When people pretend they can predict exact property price growth, there's a 0.00% chance they'll be right
Forecasters are setting themselves up to fail when they make very exact predictions about future price growth.
I've long been puzzled by economic forecasters predicting, for example, that Brisbane house prices will grow 7.4% in 2014-15 (not approximately 7% or 8%, mind you, but exactly 7.4%) and then exactly 7.5% in 2015-16 – as QBE has done with its Australian Housing Outlook report, with the help of BIS Shrapnel.
What micro-economic event will occur in 2015-16 to cause Brisbane's average price growth to limp 0.1 of a percentage point higher in that year, compared to the previous year?
Why will Melbourne prices grow exactly 2.2% in 2015-16 and then decrease exactly 0.7% in 2016-17?
What chance the forecasters will be right? I would suggest zero (and that's an exact figure I can predict with confidence).
Has this kind of forecast ever been exactly right in the past? No. I've checked and I can't find an instance of an accurate prediction.
So why do it? Why pretend you can predict city price growth two or three years into the future, down to the last decimal point, when it's not remotely possible to be right?
The most likely answer is: to drum up a bit of publicity, comfortable in the knowledge that most media outlets will publish this nonsense without asking a single question. And even more comfortable in the knowledge that no one will check in three years to see how accurate the forecasts were.
Louis Christopher of SQM Research commented earlier this year that organisations who make big claims or extreme forecasts for publicity reasons rely on "the community's collective amnesia" to cover their tracks.
In other words, they can relax, knowing that no one will bother to check on the results of their forecasts, much less hold them to account for them.
Even more alarming about this forecast report is that it is fundamentally based on data from the Real Estate Institute of Australia.
It lists Real Estate Institute of Australia (REIA) figures for the previous decade and then extrapolates forward, with predictions for the next three (financial) years.
The Melbourne figures are particularly rubbery. According to the REIA, Melbourne prices rose a tad under 20% in the financial year that ended on 30 June.
Such a figure is fanciful. All the other major research sources contradict it. RP Data says 9%, Australian Property Monitors says 10% and the Australian Bureau of Statistics says 9%.
Even more spectacular, the REIA claims Melbourne prices rose almost 27% in 2009-10.
As I have written previously in this column, the institute's figures for Melbourne growth are good for a laugh, but no one sensible gives them any credibility.
I've called the REIA to account numerous times in the recent past on the accuracy of its price data – and now the institute refuses to send me their figures. Fortunately, it's not difficult to get a copy of their reports from other sources.
The one thing that does make sense about the QBE report are the general themes, in terms of which locations are likely to rise and which are not.
For example, it suggests Brisbane will be the leading city for growth in the next three years, and that makes some sense. It also suggests that Adelaide and Hobart will do better than Melbourne, Perth, Canberra and Darwin, and that generally aligns with my own thoughts, based on our research.
It forecasts growth in some of the key regional centres, such as Newcastle, Wollongong, Ballarat, Bendigo, Cairns, Townsville, the Sunshine Coast and the Gold Coast – and that also makes sense.
But the exact numbers they have applied to those cities for price growth in the next three years? Give me a break.
Terry Ryder is the founder of hotspotting.com.au. You can email him or contact him on twitter.
