities—all the things you are talking about with the restrictions on how you build a home are in the cost of the dwelling. They would be captured in construction costs, and I refer you to some of the work that's been done by colleagues on the effect of zoning. And that gets captured in—to the extent that that is properly measured, and there are many different views on that—the cost of the land is what we're talking about. It is the price of well-located land—which is more or less fixed, unless you change transport infrastructure or unless you rebuild Warrnambool to be a city of a million people, the supply of well-located land is relatively fixed in the short-term. If the Planning Institute is saying we want you to do all of these things that are increasing the cost of construction, that will be captured in the cost of construction.

CHAIR: No, no. I guess what I'm saying is that there is a mindset that says, 'We want a million people to live here. So then we just go about rezoning land. 'Build it and we assume that the people will come.' Why don't we just leave our housing market in the same way we leave every other market in Australia? That's not to say unregulated, but it is to say that the regulation should allow buyers and sellers to meet in the marketplace and decide what they want, where they want it and how they want it?

Dr Ellis : It's an interesting proposition. The important point I would make about the housing market is that houses are not mandarins. The land underneath them is very important. You don't buy mandarins with a 25-year mortgage. Accumulated interest is not a consideration when you're buying mandarins, and the mandarins you buy this week are new mandarins and not the mandarins that you didn't buy last week. The mandarins from last week are no longer available for sale because someone's eaten them—probably my children. The laws of economics do depend on the fundamentals. Land is—

CHAIR: I'll have to stop you there, because we've heard this mandarin, and occasionally banana, argument. I guess it's 'Choose your fruit'—and you've revealed your bias in your children's fruit-eating habits.

Dr Ellis : I was referencing the fact that you used mandarins in your previous discussion with APRA.

CHAIR: Fair enough. But the housing market is not the most complex market in the world, is it? There are markets that are far more complex than this.

Dr Ellis : It has complexities. We're talking about the land. We're talking about structures that stay there after you've built them. We're talking about the role of finance and about differential margins of the quality of the property versus the number of properties. All of those things add complexity. It's a very well understood market. Our colleagues in the Bank of England will say that, yes, there's an element where supply restrictions do matter. We've also said this publicly in the past. Supply restrictions do matter, and they certainly matter in terms of the balance between price growth and additional properties when you're facing an increase in demand in the form of increased population in a particular area. None of that is disputed. The primary thing that's going on in this is the interaction of incomes and interest rates that happen over many decades, and that's also not in dispute.

If you talk to our colleagues at the Bank of England, they certainly say that the increase in prices in the UK and in many other markets is almost entirely explicable over those decades by the decline in nominal interest rates. Within that, there will be cities that are cheap and cities that are expensive. You mentioned Texas. But the cities where the house price to income ratio is the lowest are not in Texas. They're in Ohio—in Cleveland, Cincinnati, Toledo, Dayton. They're not in Texas. Actually, that's another example of where, yes, supply does matter, but what really matters is if you're in a local economy that's doing badly, where incomes are low and where people are moving away, you can buy a cheap house.

CHAIR: Sure. But there's also the example of Tokyo, where, after the 1989 crash, they instituted a massive fiscal stimulus that included what we call social housing here. That didn't shift homelessness, home ownership and housing prices. What did was planning reform by the Tokyo government in 2004. That was highlighted in the Economist only last year, wasn't it?

Dr Ellis : The Tokyo market is a very different market—

CHAIR: But so is the UK market, compared to Australia.

Dr Ellis : There is real difference—again, two things are quite important. One is that we have a growing population, and they have a declining population. That is important in terms of thinking about the balance of supply and demand, and that shouldn't be underestimated. The other point is that almost everybody in Tokyo proper would be living in apartments, so the land component of that is quite small. It's just a very different configuration.

CHAIR: Let me quote former Governor Stevens, who wrote a report to the New South Wales government. He said: 'We need to have the supply side able to respond to demand in a more elastic way. 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.' Do you disagree with that proposition?

Dr Ellis : I agree with that proposition. I just think that there's a limit. That will be helpful, but it won't undo everything. I would refer you back to my opening statement. There is a benefit to reducing general inefficiencies in the planning process, but I would also highlight that that's not a panacea that will undo everything that's happened over the last 40 years of lower interest rates.

CHAIR: Sure. The planning reform in Tokyo, for example—and Tokyo is a city continuing to have population growth as people come out of country towns, or what they call country towns!—moderated prices. It didn't reduce them. But, in 2018, the RBA issued a paper that basically said: 'Supply for housing development had a major impact on physical input costs of between 42 per cent and 73 per cent for Australia's major capital cities—73 per cent in Sydney and 54 per cent in Perth.' It further went on to say: 'The capital gains tax concession and negative gearing have an impact of prices of between one per cent and four per cent, depending on what markets you are in.' Shouldn't we be focused on the factors that are driving up prices by 10 and 15 times what the tax side is doing?

Dr Ellis : You've asked me about that paper before. Any one research paper is one perspective. It's not the whole story. It's an estimate based on a particular methodology and a particular model. I'm aware of other research that's been released since by others. There are a number of cities in Australia you would expect to see a result from because there has been a significant liberalisation in planning laws across the eastern seaboard. I don't dispute that there were—

CHAIR: Has there?

Dr Ellis : Yes, there has. In Brisbane, there has been a number of changes. This is not something that I've tracked in detail, but there have been a number of planning reforms in New South Wales, which is one of the reasons why we have so many more apartments than we used to—double 15 years ago.

CHAIR: The last time you appeared before us, you made that statement about apartments in New South Wales. So I went and dug into the figures about that. It turns out that the vast majority of those apartments in the inner-city part occurred in Green Square. Green Square was developed under Frank Sartor when he was planning minister and he removed planning control from the City of Sydney and basically approved those apartments by fiat. If you walk into any corner store in Sydney and talk to the local builder he'll tell you that planning laws in New South Wales over the last 10 to 15 years have become more constricted, more laborious, more complicated and more expensive to administer. We heard from NHFIC this morning. Frankly, no-one is actually certain what the rules are any more, much less what the charges are.

Dr Ellis : We do not hear the same story from our liaison contacts. There has been some liberalisation of New South Wales law, but that is not something that is within our remit. We did make this point in the submission—part of the issue is not that the rules are restrictive but that the rules are uncertain. It may suit some builders, if they've got the balance sheet strength to sit around and fight it in the land and environment court, to have uncertain rules. Whereas other people may not otherwise be able to compete with you because they don't have the balance sheet strength and don't know what the rules are because it's uncertain until you go to court. It may well be that what would be more efficient and useful, as I said, would be to avoid general inefficiencies in the planning system. I'm not saying that the planning system is perfect. The entire message that we've been saying is that that would help, but it's not a panacea.

CHAIR: I can very strongly argue with you on that point. The evidence before this committee from builders is incontrovertible on the point that planning regulations in New South Wales in particular, but in all the states—Queensland's included, Victoria as well—have become more restrictive over the last decade and a half. In fact, we have heard evidence on your point that builders are going directly to the Land and Environment Court because they cannot sustain the time and effort it takes for local governments in New South Wales to continuously change their rulings on different developments. That will have to be my last question because I understand that you have a flight to catch.

Dr Ellis : That doesn't correspond with my understanding, but I am not a planning expert. My impression was that there has been some liberalisation at the state level and also—

CHAIR: I'd hate to disavow you of your impression.

Dr Ellis : We can come back to you on our assessment of the apartment supply because we did do some work on apartment supply for the submission. But, again, it's not my impression that a majority of the apartments that were built were in Green Square.

CHAIR: This was in Sydney.

Dr Ellis : Yes. There has been a lot of infill development right through the middle ring of Sydney as well, in a way that you didn't see in Melbourne and Brisbane. In Melbourne and Brisbane the apartment boom was primarily in the very inner areas.


Dr Ellis : And again, that was—

CHAIR: But that's consistent—

Dr Ellis : because you can't build apartments in large parts of Melbourne. It's all in the very inner bit.

CHAIR: with your research—'your' being the Reserve Bank's research—that shows that people have found it incredibly difficult to build apartments close to the city, which has been the driver of price increases. There's also the moving dynamic chain study from Finland, which shows that, if you restrict supply close to the city, it has knock-on effects that are mostly felt by low-income people seeking housing. And the reverse was true as well. I'm happy to supply that study to you.

Dr Ellis : I was aware that you were interested in that study, and I did feel a bit nostalgic looking into some of the filtering models because I went back to that literature for the first time in a long time. The filtering models have some validity. You can find some papers that show that it works, and you can find other papers—from Canada, for example, which I think are more relevant to the Australian environment—that show that it doesn't. It really depends on the urban form that you've got. The original US literature on filtering had the presupposition that the rich people were building out on the urban fringe and building big houses and that the poor people were living in the inner-city downtown. Of course, that's not how it works in Australia and in most other countries. Again, that's an artefact of their tax system and their land tax system.

CHAIR: I don't think that's what that study shows. But, Dr Ellis, I really don't want to be accused of having made you miss your flight. I thank you for your evidence today. Was there anything else that you'd like to say before I wind up?

Dr Ellis : We've made so many submissions over the years that I think that there's probably enough to chew on there and in our opening statement, so thank you for having us back.

CHAIR: Thank you, Dr Ellis, and thank you, Dr Jones. If you've been asked to provide any additional information please forward it by 29 November. You will be sent a copy of the transcript and can make any corrections to errors you find in there. I'd like to thank members and witnesses for their time today and I declare this public hearing closed.

Committee adjourned at 13:18