Dear Property Investor…Your Best Warning!
The Australian residential property market has had a spring in its step recently. Auction clearance rates have been healthy, and rising prices have prompted media commentary on the possibility of an emerging housing “bubble”.
I’m no property expert, and so it may be wise to avoid putting an oar in on this debate, but let’s put wisdom aside for a moment and think about whether the application of value investing principles can add anything to a discussion on house prices.
We’ll start by assuming that residential property – at least the investment kind – derives its value in the same way as other financial assets (by producing cashflows), and obeys the same fundamental financial laws. This may be too much to assume for owner-occupied property, but should be reasonable in the case of investment property.
Having made that assumption, and noting that property income is fairly predictable, the valuation exercise can be approached in a fairly straightforward way. We just need to come up with three critical numbers:
• The discount rate, or rate of return required by an investor
• The net rental income generated
• The long term rental growth rate
Having estimated these numbers, we can plug them into a simple formula that will calculate the present value of the cashflows into perpetuity – following the same principles one might use to value a business.
As an aside, I note that a lot of commentators assert that investment property is justified on the basis of capital growth rather than rental yield. For me, the issue with this line of thought is: how do you estimate capital growth? To a value investor, value needs to be anchored on future cashflows (rents), not extrapolated from past price growth. In the same way that share price growth in excess of profit growth is unsustainable long term, it seems that property price growth in excess of rental growth should also be unsustainable long term.
Moving on, let’s try to estimate our valuation assumptions, starting with the discount rate. You should feel free to substitute your own estimates here and reach your own conclusions. The thing I want to illustrate is how you can go about doing the calculation.
A discount rate reflects the rate of return an investor needs to compensate them for the risk of investing in a particular asset. A good way to estimate it might be to start with the ten year government bond rate (a proxy for the rate that applies when there is no risk – currently around 4% p.a.), and add a risk premium.
In the case of equities, the risk premium is commonly thought to be around 5-6%. I think a case can be made for a lower risk premium for property, given that it has a more stable profile, but at the same time we should consider that property is a much less liquid asset than equities, and investors should demand some compensation for this. For the sake of argument, let’s choose a risk premium of 4%. This gives us a required rate of return of around 8%. Again, feel free to adopt your own assumptions.
Net rental yield is the next piece of information we need. Here I’ll refer to some data published by rp-data, set out in the chart below.
The chart shows a gross rental yield for Australian capital cities of around 4.3% p.a. currently. Note that in times gone by this figure was significantly higher (as was the 10 year bond rate).
To get to a net rental yield we need to subtract all the expenses that an investor must incur in generating the gross rent, including property management fees, maintenance, insurance etc. I don’t have a good reference for this figure, but a reasonable estimate may be around 1.0%. For now let’s adopt that figure, which gets us to a net rental yield of 4.3% – 1.0% = 3.3%.
The final figure we need is the long term growth rate in rents. The rp-data chart above sets out the long term history of rental growth rates. Having regard to this history, it seems like a figure just above 3% p.a. might be a reasonable estimate. For now let’s assume 3.2%.
Combining these numbers into a valuation can be done as follows: we divide the annual rent earned on a property by a divisor, which is calculated by subtracting the long term growth rate from the discount rate, as set out below.
Value = Net Rent p.a. /(Discount rate – Growth rate)
For the financially minded, this is the formula for calculating the present value of a stream of cash flow that grows at a fixed rate in perpetuity.
If a property generates say $10,000 per year of net rent, we would calculate its value as $10,000/(8% – 3.2%) or $10,000/4.8%, which is equal to around $208,333.
If that property can be bought today at a net yield of 3.3%, then that would imply that the market price of the property today is $10,000/3.3% = $303,030. This is clearly a significantly higher number than the valuation of $208,333 that we just calculated. On the basis of the discount rate and long term growth rate we assumed, buying property on a net rental yield of 3.3% appears hard to justify.
This analysis may indicate that the last few decades of growth in property prices is not sustainable (recall the declining yield from the chart above) and some mean reversion is likely going forward. If this is correct, it may pay to be cautious about buying on the basis of a continuation of assumed capital growth.
Of course, the conclusions you reach with this approach are sensitive to the assumptions you put into it, and the purpose here is not to argue that property is overpriced – rather, it is to set out a framework that allows some basic assumptions to be converted into a fundamental value. By doing this, you can focus attention on the things you need to believe in order to conclude that there is long term fundamental value underpinning a decision to invest.
Tim’s proposed framework may help explain the reason why the world’s longest study of house price changes reveals that houses prices largely track inflation.
It’s my favourite academic study of house prices. The Herengracht canal in Amsterdam has been a favoured strip of real estate in the city since the 1620s when wealthy spice and slave merchants showed a penchant for canal-side living. Finance professor Piet Eichholz of Maastricht University data-mined extremely long term Dutch record keeping to construct the Herengracht house index. It covers inflation adjusted real estate prices from the construction of the Herengracht in the 1620s up to 1975, after which it was extended to 2008.
Over a 380 year period – commencing with the Dutch ‘Golden Age’ – real (i.e. inflation adjusted) house prices have only doubled, which corresponds to an annual average price increase of something like 0.1%. The conclusion then is that despite short-term rises and falls, prices roughly follow inflation.
What are the implications of both Tim’s and Eichholz’s work? Well, for starters, if you buy a property and subsequently enjoy the benefits of house prices that rise much faster than inflation over a decade or two, you should consider reducing your exposure ahead of a reversion. Alternatively, if you haven’t bought, then rather than relying on an already established area to increase with a rising tide, the general areas to invest in are those where gentrification and transformation are possible within the next five to ten years. Roger M.