Average property prices in the UK are at historic highs and this is not good news for future house price gains.
Why? The short answer is that trees don’t grow to the sky. The longer answer is a bit more complicated but very interesting.
But first, a question: Why am I – a stock market investor – talking about house prices?
The answer is:
- I’m interested in market valuations in general and that extends to the property market.
- I think stock market investors can learn a lot from property investors.
- The housing market valuation methods which I cover in this article are directly applicable to the stock market as well, with just a couple of minor tweaks.
Okay, enough with the introduction; let’s take a look at the UK housing market:
House prices always go up, right?
Here’s a chart of average house prices for England and Wales based on data from the Office for National Statistics and Halifax (okay, England and Wales aren’t the whole UK, but they’re the bulk of it and it’s the data the ONS had on their website):
As you can see, average property prices in the UK have gone up a lot, from £27,000 in 1983 to £225,000 in 2017. That’s an average annual gain of about 6%.
A lot of that is due to inflation, but people rarely adjust for inflation when calculating investment returns. Most people would just look at the chart above and say “see, property goes up by 6% almost every year.”
So property prices have gone up a lot, but what about their value? In other words, I could pay someone £100 for a peanut, but that doesn’t mean the peanut is worth £100.
Price and value are two entirely different things. However, there is obviously a close relationship between the two (which is why people don’t usually pay £100 for a peanut).
How do we measure the true value of property or its intrinsic value? The simple truth is that we can’t.
There are too many factors and value is ultimately a subjective thing. But we can still make a pretty sensible estimate based on historic norms.
For example, house prices in a country are obviously going to be related to the earnings of the people who live in that country because they’re the ones who pay mortgages to banks or rent to landlords.
Average house prices share an unbreakable bond with average earnings
Here’s a chart of average earnings in England and Wales:
Like house prices, average earnings have gone up and up and up. Again, inflation is a big part of this, but even so, the average gain of 4% per year since 1983 outpaces inflation by a reasonable margin.
So what exactly is this relationship between house prices and earnings?
In simple terms it’s just the ratio between average house prices and average worker earnings.
What’s interesting about this ratio is that it’s both mean-reverting (i.e. tends to return to its historic average every few years) and range-bound (i.e. exists in a relatively tight range rather than flitting between one and one million).
Why is the house price to earnings ratio mean-reverting and range-bound?
Imagine that the average UK worker earns £30,000 and the average house costs £30,000. It would be very easy for most people to own more than one house. You could buy a house, pay it off in ten years and then buy another one to rent out or put the teenage kids in. That would push up demand, which would push up prices.
At the other end of the scale, almost nobody earning £30,000 per year could pay off a £3,000,000 mortgage (100-times earnings) in their lifetime. Banks don’t like to lend when the borrower can’t pay back the loan, so demand for houses would fall and that would push down house prices.
Obviously there’s a bit more to it than that, but that’s the basic picture.
In fact, we can do much better than simply saying that the house price to earnings ratio stays in a range between one and one hundred. That’s because the amount people are willing to pay for a house relative to their earnings, or to pay to rent one, doesn’t change by anything like a factor of one hundred.
Here’s a chart showing the house price to earnings ratio since 1983, based on the same data as the previous two charts:
Since 1983, the house price to earnings ratio has an average value of just over five, a minimum value of just over three and a maximum value of just under eight. Today it sits at an all-time high of 7.8.
This is all interesting stuff and it means we can begin to build up a picture of where house prices might go, if the house price to earnings ratio reverts to its historic average as it has always tended to in the past.
But before we think about where house prices might go, let’s have a look at where they are today.
UK house prices are in a bubble
I think the conclusion that house prices are in a bubble is pretty obvious, but the argument still needs fleshing out.
Here are a few definitions for the house price to earnings ratio:
- Depression PE = 3.3 – This is about where the market fell to in the early 90s. Prices need to increase by 50% to go from depression to fair.
- Fair PE = 5.0 – This is close to the historic average of the last 35 years.
- Bubble PE = 7.5 – This is close to the highs of the late-2000s housing bubble and the subsequent era of super-low interest rates. Prices need to increase by 50% to go from fair to bubble.
I think that’s a pretty reasonable guestimate of reality. Some people might argue that the current price / earnings ratio of 7.8 is sustainable, or that it could go even higher to ten, or 20, or who knows where. But I think they’d be wrong.
The UK housing market currently has a number of massive tailwinds, none of which are likely to be sustainable over the long-term. The main ones are:
- Record-low interest rates
- Massive amounts of money pumped into the system by the Bank of England in an attempt to motivate banks into lending more money to more people
- Crazy government schemes like Help to Buy which appears to have been created with the singular goal of pushing up house prices
When these tailwinds are removed I think it’s highly likely that the house price to earnings ratio will do what it’s always done, and that is to revert to its historic average on the back of house price declines.
Of course I could be wrong, but history suggests I’m not (although I have no idea when any of this will happen).
For example, here’s a chart of house prices laid over a “rainbow” of house price PE ratios, where red is expensive, yellow is fair and green is cheap (relative to historic norms):
Over very long period of time the ratio of house prices to earnings stays within a relatively narrow band. When it goes far above the average, it falls back. When it falls far below average, it recovers.
I think that pattern is very likely to continue and that the pull of mean-reversion is going to outlive the push of short-term government stimulants.
The upshot of all this is that as far as I’m concerned, the medium-term future for UK house prices does not look good.
UK House prices are likely to stay flat for at least two decades
In 2007, average house prices reached £174,000. Today, more than ten years later, the average stands at £225,000.
That’s a gain of 30% in ten years, or less than 3% on an annualised basis. If you take inflation into account then the average house price has stayed virtually unchanged in more than a decade.
If you look at house prices through the lens of housing PE mean reversion then this shouldn’t surprise you at all.
In 2007 house prices were at record highs relative to earnings and the gap to “fair value” was as large as it had ever been.
To a mean-reversionist the likely outcome was a big decline in house prices over the following decade, or at the very least a long sideways market while inflation pushed up the “fair” level of house prices.
What we got was a sideways-to-slightly-upwards market, largely thanks to those massive government-driven tailwinds.
This means that the gap to fair value hasn’t diminished at all, so as far as I’m concerned price declines or zero growth are still by far the most likely outcome.
If we look at a ten-year time frame then there are a few things we can say:
- If earnings increase by their historic average of 4% per year (a reasonable if somewhat optimistic assumption) then average worker earnings will grow to £42,000 by 2028.
- If the house price to earnings ratio reverts to its average value of five, then house prices will fall to £214,000 by 2028.
If that happens then we will have the mid-2000s credit boom to thank for two decades of almost zero price growth in the housing market.
Just as we have the late-90s tech boom to thank for two decades of almost zero price growth in the stock market.
On the plus side, at least ordinary people will be able to afford ordinary houses once again.