Hi Michael,
I wanted to take some time out today to provide you with a quick summary of a few of the issues I’m seeing play out in the market that could potentially affect not only your investment portfolios, but also the valuation of your business.
While we arguably have a bit more flexibility and time on our side insofar as the portfolio-side of the equation goes, I think it’s important to cover off a few issues with regard to your business valuation, and more precisely highlight some of the risks and opportunities with your exit from the business.
First, I must restate that the valuation of your business has, to date, been pretty poorly constructed. So far the evidence would suggest your company’s accountant is out of touch with standard valuation processes and metrics. But putting that to one side, what I want to cover off today is the state of the property market, and how this might affect the company’s growth trajectory in the years ahead. Bear in mind that I’m taking a very macro approach to this; in other words, I’m trying to identify trends and likely outcomes at a market or sector level, and not on an individual company-level. As is always the case, with any change to the market/business environment, there will be those who perform better, and those who perform worse.
Third, and final caveat before I get into the data, the following issues/concerns do not represent a full list of the data I track. However, they provide some good reference points for both the demand and supply-side pressures we are seeing across markets. There is quite a lot of detail behind the headlines, so if there is any aspect you would like to discuss or explore in more detail, please let me know.
I wanted to take some time out today to provide you with a quick summary of a few of the issues I’m seeing play out in the market that could potentially affect not only your investment portfolios, but also the valuation of your business.
While we arguably have a bit more flexibility and time on our side insofar as the portfolio-side of the equation goes, I think it’s important to cover off a few issues with regard to your business valuation, and more precisely highlight some of the risks and opportunities with your exit from the business.
First, I must restate that the valuation of your business has, to date, been pretty poorly constructed. So far the evidence would suggest your company’s accountant is out of touch with standard valuation processes and metrics. But putting that to one side, what I want to cover off today is the state of the property market, and how this might affect the company’s growth trajectory in the years ahead. Bear in mind that I’m taking a very macro approach to this; in other words, I’m trying to identify trends and likely outcomes at a market or sector level, and not on an individual company-level. As is always the case, with any change to the market/business environment, there will be those who perform better, and those who perform worse.
Third, and final caveat before I get into the data, the following issues/concerns do not represent a full list of the data I track. However, they provide some good reference points for both the demand and supply-side pressures we are seeing across markets. There is quite a lot of detail behind the headlines, so if there is any aspect you would like to discuss or explore in more detail, please let me know.
1: History is unlikely to repeat
To understand how we see the property market developing over the next decade, it’s probably a good idea to take a look at where we are, and how we got here.
Over the last 100 years Australia has had the best performing residential property market in the entire world, with house prices achieving long-term capital growth equivalent to 0.9% per annum above the rate of inflation. Most Australians find this number ridiculous, as we’ve been taught to believe that 5% - 15% per annum capital growth is “normal”. That’s not all that surprising, as the bulk of price growth has occurred between a period of hyperinflation (1970s) and massive expansion in household debt (since 1990). The other key growth period was after removal of WW2 rent controls.
To understand how we see the property market developing over the next decade, it’s probably a good idea to take a look at where we are, and how we got here.
Over the last 100 years Australia has had the best performing residential property market in the entire world, with house prices achieving long-term capital growth equivalent to 0.9% per annum above the rate of inflation. Most Australians find this number ridiculous, as we’ve been taught to believe that 5% - 15% per annum capital growth is “normal”. That’s not all that surprising, as the bulk of price growth has occurred between a period of hyperinflation (1970s) and massive expansion in household debt (since 1990). The other key growth period was after removal of WW2 rent controls.
2: Debt got us here: it will also take us down
Without getting too caught up in the math, it’s worth considering that since 1990 house prices (adjusted to inflation) have risen about 150% compared to a 220% increase in household debt. If we compare this against total household wealth, we see that nearly all of the change in house prices is attributable to debt (I will come back to this a little later; there is some very important data hidden in these numbers).
Of course debt isn’t necessarily a bad thing. If you can borrow at 5% for an asset yielding 3% and inflation of 2%, then you are more-or-less breaking even. It might seem counterintuitive at first, but the real problem occurs when interest rates and inflation fall, because this tends to flow through to increased borrowings and prices adjusting to eliminate the yield arbitrage of the asset. In theory, this shouldn’t be a huge problem as investors should also factor in lower inflation into their models, but when it comes to retail markets (and sectors like residential property) people tend not to behave this way. This causes major problems in the way that assets are valued.
This is a problem insofar as market efficiency goes, but where it really gets messy is when inflation turns out to be lower than they expected, and/or interest rates and/or risk margins rise.
Without getting too caught up in the math, it’s worth considering that since 1990 house prices (adjusted to inflation) have risen about 150% compared to a 220% increase in household debt. If we compare this against total household wealth, we see that nearly all of the change in house prices is attributable to debt (I will come back to this a little later; there is some very important data hidden in these numbers).
Of course debt isn’t necessarily a bad thing. If you can borrow at 5% for an asset yielding 3% and inflation of 2%, then you are more-or-less breaking even. It might seem counterintuitive at first, but the real problem occurs when interest rates and inflation fall, because this tends to flow through to increased borrowings and prices adjusting to eliminate the yield arbitrage of the asset. In theory, this shouldn’t be a huge problem as investors should also factor in lower inflation into their models, but when it comes to retail markets (and sectors like residential property) people tend not to behave this way. This causes major problems in the way that assets are valued.
This is a problem insofar as market efficiency goes, but where it really gets messy is when inflation turns out to be lower than they expected, and/or interest rates and/or risk margins rise.
3. Interest rates & risk premiums are going to go higher
Inflation is pretty self-explanatory. Interest rates and risk margins are far more interesting and (in absence of hyperinflation) more important in the context of asset pricing.
Thinking back to the expansion in household debt. Since 1990 there have been a few factors driving this growth, including tax policy (negative gearing, CGT exemptions), lower interest rates and a more toward interest-only loans. This has meant that interest repayments as proportion of household income has stabilized at around the 11% - 12% mark.
A few reasons this is almost certainly going to rise:
1: Australia’s banks have been under pressure from the Productivity Commission, APRA, and more recently the Royal Commission, to improve risk control measures within their loan portfolio. As a result we have already seen all major banks, with the exception of Westpac (no explanation why), substantially reduce the proportion of interest-only loans approved (both new and refinanced). This increases the total repayments due on the loan.
2: For every dollar that banks lend they must retain a certain amount of capital on their own balance sheet. This is usually funded by cash deposits, proceeds form Bond issuances, and foreign debt. Fortunately for consumers, the proportion of funding from foreign sources has been reduced over the last 2 years, but even so we should expect to see their cost of funding increase, which will be passed onto customers. Hard to tell how much effect this will have, but with rates being increased across North America and Europe, something around 0.2% p.a. seems about right.
3: Risk margins are also expected to increase. With very low inflation and wage growth it is likely we will see a natural decline (and probably reversal) in credit growth. Unlike the behavior we see in retail markets, risk markets tend to be quite effective at anticipating change. We know from the empirical evidence that an increase in interest rates is strongly correlated to an increase in non-performing loans. Take the GFC as an example: while central bank rates were being slashed risk margins continued to increase through to 2010, shortly followed by peak defaults. Then the margin decreased rapidly and defaults by 1/3. Given the extraordinary levels of household debt and mortgage stress we should expect to see a considerable increase in risk margins. If people were sensible, this would be 2% - 3% higher than it is today, but as nearly all of Australia’s residential property loans are non-recourse, this could be much lower (say 1% to 2%).
These factors in combination would suggest an increase to borrowing costs somewhere in the order of 1.2% to 2.5%, an increase of 25% - 50%. At a market-wide level, this equates to an increase in total repayments of 14% - 36%.
Inflation is pretty self-explanatory. Interest rates and risk margins are far more interesting and (in absence of hyperinflation) more important in the context of asset pricing.
Thinking back to the expansion in household debt. Since 1990 there have been a few factors driving this growth, including tax policy (negative gearing, CGT exemptions), lower interest rates and a more toward interest-only loans. This has meant that interest repayments as proportion of household income has stabilized at around the 11% - 12% mark.
A few reasons this is almost certainly going to rise:
1: Australia’s banks have been under pressure from the Productivity Commission, APRA, and more recently the Royal Commission, to improve risk control measures within their loan portfolio. As a result we have already seen all major banks, with the exception of Westpac (no explanation why), substantially reduce the proportion of interest-only loans approved (both new and refinanced). This increases the total repayments due on the loan.
2: For every dollar that banks lend they must retain a certain amount of capital on their own balance sheet. This is usually funded by cash deposits, proceeds form Bond issuances, and foreign debt. Fortunately for consumers, the proportion of funding from foreign sources has been reduced over the last 2 years, but even so we should expect to see their cost of funding increase, which will be passed onto customers. Hard to tell how much effect this will have, but with rates being increased across North America and Europe, something around 0.2% p.a. seems about right.
3: Risk margins are also expected to increase. With very low inflation and wage growth it is likely we will see a natural decline (and probably reversal) in credit growth. Unlike the behavior we see in retail markets, risk markets tend to be quite effective at anticipating change. We know from the empirical evidence that an increase in interest rates is strongly correlated to an increase in non-performing loans. Take the GFC as an example: while central bank rates were being slashed risk margins continued to increase through to 2010, shortly followed by peak defaults. Then the margin decreased rapidly and defaults by 1/3. Given the extraordinary levels of household debt and mortgage stress we should expect to see a considerable increase in risk margins. If people were sensible, this would be 2% - 3% higher than it is today, but as nearly all of Australia’s residential property loans are non-recourse, this could be much lower (say 1% to 2%).
These factors in combination would suggest an increase to borrowing costs somewhere in the order of 1.2% to 2.5%, an increase of 25% - 50%. At a market-wide level, this equates to an increase in total repayments of 14% - 36%.
4. Financial Stress
The obvious question: “can people afford it?”
A few things to consider here: The high-level data suggests that even with interest rates are record lows, around 30% of households are in mortgage stress, and just under 6% are at risk of default (also important to note that about 60% of non-homeowners are experiencing rental stress, so there is very little ability for rental costs to be increased to close the gap). Intuitively, you would expect any change to rates to have the greatest impact on those with lower incomes.
This makes sense, and I have written about this extensively since 2011, so I won’t get too carried away with it (basically we can work this out by adjusting gross disposable income for core living expenses, then working out the max level of sustainable debt through backward propogation). However, quite incredibly, this isn’t where we see the greatest problem. If we categorise the data by household income, debt and repayments, we find that the top end of the market is actually most sensitive to changes in interest rates. Of those with a mortgage, the top quartile by household income use about 40% of their disposable income for mortgage repayments, and the second quartile about 30%. The median is about 25%. Mortgage Stress is typically defined as 30% of household disposable income.
On a more positive note, low interest rates (and interest only loans) have resulted in a greater proportion of people contributing extra toward their mortgages. About 35% of mortgage holders have built up a buffer of 2 or more years. That’s good for them, but 50% of people have less than 6 months runway and about 25% are ahead by less than one month.
We also look to serviceability data and savings ratios. Earlier this week (along with GDP data) the ABS released savings data for the quarter ended December 2017, which showed that savings remain very weak.
Earlier when I mentioned debt to disposable income ratios I said there was something else we needed to consider, and that is the proportion of homeowners with a mortgage, relative to households that are either renting or own their home outright. We need to include this in our calculations as the debt-to-disposable income ratio is at a market level, but what we’re most interested in is the situation facing mortgage holders. Zooming in on the last 5 years, we see that the number of Australian households has grown by about 970,000. We have seen 410,000 rise in the number of households renting, 600,000 increase in those with mortgages, and a 40,000 reduction in households that own their property outright. Furthermore, when we look at this data by age category, we see significant reduction in the level of debt held by those under 35, and an increase in debt held by those in the 45-65 and over 65 age categories. The clustering of this data points to the debt problem being worse than expected, as it is less evenly spread across the population.
Together this paints a fairly bleak outlook on the demand-side.
The obvious question: “can people afford it?”
A few things to consider here: The high-level data suggests that even with interest rates are record lows, around 30% of households are in mortgage stress, and just under 6% are at risk of default (also important to note that about 60% of non-homeowners are experiencing rental stress, so there is very little ability for rental costs to be increased to close the gap). Intuitively, you would expect any change to rates to have the greatest impact on those with lower incomes.
This makes sense, and I have written about this extensively since 2011, so I won’t get too carried away with it (basically we can work this out by adjusting gross disposable income for core living expenses, then working out the max level of sustainable debt through backward propogation). However, quite incredibly, this isn’t where we see the greatest problem. If we categorise the data by household income, debt and repayments, we find that the top end of the market is actually most sensitive to changes in interest rates. Of those with a mortgage, the top quartile by household income use about 40% of their disposable income for mortgage repayments, and the second quartile about 30%. The median is about 25%. Mortgage Stress is typically defined as 30% of household disposable income.
On a more positive note, low interest rates (and interest only loans) have resulted in a greater proportion of people contributing extra toward their mortgages. About 35% of mortgage holders have built up a buffer of 2 or more years. That’s good for them, but 50% of people have less than 6 months runway and about 25% are ahead by less than one month.
We also look to serviceability data and savings ratios. Earlier this week (along with GDP data) the ABS released savings data for the quarter ended December 2017, which showed that savings remain very weak.
Earlier when I mentioned debt to disposable income ratios I said there was something else we needed to consider, and that is the proportion of homeowners with a mortgage, relative to households that are either renting or own their home outright. We need to include this in our calculations as the debt-to-disposable income ratio is at a market level, but what we’re most interested in is the situation facing mortgage holders. Zooming in on the last 5 years, we see that the number of Australian households has grown by about 970,000. We have seen 410,000 rise in the number of households renting, 600,000 increase in those with mortgages, and a 40,000 reduction in households that own their property outright. Furthermore, when we look at this data by age category, we see significant reduction in the level of debt held by those under 35, and an increase in debt held by those in the 45-65 and over 65 age categories. The clustering of this data points to the debt problem being worse than expected, as it is less evenly spread across the population.
Together this paints a fairly bleak outlook on the demand-side.
5. More stable demographics = bad news
On the supply side we have continued to see a lot of activity in high-density housing (apartments/townhouses). Comparing Victoria’s population growth against the number of new dwelling approvals we see a ratio of around 2:1. This compares to average household size of around 2.6. Though we have seen several decades of this ratio being above that over household size, this has largely been supported by the rapid reduction in average household size. AIFS research on this subject points toward household size either stabilizing or growing from this 2.6 rate, which may lead to an oversupply of housing (already at an aggregate level, Australia has a housing surplus).
On the supply side we have continued to see a lot of activity in high-density housing (apartments/townhouses). Comparing Victoria’s population growth against the number of new dwelling approvals we see a ratio of around 2:1. This compares to average household size of around 2.6. Though we have seen several decades of this ratio being above that over household size, this has largely been supported by the rapid reduction in average household size. AIFS research on this subject points toward household size either stabilizing or growing from this 2.6 rate, which may lead to an oversupply of housing (already at an aggregate level, Australia has a housing surplus).
6. Asset prices – what’s going on with property prices?
Normally assets are priced as a function of the future (lifetime) net cash flows adjusted to the investor’s discount rate (ie., as per standard DCF model). In the case of property this means we are looking at the expected rental return (net agent fees, vacancy, insurance etc), plus capital growth less depreciation.
The math for this is incredibly simple. I've provided a quick summary/explanation HERE. At the end of the day, relative to inflation, I'm expecting house prices to cool off by between 30% - 40%. There is a chance this might take the course of 10%-15% of falls with +20% of inflation.... who knows how or when this will play out.
Normally assets are priced as a function of the future (lifetime) net cash flows adjusted to the investor’s discount rate (ie., as per standard DCF model). In the case of property this means we are looking at the expected rental return (net agent fees, vacancy, insurance etc), plus capital growth less depreciation.
The math for this is incredibly simple. I've provided a quick summary/explanation HERE. At the end of the day, relative to inflation, I'm expecting house prices to cool off by between 30% - 40%. There is a chance this might take the course of 10%-15% of falls with +20% of inflation.... who knows how or when this will play out.
7. It’s not all bad news
This isn’t necessarily a bad thing for homebuilders, as over the longer term it would free up significant capital for new homebuyers to enter the market and put their money toward construction as opposed to land banking. Problem is we making it through the quiet part of the cycle when credit is tight and people aren’t building (result: builders scrambling for customers = tighter profit margins. It’s worthwhile researching the US experience post-GFC. Very tough period).
This isn’t necessarily a bad thing for homebuilders, as over the longer term it would free up significant capital for new homebuyers to enter the market and put their money toward construction as opposed to land banking. Problem is we making it through the quiet part of the cycle when credit is tight and people aren’t building (result: builders scrambling for customers = tighter profit margins. It’s worthwhile researching the US experience post-GFC. Very tough period).
8. Quick snapshot of the current market
In the last six months auction clearance rates have fallen sharply. If we take the clearance rate (ex vendor bid), the number of properties sold at auction is about 40% lower than the same time last year. Because of the way price data is reported it’s a bit less reliable, but we have seen prices soften over the last quarter. I have heard reports that the top end of the market is already recording falls of up to 20%.
About 10% of Australia’s workforce is employed by the construction sector. If you include building materials supply and manufacturing brings this number above 15%. If we see a downturn in the sector it’s likely to have far reaching consequences. Couple this with higher lending costs and low inflation and there is a better than even chance of dipping into recession.
In the last six months auction clearance rates have fallen sharply. If we take the clearance rate (ex vendor bid), the number of properties sold at auction is about 40% lower than the same time last year. Because of the way price data is reported it’s a bit less reliable, but we have seen prices soften over the last quarter. I have heard reports that the top end of the market is already recording falls of up to 20%.
About 10% of Australia’s workforce is employed by the construction sector. If you include building materials supply and manufacturing brings this number above 15%. If we see a downturn in the sector it’s likely to have far reaching consequences. Couple this with higher lending costs and low inflation and there is a better than even chance of dipping into recession.
9. Business valuation considerations (& concerns)
There are a number of ways this feeds into your business valuation, but the most crucial is the impact on your future pipeline. Even if your clients are funding their projects entirely with cash, there is nothing like a market correction to trigger a bout of cold feet.
To give some perspective, in the US housing starts fell from about 1.1 million in early 2008 to about 500,000 a year later. It took another 6 years to recover. You might like to skim through this for a few examples: http://builder-implode.com/imploded.html *Note: household debt in the US never got even close to where it is today in Australia.
The second consideration is calculating the business value. I have already criticized the valuation approached being used so I won’t go on about it, but it’s worth explaining that interest rates and inflation usually do have some influence on the sale price of the business. This is because whether we calculate on a multiple of EBIT or on a Discounted Cash Flow basis, we usually factor in a discount rate that incorporates the cost of borrowing. That’s because the value (or, in this context, the price) of the business should represent a fair rate of return that considers all risks and all other possible uses of your money. As interest rates and risk premiums rise, the value (price) should be expected to fall.
There are a number of ways this feeds into your business valuation, but the most crucial is the impact on your future pipeline. Even if your clients are funding their projects entirely with cash, there is nothing like a market correction to trigger a bout of cold feet.
To give some perspective, in the US housing starts fell from about 1.1 million in early 2008 to about 500,000 a year later. It took another 6 years to recover. You might like to skim through this for a few examples: http://builder-implode.com/imploded.html *Note: household debt in the US never got even close to where it is today in Australia.
The second consideration is calculating the business value. I have already criticized the valuation approached being used so I won’t go on about it, but it’s worth explaining that interest rates and inflation usually do have some influence on the sale price of the business. This is because whether we calculate on a multiple of EBIT or on a Discounted Cash Flow basis, we usually factor in a discount rate that incorporates the cost of borrowing. That’s because the value (or, in this context, the price) of the business should represent a fair rate of return that considers all risks and all other possible uses of your money. As interest rates and risk premiums rise, the value (price) should be expected to fall.
10. Now that I've got that off my chest...
In saying all this, you know your business better than anyone, so it’s up to you to make your own forecasts and assumptions. I just wanted to outline and explain some of the areas that have me concerned. It might also provide some comfort that leaving the business now isn’t all bad.
Finally and fortunately, much of what I’ve outlined above is usually ignored by nearly everyone so will hopefully this gives you plenty of time to make a clean exit at a fair price.
I know there’s a fair bit in there so hopefully this was not too much to digest. Happy to discuss further.
Cheers,
Joel
In saying all this, you know your business better than anyone, so it’s up to you to make your own forecasts and assumptions. I just wanted to outline and explain some of the areas that have me concerned. It might also provide some comfort that leaving the business now isn’t all bad.
Finally and fortunately, much of what I’ve outlined above is usually ignored by nearly everyone so will hopefully this gives you plenty of time to make a clean exit at a fair price.
I know there’s a fair bit in there so hopefully this was not too much to digest. Happy to discuss further.
Cheers,
Joel