Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Economics References Committee
02/10/2014
Affordable housing

EDEY, Dr Malcolm, Assistant Governor, Financial System, Reserve Bank of Australia

ELLIS, Dr Luci, Head, Financial Stability, Reserve Bank of Australia

Committee met at 08:32

CHAIR ( Senator Dastyari ): I declare open this hearing of the inquiry by the Senate Economics References Committee into affordable housing. The Senate referred this inquiry to the committee on 12 December 2013. The committee is due to report by 27 November 2014. These are public proceedings, although the committee may determine or agree to a request to have evidence heard in camera. I remind all witnesses that in giving evidence to the committee they are protected by parliamentary privilege. It is unlawful for anyone to threaten or disadvantage a witness on account of evidence given to a committee and such action may be treated by the Senate as a contempt. It is also contempt to give false or misleading evidence to a committee. If a witness objects to answering a question the witness should state the ground upon which the objection is taken and the committee will determine whether it will insist on an answer, having regard to the ground which is claimed. If the committee determines to insist on an answer a witness may request that the answer be given in camera. Such a request may also be made at any other time. I remind members of the committee that the Senate has resolved that Commonwealth officers shall not be asked to give opinions on matters of policy and shall be given reasonable opportunity to refer questions asked of the officer to a superior officer or to a minister. This resolution prohibits only questions asking for opinions on matters of policy and does not preclude questions asking for explanations of policies or factual questions about when and how policies were adopted.\

I welcome and thank officers of the Reserve Bank of Australia, who at short notice have taken the time to appear before the committee today. You have made yourselves available at incredibly short notice. I think it was the end of last week that we notified you and requested your attendance. I know you have had to move around quite a few things to be here with us today. The committee appreciates that.

I also point out that there are quite a few questions from different senators. We had a private conversation to this effect amongst committee members. While we have allocated an hour for this session there is a possibility that we will run a little bit over that. I thank you for making it clear before we started that you are able to stay back if we do run over time to give senators ample opportunity to ask their questions. I invite you to make an opening statement.

Dr Edey : Thank you for the opportunity to discuss the Reserve Bank's views on the important topic of housing affordability. This is a subject on which the Reserve Bank has published a good deal of analysis over the years. I will refer in particular to our submission to this inquiry in February and also to our most recent Financial Stability Review, released last week, which covers our assessment of current developments. My colleague Luci Ellis was the author of the February submission and she will be happy to answer any detailed questions on the contents of that in due course, but in these opening remarks I will be drawing on some key points in the two sources that I have just mentioned.

There are a number of things people might have in mind when they use the term 'affordability'. Affordability measures will differ depending upon whether we are talking about owners or renters and also on whether we are interested in some specific market segment, such as first home buyers or low-income households. For owner-occupiers perceptions of affordability will depend on many things, including price, household income, the cost and availability of finance and a whole host of factors affecting the needs and aspirations of the buyer. In any analysis it is going to be necessary to make use of summary measures, and we have to acknowledge that these will inevitably gloss over the diversity of the experience across different types of households.

Much of the public discussion on affordability is focused on home purchasers. A useful summary measure of that is the repayment on a typical new housing loan expressed as a ratio to disposable income. On that metric housing affordability in Australia has fluctuated around a broadly stable average over the past three decades, with average repayments varying between about 20 and about 30 per cent of disposable incomes. These data are reported in our submission. Currently that figure is a bit below average, but it has been rising in the recent period since the publication of the submission as the housing market has gathered momentum. Over the same 30-year period the ratio of housing prices to incomes has increased substantially. These developments in price and affordability have been interrelated. Housing prices received a substantial boost from the combined effects of disinflation and financial deregulation, which lowered the cost and increased the availability of finance. Much of the increase in the price-to-income ratio was concentrated over the 10-year period to the end of 2003, when the ratio increased by around two-thirds.

It is reasonable to think of this as a transitional impact on housing prices that will not recur. Both the shift to low inflation and the comprehensive deregulation of the financial system are things that happen only once. In broad terms, the adjustment of the housing market to this new environment seems to have been completed by around the middle of last decade. Since then, the ratio of housing prices to incomes has been relatively stable but, for reasons I have already alluded to and which I will come back to, it has been rising recently and is now at the upper end of its recent range.

To summarise these key stylised facts that I have just gone through, the ratio of housing prices to incomes now is at the top of its historical range but over time this has been more than offset by falls in financing costs, so that the typical repayment burden as a share of income is currently not particularly high. This of course does not rule out affordability problems in particular market segments or for particular types of households.

Our submission made the point that there is no shortage of housing finance in Australia. Housing loan interest rates are currently as low as they have been in a generation and households are not artificially constrained from borrowing as much as they can reasonably be expected to repay. I have already made the point that perceptions of affordability will differ across different types of households, but if there is a perceived affordability problem in Australia it is not due to a lack of finance.

An important theme of our submission was that housing prices and affordability are affected by the interaction of both supply and demand factors. The factors that I have mentioned so far—household incomes, the cost and availability of finance—primarily affect the demand side of the market. In the short to medium term it is those sorts of factors that will tend to have the predominant influence on housing price movements. The reason for that is that the supply side of the market is dominated by a large existing stock of dwellings and new supply takes time to come on stream. In the longer term, however, supply factors are critically important. It is the supply response that determines the extent to which additional demand results in higher prices over time. Our submission highlights that Australia faces a number of longstanding challenges in this area. These include regulatory and zoning constraints, inherent geographical barriers and the cost structure of the building industry. There are also obstacles to affordable housing created by Australia's unusually low density urban structure, although this is gradually changing.

Our submission does not seek to offer policy prescriptions for improving the supply-side response. The general point I would make is that we cannot improve housing affordability simply by adding to demand. Targeted assistance can certainly help particular groups, such as first home buyers, but without a supply side response any generalised increase in demand will just be capitalised into prices. An important emphasis in our submission is that due attention needs to be given to supply side factors in any policy response to perceived problems of affordability.

Let me turn to some current developments. I have already mentioned the recent strength of housing prices in Australia and I expect that the committee will be interested in the bank's current assessment of this, particularly in light of the comments that we made in our Financial Stability Review last week. Over the past couple of years, housing prices in Australia have been rising strongly. Some of this has perhaps represented an element of catch up after some earlier weakness, nonetheless, prices have continued to rise significantly faster than incomes. This has been associated with strong growth in investor activity.

I will cite a few key facts and figures. National housing prices have been rising at a rate of about 10 per cent over the past year and about 15 per cent in Sydney. The rate of growth of investor finance is significantly outpacing the growth in household incomes. Loans to investors currently account for close to 50 per cent of new housing loan approvals. Investor activity has been particularly concentrated in New South Wales and Victoria. In New South Wales, investor loan approvals have increased by about 90 per cent over the past two years. It is against this background that the bank said in its Financial Stability Review last week that the composition of housing and mortgage market activity is becoming unbalanced. The review also indicated that we are discussing with APRA steps that might be taken to reinforce sound lending practices, particularly for investor finance although not necessarily limited to that.

I emphasise that the banks in Australia are resilient and mortgage lending in this country has historically been relatively safe. APRA has, however, noted a trend to riskier lending practices and over the past couple of years has been seeking to temper these through its supervisory activities. There are also broader concerns with the macroeconomic risks associated with excessive speculative activity, since this activity can amplify the property price cycle and increase risks to households. Our discussions with APRA and other agencies on these matters are ongoing and there will be more to say about them in due course. I will finish my statement there. My colleague and I are happy to take any questions that you might have.

CHAIR: There are a fair few questions from senators, so I will get started with where I think the committee would like to go.

It is fair to say that there has been a fair bit of discussion in the media, and quite publicly, following the release of the financial stability review last week, about the use of macroprudential tools to address the concerns that were raised within that review. That is a fair statement of fact. Dr Edey and Dr Ellis, you are both nodding in agreement. Let's take a step back and asks ourselves: what are the tools that are being considered?

Dr Edey : The first thing to say about that is that these discussions are ongoing. So it is hard to be specific at this stage.

CHAIR: What is the talk?

Dr Edey : I will come to that. The second thing to say is that these are APRA's tools; they are not the Reserve Bank's tools. One reason I need to be careful about what I say is that I cannot pre-empt things that APRA might announce. What we have said so far is that we think this is a significant enough concern that we need to involve ourselves in talking with APRA about things that might be done. We are engaged in those discussions at the moment.

CHAIR: I think it is fair to say that the point that you are making, Dr Edey, is that rather than going around in circles you are saying that no decision has been made, but this is something that is likely to be discussed. Is that fair?

Dr Edey : That is very fair. These things are being discussed. Something will be done. I am confident of that.

CHAIR: What is being discussed?

Dr Edey : In the public domain, one of the things that people have brought up, which I think is not a realistic option, is loan to valuation ratios.

CHAIR: You are ruling out LVRs?

Dr Edey : I am not ruling anything in or out at the moment, but I am saying that that is unlikely to be one of the measures that we would focus on, because it is targeted at the wrong segment of the market.

CHAIR: So LVRs are probably out?

Dr Edey : I would say that those are unlikely to be in the tool kit.

CHAIR: Is the word 'unlikely' the word that you are allowed to say, Dr Edey? Is that the word that you are using: 'unlikely'?

Dr Edey : I am saying that it is unlikely. I am being careful not to rule anything in or out, because I am not a spokesman for APRA and these things have to be discussed and properly determined by the proper authority.

CHAIR: What about setting DTIs?

Dr Edey : The additional thing that I wanted to say is that the tools we are talking about will need to be carefully targeted. When we have talked about the nature of the problem we have said that we think there is an imbalance in the form of excessive activity by investors in the market. That is out of proportion with their normal share in the housing market.

I cited some facts and figures on that. Investor finance is roughly half of new housing finance at the moment. That is out of proportion with its normal part in the market. One thing that I can definitely say is that the measures that are implemented will be targeted at the imbalance.

CHAIR: So you are saying that the measures will be targeted at the investor component of the housing market?

Dr Edey : They will be targeted at high-risk and problematic areas of housing market activity. And the prime candidate for that is investor housing activity. I said in my opening remarks that it will not necessarily be limited to that, but that is the prime candidate.

CHAIR: So where are we in terms of the process? You mentioned APRA before. Just let me get my understanding right: the process is that you are in discussions with APRA; you are in the process of determining what the tool kit we are going to use will be; and then you will make a decision about whether or not you are going to implement that tool kit? Is that fair?

Dr Edey : That is not entirely accurate. As I said before, these are APRA's tools so the final decision rests with APRA as to exactly what they do with the supervisory tools that are at their disposal.

I think the other thing I want to say is that some of the public commentary gives the impression that suddenly we have lurched in a new direction here, that we are doing things that have never been done before and that it is quite a different approach to policy. I want to emphasise the continuity with what has already been done. We have been talking about risks to housing for quite some time—for at least a year. We talk to APRA about these things on an ongoing basis and APRA has been doing things in the course of the past year already to try to bring about better risk management practices in the lenders.

CHAIR: But that is not quite fair in the sense that there is nothing wrong with an organisation of the size and nature of the Reserve Bank adopting a new position and changing its position on this front. But you only have to go back to reports you have done previously—and Dr Ellis will probably speak to this—which have, reasonably, highlighted the concerns about using macroprudential tools. To go further, the Governor of the Reserve Bank, Glenn Stevens—and these are his words, not mine—referred to them just in August as, I think, an 'international fad'. So I just want to put it in context. Previously the Reserve Bank has expressed some concern about using macroprudential tools, and the concerns that you have raised have been in the context that these, by their nature, can be fairly blunt instruments and difficult to target and that there will be unintended consequences.

Dr Edey : I would welcome the opportunity to put our thinking on macroprudential tools into some context if you would allow me to do that.

CHAIR: Perfect.

Dr Edey : Luci might amplify what I have to say about that. You mentioned the governor's comments where he used the word 'fad', and I think he is right: there is a faddish element to these things in the way that they are discussed internationally. I think what he was reacting against was some of the highly prescriptive and overengineered approaches that are being advocated in some of the international debate. But I did go back and check the record of what the governor said then, and the sentence that followed the use of the word 'fad' immediately said, 'We don't rule out the use of these tools.' So I think you need to keep it in context.

CHAIR: No, I am not saying he has ruled them out. If we go back, Dr Ellis, to your July speech, which you are obviously well aware of, you outlined that we recognise that some of the steps that were taken in the sixties and seventies have not always worked. I just want to get my head around that. Your point is that these are things that have been discussed before. This is not new, and no-one is saying it is new. I suppose the difference now from where we have been in the past is that in the past we have turned around and said we are not going to use these kinds of macroprudential tools and now you are saying not only that they are on the table but that, as we speak, you are meeting with APRA about implementing them. I am trying to understand what has changed.

Dr Edey : I would not accept that characterisation, because we have not said that we will not use these tools.

CHAIR: But we decided not to in the past, or we have not used them.

Dr Edey : Let me complete the answer. We have expressed some scepticism about international approaches to using these tools; I think that is a fair characterisation of what we have said. Luci has said those things in speeches. I have said in speeches of my own that we are sceptical about current international approaches to the use of these tools. The particular scepticism that I have expressed and that I think other members of the Reserve Bank staff have expressed has been towards an approach that seeks to be overambitious and to overengineer the use of these tools. You have had people talking about setting up entirely new frameworks where you have new institutional arrangements and new oversight committees, trying to almost mechanise the relationship between instruments and objectives in a way that we think is unrealistic. We have talked in very sceptical terms about those kinds of approaches to policy. That is why I emphasise the continuity between what we are doing now and what we have said in the past. What we have always said is that we do not need a radically new approach in Australia. We have the tools and we have institutional arrangements that are capable of dealing with systemic risks as they arise. What we believe has happened over the past year is that a concentration of risk in the housing market has come up, it has gradually become more severe and, as that has happened, we have turned up the dial in the response to that, both in our rhetoric and in the way that we have engaged with APRA as to the sorts of supervisory responses that are needed.

CHAIR: I have one or two quick process questions to get my head around. Has the Council of Financial Regulators met about this?

Dr Edey : The Council of Financial Regulators meets four times a year.

CHAIR: Or when needed—I think it is 'responses to potential threats to financial stability'.

Dr Edey : Yes. We do not publicly disclose the dates of the meetings, but it is no particular secret. We normally meet four times a year. The last meeting was early in September.

CHAIR: So last month?

Dr Edey : Yes. I can tell you that the first item on the agenda of a typical council meeting is an assessment of current systemic risks, so these sorts of matters are always going to come up, and the agencies talk about what the risks are and whether there are any responses that need to be made.

CHAIR: So in early September this year, at the Council of Financial Regulators, this came up as an issue?

Dr Edey : Yes, these things were discussed at that point. Please do not ask me to go into too much detail about council discussions—

CHAIR: You know I am about to.

Dr Edey : because these are confidential discussions amongst the regulators.

CHAIR: You produce the financial stability report on a quarterly basis?

Dr Ellis : Half-yearly—thank goodness!

CHAIR: That goes to government a few days before—or how does that work?

Dr Ellis : It is shared among council agencies. It does not go to the government, in the sense of the minister, but the agencies that are council members receive a draft a few days before and they comment on it.

CHAIR: The Council of Financial Regulators—correct me if I am wrong about this; it is all kept a bit hush-hush—consists of APRA, the RBA, ASIC and Treasury?

Dr Ellis : Correct.

CHAIR: And Treasury represents the ministry? During the GFC, Wayne Swan used to go to the meetings. Does the minister attend the meetings?

Dr Edey : Not normally. Treasury represents itself as a regulatory agency. From time to time, for example, the council will form a view that some collective recommendation should be made to the government. That happened quite a number of times in response to the crisis. We do not see the government itself as being part of the council, but the council is a collection of agencies that can advise the government.

CHAIR: At the end of that, you produce minutes and that goes to the members. Are decisions made at those meetings, or are they just the place to discuss risk and what the framework response to risk will be?

Dr Edey : Decisions can be made. But you have to remember that the council has no powers of its own; it only has the individual powers of the individual agencies. It is a coordinating body, rather than a body that has powers that it can enforce. But sometimes it will make a collective decision to make some recommendation to the government or to engage in a collective consultation, for example.

CHAIR: In your opinion, is this becoming the largest risk facing us? You said you used to get together and discuss risk. You are out in the media at the moment talking quite openly about the housing market, about macro-prudential regulation. How much of a risk is this? You have outlined that you think it is a problem. Your own Financial Stability Review said: 'The risks associated with the lending behaviour are likely to be macro-economic in nature rather than a direct risk to the stability of the financial institutions.' Could you explain what some of these macro-economic risks really are? You have outlined that there are risks and you have outlined the data identifying that the risks are increasing, but what are the consequences of not tackling this risk?

Dr Ellis : The share of investor housing is an important indicator of what we are seeing and the price dynamics being so concentrated in particular markets. When we were framing what we wanted to say in the financial stability review, we really focused on: what is the problem that we are actually worried about? We realised that this is not really an issue of it being a threat to, for example, the stability of the banking system, what people would ordinarily think of as financial stability; it is really about the financial stability of households. In particular, the more you have an upswing in housing prices now, being driven by investors in particular, the more likely it is that the end of that boom will be a more severe decline in housing prices. That can catch out certain households—potentially, not the ones that were engaged in bidding house prices up to begin with.

It is important to remember that, in any given period, there are always some households that experience unfortunate events—one of them may become sick, or die, or lose their job, or they may become divorced—and suddenly they are no longer able to manage the financial obligations that they were once able to manage. With that background level of individual level risk, there is a much worse outcome if house prices should fall because suddenly you might find that, instead of 'I can no longer manage the mortgage; I have to sell the house, but at least I'll walk away with some equity,' it is now a situation where they are in negative equity and they will actually default. That is likely to have a much more severe impact upon their spending, and on the people who are relying on their spending for their own livelihoods, than it would if that particular unfortunate event happened to that individual household in a situation where house prices were rising and they still had a base load of equity. So what we are concerned about is the pro-cyclicality of housing prices and what that could do to household balance sheets.

The important thing to remember is that distribution matters. A one per cent fall in household wealth does not make much difference if it is evenly distributed. But if it manifests as, for example, one per cent of households losing their homes, that will have quite a severe effect on household spending and welfare. So that is what we are really concerned about. It is a slightly complicated transmission from the current developments to the shock that we are worried about. It is very hard to calibrate exactly how large that is because it is not something we have seen in Australia before. But it is certainly something I do not want to find out how big it is when it happens; I would prefer to avoid it.

Senator EDWARDS: Have we been in these times in Australia before?

Dr Edey : We had quite a pronounced housing boom a bit over 10 years ago.

Senator EDWARDS: And what were the levers that you introduced then to manage that?

Dr Edey : At that time, there were a number of things that were done. Mostly, we are talking about 2003. That was a period when we had very rapid growth in house prices. In some ways it was similar to what is going on now; very strong growth in investor activity was a part of that. The same sorts of things that are being done or talked about now were being done then. The Reserve Bank was very strong in its rhetoric, warning people about the risks. APRA was doing things to tighten up the supervisory framework for risky loans in that sector. Towards the end of that year was a period when the economy was strengthening, and interest rates ended up going up at the end of that year.

Senator EDWARDS: But you used rhetoric to warn the market rather than actually implementing macro-prudential levers?

Dr Edey : That is not entirely true—

Senator EDWARDS: I am just putting it to you as a contention.

Dr Edey : I think there were some similarities with the current situation. APRA did do some things to tighten up the supervisory standards for high-risk lending in the housing market.

Senator EDWARDS: But they are still in place; they are ongoing, and continual improvement in that space is obviously what we aspire to as a community.

Dr Edey : Yes, that is true. But I think the general principle here is that APRA and the Reserve Bank in general need to be alive to developing risks and to be able to have flexible responses to those. What is going on at the moment is just another example of that.

Senator EDWARDS: How do you manage the regionality of this? Obviously there is a market in Sydney, a market in Melbourne, a market in Hobart, a market in Adelaide and a market in Brisbane—and all of them are at varying states of buoyancy, despair or whatever. You take a big stick to this whole issue. In sorting out a problem in, say, Sydney or Melbourne do you actually create a problem in, say, Adelaide, Hobart or Brisbane? How do you manage that?

Dr Edey : I think we do need to be mindful of that issue. That is why I emphasised earlier that any measures we take have to be well targeted. Exactly how that is done will become evident in due course when the final decisions are made. We have to strike a balance between being overly prescriptive and trying to micro-manage the market and coming up with measures that could be broadly effective. I would just make the mathematical point that most of the growth in investor finance at the moment is coming out of lending into the Sydney and Melbourne market, so any measure that targets investor finance in total is going to have its major impact there because that is where most of the activity is. Whether we need to do something even more than that to target it even more tightly is something we will need to think about. As a general proposition, something that is targeted at the investor market is going to have its main impact in the areas where the largest imbalances are at the moment.

Senator CANAVAN: Are you considering implementing tools which would require banks to place more stringent lending requirements on lending in Sydney and Melbourne but not in, say, Roma or Thargomindah?

Dr Edey : I have tried to be careful not to rule anything in or out.

Senator CANAVAN: I said are you considering that.

Dr Edey : All I really want to say is we are considering carefully the targeting question. I do not particularly want to say anything about geographical targeting. The main thing I want to say is that the prime candidate for being targeted is the investor market. It happens to be the case that most of that activity is happening in Sydney and Melbourne at the moment.

Senator CANAVAN: But you are not ruling out geographic targeting?

Dr Edey : I am not ruling it out or in.

CHAIR: Is it fair to say, Dr Edey—to use your own words—you are prepared to say that it was 'unlikely' that you would go down the path of setting LVRs. Is it 'unlikely' that you would consider going down the path of regional targeting?

Dr Edey : I do not want to be unhelpful here, but I really just do not know how likely or unlikely that is because—

CHAIR: You knew LVRs were unlikely.

Dr Edey : That was a pretty clear case because you could see that, if you target LVRs, you are going to hit the wrong part of the market. This one is a bit—

CHAIR: Regional targeting is on the table?

Dr Edey : That might be too strong a way to say it. What I said was—

CHAIR: It is not on the table; it is not off the table.

Dr Edey : I just do not know how likely or unlikely it is.

CHAIR: Okay. Senator Edwards.

Senator EDWARDS: I asked that. Senator Canavan asked that. You have asked that. I think you are not ruling it in or out.

Dr Edey : That is correct.

Senator EDWARDS: I will go to my learned colleague on my left, with the rest of my time that is available, because I know everybody want to participate. But before I do, I will talk about: New Zealand, Canada and Korea. They introduced macroprudential tools from the period of 2008; and more recently New Zealand in 2013. Have you been monitoring that, benchmarking it, seeing what the effect of those instruments have been in those economies, and whether it has achieved what it has set out to achieve or not? And how do you manage that in the context of what you are talking about?

Dr Ellis : Absolutely, we monitor what other countries are doing in this space. Whether they brand it as 'macroprudential' or not, certainly there are a number of countries that are engaging in particular measures. I think it is important to understand that, with the Canadian case, it is really about the government deciding how much credit risk in the mortgage market it wants to take on itself, because the government owns the main lenders mortgage insurer in that country. So what they are really doing is, like any lenders mortgage insurer, deciding what lending standards they are comfortable with and what kinds of loans they will insure.

The kinds of measures that have been announced in Canada are exactly the kinds of things that the private sector lenders mortgage insurers would say about their own business here in Australia. They decide how much credit risk they are comfortable taking on. So they would not necessarily be seeing that as a regulatory measure; it is about managing the government's credit risk.

Senator EDWARDS: That is not dissimilar to here; because in 2009 the government decided to guarantee deposits of our major banks, and indeed our secondary banks in this country, so maybe that might be an issue again.

Dr Ellis : At a far remove, you could make that analogy. But I think it is a different situation in Canada because in that case—

Senator EDWARDS: I get that.

Dr Ellis : the government is guaranteeing the mortgages directly—that are being bundled into securitisations and being sold to investors. So that is a direct government guarantee of a specific loan. And, like any lenders mortgage insurer, the Canadian government, through the Canada Mortgage and Housing Corporation, is entitled to decide what lending standards it will accept.

New Zealand is another situation. Certainly we have been very closely liaising with and discussing with our New Zealand counterparts. We have a very close relationship. We have done things a little differently. That is partly because of the different institutional arrangements we have in this country. In New Zealand the RBNZ is also the prudential supervisor; in Australia they are different. I sometimes worry that, when we say 'we', it may not always be clear to you that sometimes we mean 'we' as in the Reserve Bank and APRA or the council, rather than 'we' the Reserve Bank. I think it is important to remember that it is not 'we' who will be instituting any kind of supervisory tools. It will be APRA, and Malcolm has already made that point.

In New Zealand they have gone more along the lines of having a formalised framework where you prespecify what tools you might need, regardless of what risks might come along, these are the tools that we have already chosen, and they have gone through the formalisation. We understand that that is the trade-off they had to make between flexibility and having buy-in, but in the Australian context I think we would prefer not to say, 'Here are the four tools that we would call macroprudential tools', and then risk seeing every risk looking like a nail because all we have is four hammers.

I think one of the observations we have made about the macroprudential frameworks that other countries have put in place is that they are always about housing. One of the first things that was mentioned in this hearing today was LVRs and DTIs; they are the sorts of things that are discussed in the international comparisons of what different countries have done. But here we have a situation where it is not high LVR lending that is the issue; in fact, high LVR lending—LVRs above 90—are a smaller share of new lending at the moment than they were a couple of quarters ago. That is not the problem we have, so we do not want to use that tool, and we do not want to have that as the only tool in our toolkit. That is why we said we see macroprudential policy as being a state of mind rather than a suite of prespecified tools. We do not want to carve out a few tools from the framework.

So we understand completely why the New Zealanders have done what they have done. In New Zealand there has been an increase in high LVR lending. That is the risk that they see that they have. We see ourselves as having a different risk.

Senator EDWARDS: Are there any policy settings that have been ongoing, which you believe are exacerbating this problem of investor markets and that exist outside economics, whether it be migration policy or any other policies that you can speak of, which you could use rather than the macroprudential?

Dr Ellis : I would refer to our submission, where we did go through quite a number of the demand factors that would be adding to demand generally for housing, and perhaps making investor housing particularly attractive. I do not particularly want to comment as a central banker on the government's migration policy. We did lay out some facts in there.

Of course the Reserve Bank has in the past—back in 2003, we made a submission to the Productivity Commission where we did discuss some of the other factors that may be affecting investor housing demand. One of the things that really struck us during that 2003 period was that not only were we aware that this was very much an investor housing driven phenomenon, at that time as well, but this was not something we really saw in other countries at the time. We engaged with a number of countries during that 2002-03 period to find out what their experiences were, and not only were they saying, 'it doesn't really seem to be this buy-to-let market'. But in fact they did not have the kind of data that we have. One of the advantages that Australia has had in all this period has been that we actually had a good handle on the split between owner-occupiers and investors and we have done for many years.

CHAIR: Senator McLucas?

Senator McLUCAS: Thank you. Thank you for appearing today. I want to go back, Dr Ellis, to your comment: 'What is the nature of the problem?' Is it just in the investor market and just in Sydney and Melbourne? Can you describe the nature of the problem, in your words?

Dr Ellis : Malcolm might want to chime in here more eloquently than me. The nature of the problem is primarily concentrated in the inner parts of Sydney and Melbourne in terms of the investor demand. But I think we would also see that as a bit of an indicator of speculative intent more generally. But it is the bit you can get a good handle on. So, for example, if someone wants to speculate on rising house prices and they do a 'buy, renovate and flip' and they are living in the house, that will be recorded as an owner-occupier loan. We do not know how important that is at the moment, but surely there is some of it going on. Similarly there is certainly an attempt by the real estate industry to attract investors to other markets, and say: 'Look at this nice low price'; you can see them particularly in a number of regional areas where the lower priced houses are marketed on the websites as being 'suitable as a first home or for an investment' and they are really trying to encourage that.

But I think we would see the imbalance as being primarily in the Sydney and Melbourne markets. But of course what we do not know is whether the down-swing will be quite so concentrated. Once you have the cycle in house prices—at the moment the increase in house prices is primarily concentrated in those two cities. To be honest in some other parts of Australia house prices are rising a little bit faster than incomes but not by a whole lot and not by the kind of growth rates that would cause you to be too concerned. But it is part of a general cycle in house prices and we are worried about what the downside of that house price cycle might look like.

Senator McLUCAS: So, in responding to that—and I take your point, Dr Edey; you do not want to tell us what potential tools you might employ—how do you make sure that whatever is done is very targeted so that we do not end up with unintended consequences in other parts of the country. How do you model that so that you make sure that you do not make a mistake that will affect a market in another part of the country in a negative way?

Dr Edey : I think there are a few principles that have to be kept in mind. One is that it has to be targeted. We have already talked a bit about what that means: identifying what the problem is and designing measures that specifically address that problem. It has to be proportionate. We are not trying to kill the investor market. We are not against investors; we are just against imbalance, so it has to be proportionate. Yes, I think those are the two main things really. It has to be well targeted. It has to be commensurate to the problem that we are facing.

Senator McLUCAS: Do you have or does APRA—or any of the regulatory agencies; I dare say it is APRA—have the capacity to say, 'Okay, if you're an investor and you're going to invest in a property in these postcodes in Sydney and Melbourne then these tools will affect you'? Can you target in that way, or is it just to the general risk that might be presenting itself?

Dr Edey : It will not be that narrowly targeted.

Senator McLUCAS: I do not think you have the flexibility to do that.

Dr Edey : We are not social engineers who are going to get into the postcode level and tell you what market can grow and what market cannot. The other thing that I meant to say was that it needs to be incentive based. We cannot have hard rules that say, 'You can lend this much here and that much there.' That is not going to be effective—

Senator McLUCAS: That is right.

Dr Edey : and that is not the way that APRA's tools work. It is incentive based.

Senator McLUCAS: How do you stop then? If you bring in some tool, whatever it might be, which is trying to dampen investor activity in markets in the country, there are parts of the country where it is a good thing to be encouraging investor demand—parts of Queensland, for example, where there has been a downturn in investor housing and it is generally agreed that it would be a good thing for us to have some more investment in the residential property market. If you bring in a tool and you say it is targeted, how do you make sure that you do not end up with a very negative outcome in other parts of the country?

Dr Edey : I was trying to make this point before. Even without geographic targeting, anything that reins in the growth of investor finance is necessarily going to have its main impact in Sydney and Melbourne because that is where all the growth has been in recent times.

Senator McLUCAS: That is my concern, Dr Edey, because you are saying that they will all be dampened down. That is fine in Sydney and Melbourne, and that is your intended outcome, but there are other parts of the country where that would be a negative thing to happen to the housing market.

Dr Edey : All I can say is that we need to be mindful of that risk in designing the tools. Because the details have not been worked out yet, it is very hard to comment specifically on that, but I think there is an important principle here that the tools need to be risk based and incentive based, so they are not based on hard, quantitative restrictions. If they are incentive based, it means that they will allow flexibility for banks to allocate their response in a way that is proportionate to the risks in the different sectors. That is the objective.

Senator McLUCAS: I just do not know how you do it.

Dr Edey : Luci might want to comment further on that.

Dr Ellis : Unfortunately, I do not really want to delve into the details of the capital framework because, although we have agreed to go over time, I do not think we have that kind of time. But one thing that is worth bearing in mind is that part of APRA's framework includes the ability to impose capital or supervisory guidance around concentration risk. If the risk concentration is in a market that is growing quickly, that is what would attract the supervisory response, whichever kind of measure they may choose to take up, so that would be the way you would frame it: in terms of concentration risk. Here is a bank that is filling its boots on lending in a market that is growing strongly—let us not name names—and of course, if it is framed in that way, it would automatically not really impinge on a market in which there is not that kind of concentration risk.

How exactly supervision works is not something we have time to go into today, and I do not want to necessarily say that that is precisely the structure that they would use, but there are ways to target these things within the framework. People have to go through and say, 'Is this a pillar 2 thing or is this a pillar 1 thing?' Unless you really know what that means, it is probably not that helpful, but there are ways to target things that are not quantitative, in the same way that Malcolm mentioned, and there are things that APRA can do in that regard. Because we collectively are mindful of the very concern that you have just raised, it is not a simple thing to say: 'We're concerned about this. Okay, next week we work out exactly what to do.' It requires some careful design, precisely to avoid those unintended consequences. We understand that those are consequences that we do not want to have, so we can design around them.

There are also the unanticipated unintended consequences, and we have to be mindful of the possibility that we might create some of those. That is why it is a careful design process by APRA in consultation with the Reserve Bank and the other council agencies.

Senator McLUCAS: I have just one little question. When you do that sort of modelling, is it just done across the whole of the market nationally, or do you drill down to certain locations and have a look at what would happen in that location?

Dr Ellis : To the extent that data are available, we do drill down as best we can. It is interesting. The way our analysis has evolved over the last four or five years has been that it was always thought that financial stability is about the system, so you just look at 'the system' and its aggregate, and I think one of the things we learned as a consequence of the crisis and particularly what happened in the US is that you cannot just look at averages and aggregates.

Senator McLUCAS: That is right.

Dr Ellis : The median borrower is never the one that causes the trouble by defaulting. The distribution really matters. I think that increasingly we have started to try to understand geography. We do a lot of work on income distribution. We do a lot of work on what the distribution of gearing ratios amongst corporates is. We look at individual bank data. You cannot just look at the average to really understand the risks, and that is something we emphasise. But of course, once you get down into that level of detail, the data are not as good, or they may be quite noisy, so we just have to be very mindful of that.

Senator McLUCAS: Great.

Senator CANAVAN: Is there a bubble in the Sydney housing markets?

Dr Ellis : I do not think that is a particularly helpful way to frame the problem. There is always going to be some speculative demand in any market, and even in the hottest market there are always going to be people who are just living their lives and upgrading and downgrading and moving closer to their families. You will never get a good model. People desperately try and come up with a good model of fundamentals, and then they say, 'Well, prices have departed X amount from these fundamentals.' Well, that could just be because your model is not very good. So we have never really thought of that as a helpful way to frame the issue.

What matters is how much speculation there is in the market and what that might mean for a subsequent price cycle. At the moment there is more speculative activity than we are comfortable with, as we described in the investor market, but we do not find the label to be particularly helpful.

Senator CANAVAN: So you admit that you cannot identify bubbles ex ante?

Dr Ellis : I think what happens is that you end up with a debate about whose model is best, and the people who confidently pronounce that prices have deviated X amount from fundamentals are usually using some pretty simple metrics, which usually turn out to be quite misleading, to derive that. I think that in some ways not worrying about whether you can identify a bubble or not is helpful because otherwise you say: 'Well, we can't see this bubble. After all, there are all these other fundamental reasons why demand is strong.' And that is true. We have strong immigration. We have had a period where supply has been quite weak. And those become excuses. So we would much rather focus on the risks than the label.

Senator CANAVAN: Just so that I am clear in my mind, I am trying to repeat it back to you. You do not want to get into the semantics of whether there is a bubble or not, but you do believe that you can identify whether there is too much speculation in the market.

Dr Ellis : That is right. I think the problem is that, once you try and frame it as 'these kinds of fundamentals', you then get into the econometric problem of whether you have a good model. People will confidently proclaim that they can identify bubbles, but usually they end up identifying 18 of the last three crashes. I think that, for example, just purely looking at the run-up in prices is often not helpful. The run-up in prices in the US was not big compared to some of the markets that did not have the same kind of crash.

You have to look at the whole picture, including whether there has been overbuilding, which there was in the US and places like Ireland. You have to look at what the lending standards were and what the resilience of the household sector is to any kind of fall in house prices.

Senator CANAVAN: Sure. Can you provide the committee with the RBA's definition of macro prudential tools?

Dr Ellis : The distinction that I would make is that in Australia we have a financial stability policy framework, which is as described in the paper that APRA and the Reserve Bank published in 2012, where we both have financial stability mandates and we respond to the risks as we see them with the tools and the measures that we have. We note that in some countries the way that has been framed has been to pre-specify what specific bits of the prudential framework you can carve out—they are macroprudential—and then whatever else the prudential supervisor does has to be microprudential, that is, focused only on the risk to individual banks.

One of the things that we have been discomfited by is not so much the word 'macroprudential' as the word 'microprudential'. When you describe how a microprudential supervisor is meant to work, it is very narrowly focused on, 'Is this bank within the rules? Is this bank safe and sound?' I think that the way APRA's mandate is framed and the way we see the policy problem mean that it is just much better to treat prudential supervision in an holistic way, taking the whole system into account and situating it as part of a broader financial stability framework, which our council arrangements allow us to do.

So there is a macroprudential or financial stability policy framework but you can also think of pre-specified tools. And the idea of a separate committee deciding on those is the thing that we have been discomfited by about the international debate.

Dr Edey : If I may add to that? I think there are really two key elements when you are talking about macroprudential. One is that it is focused on the system as a whole, not just the individual institutions. And the second thing is that there is a time-varying element to it. So, if some new risk develops that affects the system as a whole then you have a flexible response to that. That is why people use the term 'macroprudential' for what we are doing at the moment.

Senator CANAVAN: If this is all about systems as a whole, why are we just focused on speculation in one market in one city in the country?

Dr Edey : It is a segment of the market that the system as a whole is actually feeding into that market. To use a rough shorthand, all banks—the banking system—are putting a lot of concentrated risk into investor housing finance. So that is a system-wide problem, but it has a targeted—

Senator CANAVAN: But I would put to you that yesterday RP Data came out with their figures—and you would probably be aware of this—that four capital cities have had falls in house prices in the past month. The talk here is that we want to introduce tools and restrictions on lending because of overspeculation in housing markets. How does that analysis square with the fact that in four capital cities house prices have gone backwards in the last month?

Dr Edey : Well, those are monthly figures. It is very dangerous to draw conclusions, over the policy frame that we are talking about, based on one month's data. So—

Senator CANAVAN: I would agree with that. So, in the last quarter Perth has gone down by 0.6 per cent, Hobart down by one per cent and the rest of state, which is regional areas, are both down by 0.6 per cent in the last quarter. So there is no overspeculation there. I ask again: how do you define—or how do you assess—these system-wide systemic risks? In your analysis, they appear to be emanating from one or two cities in the country that require a macroeconomic response?

Dr Edey : We have said that this is a risk concentration problem, but it is big enough to have impacts on the national economy. That is what we are worried about in the end. Our mandate is the performance of the national economy, the stability of the economy and, eventually, the financial system itself. Sydney and Melbourne are a big part of the economy; they are a big part of the lending focus for the banks.

Senator CANAVAN: How much are they? Sydney and Melbourne for lending finance?

Dr Edey : Luci is just quoting some figures here—

Dr Ellis : They are 55 per cent-ish of the urban population and a little bit less—closer to 40 per cent—of the total population. That is a reasonable fraction. They would be more than that of lending, although I could not swear to the exact figure, because incomes and housing prices are higher in those two cities.

Senator CANAVAN: Okay. But you are making these judgements. You have been put in this position and you have great responsibility, but you cannot tell me how much Sydney and Melbourne account for in our housing lending?

Dr Ellis : Just not off the top of my head.

CHAIR: Take that on notice?

Dr Ellis : Yes. The data is actually by state, but the bulk of it would be the centres. So, I mean—

CHAIR: Take it on notice.

Dr Ellis : What I can tell you is that New South Wales and Victoria collectively were about $6 billion of new loans to owner occupiers, compared with $4 billion to the rest of the country. For loans to investors—this is per month—it is nearly $7 billion—goodness me!—per month in New South Wales and Victoria versus about $3½ billion.

Senator CANAVAN: Do you have figures for Sydney and Melbourne?

Dr Ellis : These are the ABS figures for housing loan approvals, so they are not actually at a city level.

Senator CANAVAN: Do you have those—I know they are not publicly available.

Dr Ellis : No, it is only collected at a state level.

Dr Edey : We can get you some figures on that, but the fact is that particularly in the investor market, New South Wales is an outsized proportion of the lending activity in the national market at the moment.

We have a balance to strike here. We have talked about being targeted and we have also talked about having a system-wide response. What is going on here, even though it is in some ways concentrated, is a big enough event to affect the macroeconomy as a whole. That is why we are concerned about it.

Senator CANAVAN: I just want to pick up on that point about 'targeted'. I am a bit worried about the language there because possibly we seem to be falling into the trap that George Santayana said: those who cannot remember history are doomed to repeat it. You are trying to manage these macroeconomic risks by, what I can assess from what you said, making more stringent lending standards in certain cities to certain types of lending. I thought all the experience we had with policies like quantitative restrictions in the sixties and seventies and regulation queue in the states was that the people who banks employ tend to be smarter than us and probably smarter than you. They get around those regulations and they find other ways—

CHAIR: I hope that is in relation to yourself, Senator!

Senator CANAVAN: I am certainly not earning as much as they are! They get around those restrictions and the bubble just comes up somewhere else. How are you going to prevent that happening today, when it clearly happened to things like regulation Q in the seventies and eighties?

Dr Edey : I have been trying very hard to avoid saying what you have just accused me of saying, which is this idea of tight geographical targeting. Again, at the general principle that I am trying to put across here is that there will be an appropriate level of targeting at the problem. In the first instance, the problem is excessive growth in risk exposure in the investor market.

Exactly how you come up with policies that target that, it is very hard for me to be specific about at this stage. But what I have tried to say is—

Senator CANAVAN: Let me just put this to you quickly: if you did that to investor housing, surely a bank could just simply design a product which gives lending to a trust or some other vehicle outside your framework, and then they go and invest in the housing. My understanding is that you are not talking about limiting lending to trusts, partnerships or mutual funds only—basically, only to mums and dads who are investing in housing. Surely, smart people will be able to get around that very easily?

Dr Ellis : I think that one of the important barriers against that particular approach is that the risk weights on lending to mums and dads to invest in housing are lower anyway than lending to a trust. I do not think that would be feasible. Hypothetically speaking, if you increase capital or you required some other sort of restriction on a particular kind of activity in lending to individuals for housing purposes, you would have to go a long way for that to then be higher than lending to a small business for the exact same purpose. For example, there are some quite stringent restrictions for self-managed super funds. The know-your-customer requirements would get around that as well.

Also, it is partly because lending to individuals is also covered by the National Consumer Credit Code, which is administered by ASIC. there has already been a fair bit of work to ensure that there is not avoidance activity by magically declaring yourself to be a business. But, in any case, the way the capital framework works, that particular avoidance activity would not be that effective.

Dr Edey : The general point you are making is: we need to be careful to avoid stimulating avoidance activity. We are mindful of that. I think one of the things that comes out of that is that any response needs to be incentive based; it cannot be based on hard rules that create cliff effects and easy avoidance opportunities. These are things that need to be taken into account carefully as we design the tools.

Senator CANAVAN: Can I just ask one more quick thing? I am quoting from an IMF document here, Key aspects of macroprudential policy, from June 2013. I just want to ask you for your opinion on what they say here:

… macroprudential policy is not well-suited to control asset prices, including the prices of securities (stocks and bonds), or interest and exchange rates.

They go on to say:

… affecting them should not be seen as a primary aim of macroprudential policies.

Would you agree with that statement?

Dr Ellis : They are referring to financial market prices—stocks, bonds et cetera. I think the reason they note that is because many of those kinds of activities are not financed through banks. Whereas—

Senator CANAVAN: So you think that an objective of macroprudential policy should be to control housing asset prices?

Dr Ellis : I think the objective of financial stability policy should be to control risk and to manage risk, some of which may manifest as the risk of a severe downswing in housing prices. We are not seeking to control housing price upswings in and of themselves; we are seeking to control, or to manage, the risk that there could be a bad effect upon households because of the subsequent downswing. What we are seeking to do is just avoid the scenario that the upswing in housing prices will be of a nature that then creates a sufficiently strong downswing that harms households.

Senator KETTER: Dr Edey, I understand your reticence to go into too much detail, but I would like to have a peep back inside your macroprudential tool kit for a second. You have said that the use of LVRs is unlikely. DTIs: would you also say that their use is unlikely?

Dr Edey : DTI?

Dr Ellis : Debt-to-income ratios—maybe you want me to take that one?

Dr Edey : I will pass to you in a moment. Again, that is something that I do not want to rule in or out. When I was speaking earlier I was trying to emphasise the continuity between what is being done and what is going to be done. APRA has already done things to tighten up the prudential framework for banks over the past year or so. I see what we are talking about now as very much a turning up the dial for that. I would not at all rule out that there can be more than one announcement here; that the dial might be turned up now, and if that is not sufficient it gets turned up some more later on. That is what flexible response is all about.

I think we need to distinguish between two kinds of tools. There are tools that are specifically focused at the risk that is emerging at the moment. I think those sorts of tools would be looking at ways to provide banks with incentives to lend less in the areas that are problematic and, particularly, investor finance. They are also—

Senator KETTER: Sorry—can you say that again? I did not quite catch that. Can you say that last phrase again? You are looking to lend less—what did you say?

Dr Edey : To restrain the growth of lending in the currently problematic area, which is investor finance. There are those sorts of tools that are targeted at the current problem. But I know that APRA is also concerned about risk management practices in some areas of the housing market more generally. It is open to them to take the opportunity to look further at tools that might be in place for a longer period of time to tighten up risk management.

Senator KETTER: From my limited understanding, the DTIs do have a disproportionate impact on the investor sector of the market.

Dr Ellis : Let us just be clear whether we are talking about debt servicing to income or loan to income. I think this is one of the problems with the international debate on macroprudential. There have been a number of attempts to come up with something you can graph. And LVR is something that you can easily graph. We have graphs of that in our own financial stability review, and we find that very useful as an information variable. But does it actually map into best practice credit risk management?

A debt servicing ratio—the repayment-to-income ratio—is a good way of thinking about the appropriate size of loan to extend to a particular individual. Certainly, that is the kind of thing that banks think about when they are deciding how much to lend. But there is more to it than just saying, 'Thirty per cent debt servicing to income is the limit.' Again, that is one of the reservations we have had about the way macroprudential has been framed in the international debate. We would much prefer to see the serviceability dimension of lending standards be seen in their totality. Thirty per cent of income might be quite a struggle for someone who has a lot of other obligations, but it might be really easy for someone who has quite a high income with quite a high discretionary income who is willing to make the spending sacrifices in order to service that particular loan and who is quite willing to cut back their spending to manage a larger repayment. So setting a ceiling on that does not really map into the risk management for individuals. That is why Malcolm said the kinds of measures APRA have in mind are really incentive based and within its existing prudential framework rather than setting these ceilings.

Certainly serviceability is likely to be more binding for an investor than loan-to-valuation ratios. Typically, investors have a reasonable amount of equity in their property. They are not the high-LVR borrowers. They are more likely to push the serviceability because they are trying to maximise the tax benefits of whatever it is they are doing, but that kind of measure would also hit owner occupiers, particularly the people who are stretching to afford that home knowing that that will get easier over time. So that is one where it would bite for investors more than LVR caps would.

These are important questions of design, and the sorts of things that get discussed in the international debate are usually about housing generally. They are thinking about it in terms of the housing market in general. I think we have been more focused on what the risk and the problem actually are. That is why we and APRA have had a more focused approach on the current events. As Malcolm mentioned, we are also thinking about practices generally in credit risk management for the mortgage industry, which has a number of dimensions, not just LVRs or debt servicing ratios.

Senator KETTER: How do you design a measure to address some of those points but at the same time not create new barriers to market entry for would-be first home owners?

Dr Ellis : Both Malcolm and I have been reluctant to go into a lot of detail about something that is already in the process of being discussed and designed. Some of those complexities and how you design around them are being thought about, but in some ways we are very aware that the rest of Australia is not seeing the same kind of conditions as we are seeing in Sydney and Melbourne. Sydney and Melbourne are too big to ignore, but they are not the whole market. We would not want to be Sydney-centric. We have been accused of that before and certainly would not want to do that again. I think it is important to focus on where the risks are—where are the risks, what are the concentration risks, what is growing quickly—and design flexible measures that are focused on what is growing quickly and where the risks are.

Dr Edey : It is not our objective to structurally favour one part of the market—first home buyers versus investors and so on. What we are concerned about is imbalance. We see an imbalance at the moment. The sorts of measures we are talking about should if anything be good news for first home buyers because one of the complaints at the moment is that they are being priced out of the market by investors ramping up property values. In that sense, they should see it as good news if there are actions to limit activity in that area.

Senator XENOPHON: Bank of America economist Saul Eslake was in the media this morning talking about a number of factors in terms of the housing market, saying that negative gearing was a factor in squeezing out people trying to enter the housing market, pointing out that there are, despite record-low interest rates, fewer people being able to get into the housing market, particularly at the lower end. Dr Edey, you said earlier that you are not against investors; you are against imbalance. Do you see the current structure of negative gearing as creating or allowing for an environment of imbalance?

Dr Edey : We made some extensive comments on the tax treatment of housing 10 years ago, and our general position has not really changed. If you have a tax system that excessively favours leveraged investor activity then that is going to cause distortions in the market.

Senator XENOPHON: Including the imbalance that you referred to.

Dr Edey : It might contribute to that, but the thing to say in counterbalance to that is that this tax regime has been in place for a long time. It is not something that has newly triggered the recent bout of activity, so it is hard to blame it for that.

Senator XENOPHON: But, if it were tweaked, it may have some. I am not talking about getting rid of it; but, if it were tweaked and nuanced, that could have some impact.

Dr Edey : Tax changes could well have some impact on the current impact dynamics. We are not giving any particular tax advice to the government at the moment, and I think I should also stress that it is more complicated than just negative gearing; it is the way negative gearing interacts with other aspects of the tax system.

Senator XENOPHON: Of course. I go to the issue of mortgage stress. I think the debt-to-income ratio has increased fairly markedly in the last 10 years. It has more than doubled. Has the Reserve Bank done an analysis on what stress it would put on heavily leveraged households if interest rates went up one, two or three per cent?

Dr Edey : Yes, we do a lot of that kind of analysis. Generally what that shows is, first of all, as I said at the beginning, the current interest burden is actually lower than average. Interest rates are so low that, despite the fact that the debt levels are high, the interest burdens are actually still quite low at the moment. Second, the analysis that we have done shows that the people who are holding the debt typically are the people who are best placed. This is a general observation; there will be people who are exceptions, but generally the people who are holding the debt are the ones who are best placed to service it.

Senator XENOPHON: At what level would you be concerned about housing stress in terms of interest rates going up? Two per cent? Three per cent? Four per cent? Would you be able to proffer a figure that you think would get to uncomfortable levels of mortgage stress in the marketplace?

Dr Edey : I would not like to put a figure on that, because it is very hypothetical. That is one reason. Another reason is that, when monetary policy decisions are being made, they are made with the intention of having particular effects, taking into account the sorts of adverse stress impacts that you are talking about. In some ways it is a bit artificial to say, 'What would happen if we did something silly with interest rates that then caused a whole lot of stress?'

Senator XENOPHON: I am not suggesting the Reserve Bank would do anything silly. My final question relates to manufacturing and jobs in my home state. This also applies to Victoria. We have seen the demise of three original car makers in this country. There is a concern in my home state particularly about shipbuilding jobs in terms of what decisions the federal government will make. In the auto sector some of the key industry groups are concerned about many thousands of jobs being lost in the next two to three years and potentially tens of thousands of jobs if there is not a decent transition. To what extent do you take into account those pressure points in terms of the housing market? If you do have many thousands of people being laid off and people not being able to maintain their mortgage repayments and losing their homes, at what point does that impact on the housing market?

Dr Edey : Obviously that is a very significant stress factor for the South Australian economy.

Senator XENOPHON: Are Holden, Ford and Toyota exiting the country in terms of making cars on your radar?

Dr Edey : We do a lot of work on analysing regional developments and taking those into account in the broader picture that we build up for the national economy. Those things are very important to us.

Senator XENOPHON: So, just as you may be concerned about, say, an overheated market in one region, are you concerned about the market suffering significantly as a result of a regional factor with a significant spike in regional unemployment?

Dr Edey : Certainly we do not play regional favourites here. The general principle is risk management. We try to identify the risks and, if a risk is big enough, try to promote targeted action to address that. That is what we are talking about at the moment.

Senator XENOPHON: But it is on your radar that there could be many thousands of jobs lost in the next couple of years?

Dr Edey : Yes, very much so.

CHAIR: I want to make sure we are reading this and hearing what you are saying right. The word you keep using is 'imbalance'. The choice word you have chosen is that there are 'imbalances' in the housing sector, and that is presenting macroeconomic risks. We have kind of explored all that. I am quoting comments by Mr Jeremy Lawson, who is known well to you, Dr Ellis. You wrote a paper with him on housing leverage, and he worked on the OECD. Is it true the Reserve Bank 'underestimated how strong the investor pulse would be'? Clearly you are saying that there are imbalances at the moment. You are saying there are imbalances in sections of the housing market. We are saying that we are going to look at the tool kit that is available to the regulators to address those imbalances. This goes to the comment made by Mr Lawson. Isn't it fair to say, then, that you 'underestimated how strong the investor pulse would be'? Isn't a lot of this a natural response to what is going to happen when you have a record-low cash rate of 2½ per cent? How did we not see this coming as a problem?

Dr Ellis : If you look at the September 2013 financial stability review, we were already discussing this. The language in the September 2014 financial stability review was stronger than in the previous couple of issues, but we were already drawing attention to the rising share of investor housing lending in New South Wales in that particular—

CHAIR: Hang on; I have it here. You are referring to September 2013. That is the same time that you came out and made some pretty strong comments about bubble terminology and how, every time the rate of increase in housing prices is higher than average, you are just going to be unrealistically alarmist by making that call. Is it fair to say you went out pretty strongly in September 2013 to make the point that there is not a housing investment bubble? I know you are not in favour of the terminology.

Dr Ellis : Let's look at what housing prices were actually doing at that time. We have just had a discussion about what is not happening in other regions. We actually talked about housing prices then. They had come up. We said that was to be expected as a consequence of monetary policy, but on page 50 of the September 2013 financial stability review we noted that the increase in investor activity in New South Wales had been particularly sharp. I am reading through parts of this particular paragraph:

An increase in housing market activity more generally is not surprising given reductions in interest rates. However, it is important that those purchasing property maintain realistic expectations of future dwelling price growth; in contrast to the decades leading up to the crisis - when dwelling prices grew rapidly in response to disinflation and financial deregulation - long-run future growth in dwelling prices might be expected to be more in line with income growth.

I concede that the tone of that was less strong than the tone that we had in the more recent financial stability review.

CHAIR: In hindsight, did you get it wrong?

Dr Ellis : No. We were warning about this a year ago.

CHAIR: You say:

… we shouldn't be rushing to reach for the bubble terminology every time the rate of increase in house prices is higher than average…

You're just going to be unrealistically alarmist by making that call every time that happens.

Dr Ellis : We are talking about two different indicators.

CHAIR: Let me put the question to you. It is now 12 months later. We are proposing—these are my words, not yours—extraordinary macroprudential brakes to address imbalances. I am using the word 'extraordinary' because they have not been used before. I am not saying other tools have not been used before, but now we are proposing new tools. Frankly, with the benefit of hindsight, how helpful was it to be encouraging people in your public comments not to worry about these issues that are emerging?

Dr Ellis : Let's go through that. Firstly, as Malcolm said, we do not see these as being extraordinary tools. This is a ramping up of the framework that we already have.

CHAIR: The tools we have not used before in this way.

Dr Ellis : Not necessarily. It just may be more targeted than what we have. APRA has not been given any new powers.

CHAIR: Sure. That is word. These are new tools.

Dr Ellis : I would see them as being a further refinement of what is already there.

CHAIR: So they are tools that have not been used before. We can play with semantics all we want.

Dr Ellis : So that is fine. The other thing I would point out is that we are talking about two different things here. We care about imbalance and we care about the risk to households, so we were warning about the investors. We said, yes, you should expect there would be some increase in housing market activity because of interest rates. We were alert to that. We did not miss that one.

CHAIR: So why didn't we do this a year ago? Why weren't we using these tools to address this a year ago?

Dr Ellis : Throughout the past year APRA has been working on its supervisory intensity on the banks, but it is just not something that is seen particularly publicly. I would point, for example, to the Prudential Practice Guide that has come out. There has been a number of steps, both public and not so public, in the supervisory activities of APRA that have been turning up the dial, as Malcolm said. One thing I would—

CHAIR: You are saying that you have not underestimated? You do not believe that the RBA underestimated how strong the investor policy would be a year ago?

Dr Ellis : I do not think so, because we were already warning about it a year ago. We were already pointing to the fact that as, primarily, Sydney and Melbourne—

CHAIR: But we did not do anything about it then?

Dr Ellis : APRA was doing things about it and we were—

CHAIR: But you were not doing anything about it?

Dr Ellis : I think we were pointing to some of the risks. I would also point out that housing prices had not really risen that much at that stage. An important point I would make is that we are concerned about the imbalance; the imbalance is the concentration in the investor activity and that—

CHAIR: But when you have interest rates at 2½ per cent, isn't this, in a sense, a sensible response for investors when you have a cash rate that low?

Dr Edey : That is a very big question and if you want me to I can go into why we have got the interest rate configuration we have—

CHAIR: No, I am just asking the question. I think my questions are very simple. Fundamentally, what we are here talking about is the investor policy: how strong it is, how strong it is not, how much you warned and how much you did not. Some of this is a kind of sensible market response to the fact that you have interest rates at 2½ per cent, isn't it? This is what is going to happen when you have interest rates there. I am not saying it is a bad thing, I am just saying, 'Isn't that just a reality?'

Dr Edey : The proposition you are trying to put to us is that we have completely missed all of this and suddenly we are panicking and scrambling to catch up—

CHAIR: No, that is not what I put to you at all.

Dr Edey : I do not accept that characterisation and I do not think the record supports that. We have been talking about this for a year. As the risk has built up our rhetoric has become stronger and APRA's responses have become more comprehensive. We are moving further in the same direction now, but it is a continuation of what we have already been doing.

CHAIR: The other bit that I want to get my head around, if you could just explain it to me, is that in the Financial Stability Review you argue that the housing risks brewing in Australia are—I think the term you used—macro-economic. I want an understanding of to what extent this is a present material financial stability concern?

You spoke a little about this earlier. It is hard to reconcile on one hand a part of the rhetoric, which is that this is a big issue, it needs to be addressed, we have been addressing it, we continue to address it and this is a concern. And on the other hand is the other issue, which is that it is not really a concern; I think the words you used were, 'The banks aren't in crisis and we are not risking the banking system collapsing on itself.' Especially when you look at—again I am quoting some of the stuff from Mr Lawson, but this is the RBA's own research—'If Australian house prices only track historical disposable incomes into the future, they are currently about 20 per cent overvalued.' They are his figures, we can get into a debate about that. He says that home loans represent 65 per cent of bank assets, the housing debt to income ratio is at an all-time high of 137 per cent and housing prices relative to incomes are way above what long-term averages are, and that we have a bigger boom in speculative investor activity than in the last 'worrying cycle' of 2002 to 2003. So my question is: to what extent is this a present material financial stability concern?

Dr Edey : I have made the point that the banks are resilient. These developments are not going to drive the banks broke in the foreseeable future. The banks are not going to get into that kind of stress immediately as a result of these developments. The initial risk is to households. That is what we have set out in our statement. The longer this kind of development—

CHAIR: So there is no risk for the banking sector in this?

Dr Edey : Just let me finish. The longer this kind of development goes on, the more households will be at risk, the greater the risk to the macro-economy and the potential for that to feed back to the banks and amplify the whole thing. So it is a financial stability issue, but we are not trying to panic people by saying the banks are at immediate risk. What we are saying is we need to take timely action to stop a developing risk from getting out of control.

With the sorts of questions that we have been receiving today, we are very happy to answer them; but I think it is instructive to note that we are getting two kinds of questions. We are getting questions that ask, 'Why do you need to do anything at all? Aren't you panicking; we do not need to do anything.' And then we are getting another set of questions that ask, 'Surely, this is all inadequate and you should have acted sometime ago. This is all terribly too late and we have missed the boat.' To me it is not surprising that we are getting criticism from both of those sides, because we think we have struck a sensible middle course.

CHAIR: Dr Edey and Dr Ellis, I think we have been asking that.

Senator CANAVAN: I want to pick up on what the chair is saying. To the extent that you are concerned about overinvestment, wouldn't typically your tool be interest rates? That is the tool you have to deal with overinvestment in markets.

Dr Edey : Again, this is a big question. Interest rates are set on the basis of a whole range of factors for the economy as a whole not just for housing. What we are saying at the moment is, 'We think we've got the right interest-rate setting, but this happens to be an environment driven by global factors which are generating a lot of speculative activity in one area'.

Senator CANAVAN: Do you believe that you are adding an objective here to your policy approach in that, typically, you would be just looking at the economy as a whole but now you are looking at the conditions in particular markets. Given you have another objective and you need another tool, as per the Tinbergen Rules, are these additional tools to meet an additional objective?

Dr Edey : I think that is an overly simplistic way of looking at it. We have had a financial stability objective for a long time. It has been well recognised. The overall objective here is not some new market that we are suddenly targeting; it is risk management. We are saying that a risk has emerged that is material for the economy as a whole.

Senator CANAVAN: In August this year in your Statement on Monetary Policy I think there are comments you have made over the last few years that we have a mining slowdown, and one of the ways that we hope to perhaps pick up the slack is through increased construction activity, particularly housing construction. Briefly, you said in August:

Mining investment could decline more sharply than anticipated. On the other hand, it is possible that consumption and non-mining business investment could, in time, be stronger than expected. In particular, an extended period of low interest rates could lead to further strength in the housing market and stronger-than-expected growth …

I am a little bit confused. On the one hand you are seeking here, hoping almost, that there would be a pickup in investment in housing to pick up that slack, and on the other hand we are saying that it is too much and we need to slow it down.

Dr Edey : Those comments are referring specifically to house-building activity, and you are quite right in saying that we saw benefits in upswing in that market, and we have been seeing that. What we have been mainly talking about today are prices and speculative activity and it is possible for there to be too much of that.

Senator CANAVAN: Sure, but, Dr Edey, I presume you are a PhD in economics and you probably understand the Q theory of investment. To get more investment in new stock of capital you need a higher price for the capital firstly, don't you?

Dr Edey : Yes, certainly, but it is possible for these things to go too far.

Dr Ellis : It is worth pointing out that the activity in investor lending is primarily for established dwellings. We can split it out between newbuild and established.

Senator CANAVAN: Sure, but my point is that there our second-order effects by increased investment in established dwellings.

CHAIR: Can we just touch on that again?

Dr Ellis : The confusion in the terminology, which Malcolm made clear, is that the Statement on Monetary Policy is referring to housing construction and the concerns that we have are about investor purchases, buy-to-let purchases of primarily established dwellings.

CHAIR: Going back to what Dr Edey said earlier, is it fair to say that part of what you are going to try is to find a way to target. You are obviously conscious of and talk about supply. It is a longstanding position of the RBA that there is a supply problem. We are looking at this as part of our inquiry. There are councils, local governments and a whole PhD thesis in what is wrong with the Sydney and New South Wales supply of the housing market. Are you saying that, with the tools you are looking at, you are very conscious of the fact that you do not want to dampen investment in new supply?

Dr Ellis : We would try to avoid that to the extent possible.

CHAIR: You believe that is possible in targeting?

Dr Ellis : We will have to see. Again, those tools are not been designed. Malcolm might want to expand on that.

Dr Edey : I think the thing to say is that there is a supply response actually happening at the moment. We are getting an upswing in house-building activity; it is a good thing.

CHAIR: Of course.

Dr Edey : What we are aiming to do is to rein in growth in financing to the highly speculative part of the market, which is where the risk concentrations are being built up.

CHAIR: I am sure you are well aware of the public criticism or concern that, in tackling the speculative investor market, you run the risk of an unintended consequence of actually stymieing investment in new housing stock. It is hard enough to get investment in housing stock because of all the supplier, council and local government problems in Sydney. You are saying you are aware of that concern, but you have not yet said that your tool will be able to address that.

Dr Edey : All I can really say is that we are aware of that general issue and we need to be mindful of it. I talked about imbalance because we are not trying to kill the investor market, we are not trying to kill construction activity or anything like that. What we are trying to do is mitigate risk in an area where excessive risk might be.

CHAIR: That is the danger of an unintended consequence, isn't it?

Dr Edey : That would be an unintended consequence if the measures that end up being taken are too severe, so we need to be careful to make sure that the measures we do take are proportionate to the risks that are out there.

Senator McLUCAS: Forgive me if you have already answered this question: what is the time frame of the negotiations or discussions between yourselves and APRA around what will be decided?

Dr Edey : I expect there will be at least a preliminary announcement before the end of the year.

CHAIR: Before the end of the year?

Dr Edey : Yes.

Senator McLUCAS: Thank you very much.

CHAIR: I thank Dr Edey and Dr Ellis for making themselves available. I very aware that it was incredibly short notice to appear, and once again the RBA has made themselves available to the Senate Economics Committee. On behalf the committee we really appreciate that.

Committee adjourned at 10:07