Feeling No Pain
DG APRA a few years ago had launched a project called Project Panama which required all of the home loan lenders in Australia to look at a stressed scenario which involved increasing the likelihood of somebody defaulting by a factor of 6 and applying a 30% reduction in house prices to the portfolio.
MP Which is a pretty disastrous sort of scenario.
DG It’s a very extreme scenario and in fact the only other place that I could find where anything remotely approaching that had happened was the UK in the early 90s so it can happen but it’s a very extreme scenario. Having said that though we wanted to understand on the basis of the Basle-II principles, what would that mean, not just at portfolio level but could we actually apply those stressed risk factors and that decrease in property value at an individual level and assess for every single one of our customers what their likelihood of default would be, and if they did default, what would be the amount of loss. So that’s what we did, and the big change of course, is that instead of assessing the impact at a portfolio level because we’ve now done the work as an individual customer level, we can begin to understand some patterns in our customer base that suggests to us that the impacts of a major macro-economic change would in fact not be uniform across all of our customers.
MP Well what were the headline findings. What were the surprises.
DG The first surprise I have to say is that the level of loss that we were expecting to see was substantially less than the level of loss that we had calculated using the previous portfolio approach. So I think the number that we communicated into the market is around about .25 of 1 percent of our total book so we were very surprised that the level of predicted loss was as low as that.
MP Why, why was it so low.
DG The main reason why it was so low, is that the driver of loss is more related to the equity position that you have and if you apply a 30% decrease in home value, then if your home has appreciated by 30% then the amount of loss is zero but obviously if you are in the position where you don’t have that amount of equity in your home, then that’s where the losses kick in. The vast majority of the Commonwealth Bank’s book is at a loan to value ratio very very different from the 30%.. 30% buffer so we’ve got a lot of customers that have had loans for quite some time with us and with property appreciation most people are sitting on some equity.
MP Does that mean the bank can lend to more risky customers knowing you’ve got so much margin of the overall book, well you have to stick with individual judgements, not the portfolio judgements.
DG We actually stick with an individual judgement. A lot of the tools that we’ve now got to make that individual judgement are better than tools that we have in the past, so it’s possible for us to extend loans into people that perhaps previously we may have said no to. But equally because of the same tools and the data that we’ve now got available we may be saying no to previously we might have said yes to, so I guess it’s a double edged sword.
MP That’s looking at it from the bank’s point of view. Looking at it from the customer’s point of view or potential investor’s point of view, what does it mean.
DG Yes. I think the customer experience through an economic downturn is going to be highly differentiated and that was the second major finding from the work that we did. The surprise to me was that.. and remember this is in the context of relatively low losses, but the surprise to me was that it wasn’t just the areas that you would expect to be [ ], so for example inner city apartment lending. There’s been a lot of publicity, a lot of press around, is that going to be where the pressure point is. It’s certainly the case that there would be some losses relating to recent loans and particularly for apartments where they haven’t appreciated in value but the real call out, the real interest was the impact of potential loss in rural and regional Australia and some of the impacts on mining towns, and what we did was to map the distribution of losses and while certainly some of the inner city areas, the areas with recent apartment growth came out as likely areas of pain. Likely areas of loss. It was also very clear that other parts of Australia that we hadn’t really thought about too much I have to say, did appear on the maps very clearly, and if you look at those areas then the principle reason for them appearing as points of pain is that those places haven’t experienced the rates of property value increase that we’ve seen in the major metropolitan areas so it comes back to an equity issue.
MP Is that a fairly extremely scenario. 30% fall in property prices. Did you also do it at less extreme levels and find similar results.
DG We did. We progressively stressed the portfolio. If you can imagine a little matrix where we’ve got the probability of default stressed by a factor of 2, 3, 4, 5 and 6 and then on the other axis of the matrix we’ve got the property value decrease which is 10%, 20% and 30% then the interaction with those two most extreme conditions, the 6 times default and 30% drop in property values, that was where we certainly did notice some impact, although again I say, it’s in the context of relatively low losses overall but when you only stress by 10% you actually find that the change from the conditions that we have today is really quite marginal. So for there to be a noticeable impact you would actually have to have a pretty major recession.
MP Does that mean that some of our concerns about the housing bubble bursting or misplaced.
DG I think you’ve got to be careful before you jump to that conclusion because you never quite know what interaction there’s going to be between an overheated housing market and the triggers that that might provide for macro-economic change so I think the way that the housing bubble has now come in, as a kind of soft landing, is really very encouraging. Because whilst there’s obviously residual risk, there always is risk when you [lead] it really would now take some very extreme circumstances for the kind of pain that I think people were very worried about at one point, to become evident.
MP So even some of the sectors of the market that have come down by 10 or more percent, have not been a problem for the banks.
DG In itself, a 10% drop in home prices isn’t going to lead to widespread distress and we can say that with a great deal of confidence because of the work that we’ve done. If you look at an interaction between a 10% decrease in home prices and other factors in the economy that lead to an increase in the likelihood of default, so for example much higher interest rates, much higher rates of unemployment, plunging consumer confidence.. if you have an interaction of those factors which leads people to default are a greater rate than they do at the moment, then obviously that does provide you with a slightly worse scenario but what surprised me was just how robust the housing book was and that you really had to try very hard, stress it to the max, and decrease property values to the max before you really got any sort of meaningful patterns emerging.
MP Why was that? Is that because even when people are in negative equity on their home loan they’ll still stick with it, or do we have such a real estate investor culture in Australia, there’s always a buyer.
DG I think you raised an interesting issue actually. I’m not sure this is directly answering your question but one of the areas that we did have a careful look at, and we’ve actually got to do some more work in, is in looking at people who are in negative equity and seeing how that actually interacts with their likelihood to default on a loan and obviously what we would like to be able to do as a bank is to provide proactive credit policies for people who might find themselves in those circumstances which means that the obvious answer is not necessarily, well let’s just put the keys through the door of the nearest Commonwealth Bank branch. So I think what we’re able to do is to try and move towards damping down some of the excesses that might otherwise emerge.
MP What do you mean.
DG I think there are things we can do. If you start with the customer and look at a customer who has actually been a good customer with you but because of economic change you suddenly find that the customer is now in difficulties.
MP Meaning unemployed, meaning unable'¦
DG Meaning unemployed, loss of job, faced with a home that perhaps didn’t have terribly much equity in it, then what we’ve been looking at is exploring a range of alternative options under that macro-economic downturn scenario where we can say well let’s have a look and we can take a longer term view so as a bank, we know that historically macro-economic downturns don’t last forever and we clearly provision against those and conventionally provisioning has been against a loss experienced by the bank, but what we’re now looking at are policies which would actually enable a customer to refinance. Potentially have an interest holiday. Potentially extend their loan over a longer time period and really try and look at a way that we can support a customer through difficult times which is clearly in the customer’s interests but it’s also an interest that all banks would have, so that you don’t have a collapse of the property market and everybody’s home is suddenly up for sale.
MP So it’s not a return to 1930.
DG Absolutely not. No.
MP From a [vulture] investor’s point of view, do you find that the areas that do hurt most, that do come down most, do they remain scarred. Are they a problem to investors?
DG I think a lot depends on your time horizon as an investor. If you consider that there are always going to be housing cycles and there are always going to be economic cycles, and if you take a long term view, then I think the housing market still represents an area that would be of interest. But if you are looking for a short term gain and if, in particular, that short term gain comes from capital rather than rental income, then the current conditions in some parts of Australia aren’t really conducive to generating that kind of gain. But even though there are properties which have decreased in value I don’t see that that’s necessarily going to provide a permanent scar and that there’s no chance of recovery. I think we’ve all seen in countries around the world at different points in time that the housing market does indeed go up and down, so I don’t think there’s any cause for concern that the housing market’s going to be impaired for life.
MP Again from the investors point of view, do those findings that parts of rural and regional Australia that haven’t had the equity growth, does that mean they’re also riskier places to buy in?
DG If you’re looking for appreciation in property values then you probably wouldn’t invest to get a capital return in rural and regional Australia. You would invest where there is population growth. Where there is very clearly economic growth and currently that would actually direct you more towards Western Australia and Queensland than anywhere else so if you want to have returns that are based on an increase in property values than you really ought to be looking to where the economic motors in those regional economies are running strongly. I think it’s fair to say that for vast areas across rural and regional Australia, those economic motors aren’t running as fast. Clearly you’re tied into an agricultural'¦ a primary production cycle economy and you tend to get relatively low levels of growth. And again seasonality with this is going to impact. So if you’re looking for investment opportunities it’s going to be the urban areas I’d have thought and it’s likely to be the urban areas in those states or sub-regions where the economy is growing fasted.
MP Having done that research, having found the areas that would hurt most, how does that impact on the bank’s lending policies. Are you less willing to lend?
DG The true answer to that is no. The current economic conditions that we have and the assessment of risk that we make in current conditions suggest to us that we should continue to provide the service that we do and obviously that consumers would like us to provide and I think that the difficulty in predicting when the next macro-economic downturn is going to occur means that probably the better way to go, is to continue to provide a service that’s in demand but to be very aware of what the consequences would be in the event of a downturn. And that means provisioning both from the bank’s capital and financial regulatory point of view but it also means provisioning from a social and customer point of view and I guess it’s into that second area that we’ve suddenly become much more active and interested as a result of the findings of the research.
MP So lend to someone in that non growth rural regional area, you’re not more likely to say look we only want to lend 80% of valuation, not 85.
DG We’re not.. we’re not differentiating lending policy on the basis of the research results that we’ve found. We’re making risk assessments that are appropriate to the current environment.
MP Does this all add up to the strongest rebuttal yet of the Economist magazines doom and gloom about the housing bubble that is all going to go to hell in the house with negative equity.
DG Well I think it certainly provides context and The Economist magazine doesn’t have access to the facts and figures that we have access to. And having looked really hard at the underlying characteristics of the housing book that the Commonwealth Bank has, obviously I can’t speak for other banks, but certainly on the basis of what we see it would require a very extreme combination of circumstances to trigger even a mild feeling of pain across our customer base and I can’t imagine what circumstances we would have to model to bring about the kinds of scenarios that I have read about through The Economist and perhaps another commentator.
MP So you’re really talking about a worldwide depression, not a recession. That happens we’ll have so many troubles housing will be the least of them.
DG Well I think that’s right. I mean it’s interesting to be able to say we now have an objective means by which we can come to these conclusions so these are conclusions.. the conclusions of the research are conclusions that are built on facts. It’s very easy to try and build conclusions that are based on opinion but I have to say I’d far sooner have my views grounded in fact.
MP The flip side of it, the reverse, given a mild downturn, which areas fared best. Which areas felt no pain or very little pain.
DG The areas that feel least pain are areas where there is very little risk of default and where there is substantial equity that’s been built up in the housing stock so those areas typically are stable, settled, middle class and upper middle class suburbs. And that is very clearly where there is least pain felt.
MP That research is about the safety of the bank and the comfort level of the bank. Out of it does there come anything about the comfort of investing in housing.
DG I think there is. We obviously didn’t direct the research that way but I think the main comfort that comes out of the work that we’ve done is in the generally low level of loss that we would expect to see from the housing book even under extreme circumstances. So as far as investment is concerned if our modelling is correct and I believe it is, then the amount of risk which an investor would incur even under conditions of extreme downturn are relatively low. That’s not to say they’re non-existent but the kind of risk is relatively low. The issue for an investor if we were to move into an economic downturn would be to understand that the type of investment actions that he or she may take in that emerging scenario where there’s a recession around the corner, there would have to be some very careful thought given as to where the losses are most likely to arise from and to tailor your investment actions accordingly.
MP So what’s the implication for mortgage insurance? That it’s not as necessary. As perhaps was previously thought?
DG I don’t think we’re going to change our policy on mortgage insurance but the essence of a mortgage insurance business is to do reasonably well. Sometimes very well in good times but the payments that you have to make in the bad times is such a drain on capital that you’ve got to have a lot of good years to enable a mortgage insurer to weather any storms. And I guess that’s a trade off which mortgage insurance companies have got a better handle on than I have. But certainly we have been surprised at the relatively low level of losses and I’ve no doubt that that’s something that we’ll be having conversations about with mortgage insurance providers.
MP This is a new area of research. New modelling. New matrix. Has it extended to other parts of the bank’s book. To margin lending. Other products?
DG No, it’s a very interesting point. That’s work in progress. At the moment we’re in the bank so there’s a section of my team that’s now looking at what is likely to happen in terms of personal loan and credit card losses. Margin lending. So all areas of lending, not just the traditional home loan. And we expect to have that work completed in the next month or two and that again is going to give us I think some very useful insights particularly from a customer perspective as to what we would find, not just from a banking point of view but what are we going to see from a social perspective. And that’s going to be really important in terms of driving policy.
ends