Money Property Good luck forecasting house prices, but one worm has turned
Updated:

Good luck forecasting house prices, but one worm has turned

House prices
While forecasting house prices is a mug's game, at least one indicator has some interesting things to say about the market's future.
Share
Tweet Share Reddit Pin Email Comment

As every non-charlatan knows, forecasting is a mug’s game – but that’s not stopping everyone with a crystal ball and keyboard piling onto the “housing prices will fall X per cent” bandwagon. 

The reality is nobody knows how much further Sydney and Melbourne housing prices will fall. There are, at best, some educated guesses, but they remain guesses.

It’s not possible to know when prices are going to be low enough to tempt enough potential buyers off the sidelines to stop the falls, or when FOMO (fear of missing out) turns into FONGO (fear of not getting out) and vice versa. 

That said, one interesting but rough rule-of-thumb guide has turned from negative to stable. I’ll get to it in a moment but must first add a few more caveats about predicting the future.

Most market “forecasts” tend to be a simple extrapolation of whatever has just happened. Someone forecasting a 5 per cent fall will tend to move to 10 per cent when 5 is reached and then onto 12 to 15 and so on until a market stops falling, the apparent faith in their forecasting ability not dulled by being shown to be consistently faulty.

It’s easy to observe this in foreign exchange commentary. When the Australian dollar is moving in either direction, the consensus forecast is likely to be another couple of cents in whatever direction it’s already moved. 

Still demand for forecasts

At the absolute most optimistic and over-simplified best, there’s only one chance in three of any given forecast being correct. A forecast can be too high, too low or on the money. Odds are that it won’t be on the money.

This reality doesn’t stop people wanting forecasts, or that demand being met by various types of forecasters.

We’re hardwired to want to know the future, to have an idea of what’s going to happen next, to not be in the dark about tomorrow. That explains astrology having adherents down through millennia, of people supporting tea leaf readings, chook entrails and Tarot cards. Some suggest it is why our species is so keen to have a religion.

Another factor in the forecasting marketplace is there’s not much kudos and less publicity to be gained by sitting in the consensus middle. There’s some comfort in numbers, in having company when you’re wrong, but there’s no praise or notice for saying what “everybody else” is saying.

This leads to the outlier forecasts, the predictions above or below the consensus. These are the ones most likely to get the forecaster’s name in the newspaper and, thus, there is a temptation to go a bit further than the herd.

To again use the Aussie dollar as an example, an outlier forecast tends to be right more often than the consensus – but it seems to be a totally random matter whether the successful outlier is on the high or low side. 

So what’s the point of paying so much attention and devoting so many headlines to forecasts? The justification is they should make us think, to ponder the possibilities and thus prepare for them. 

Good forecasts come with some sort of reasoning to encourage us to think – a forecast that Bitcoin was a bubble sure to burst needed to come with a rational discourse that hopefully would warn mug punters not to be taken in. Similarly, those forecasts that Bitcoin would rise forever came with justifications that, upon examination, should have warned off anyone not a mug.

My favourite forecasters work in the Reserve Bank, not because they’re proven correct most often but because they are prepared to acknowledge how often they are incorrect – they put their hand up and wear the pain, unlike the vast majority of non-RBA forecasters who like to highlight their occasional successes and forget all about their mistakes.

The RBA’s quarterly forecasts for economic growth and inflation, for example, come with a neat graph showing the bank’s range of certainty – or uncertainty – about their predictions. That’s a rare level of honesty. 

An RBA graph of forecasted GDP growth.
Growth forecasts conducted by the RBA highlight the difficulty in trying to predict the future.

For planning and policy purposes, it’s necessary to have a core forecast, but wise to remain flexible about the way it evolves.

Right now, hectares of newsprint and countless electrons are being devoted to forecasters outbidding each other on how far Sydney and Melbourne housing prices might fall.

Depending on who you want to read, a dip has turned into a correction and, with Sydney down nearly 10 per cent from its silly peak, the “crash” word is being rather recklessly used.

Not long ago, the consensus forecast was for a 10 per cent Sydney correction from peak to trough. Now the fall has been measured at 9.5 per cent, the consensus has moved out to between 15 and 20 per cent. 

Ignoring the perma-bears who have been forecasting a crash forever without success but still get coverage for some easy clickbaiting, it strikes me as a little silly to be shouting “crash”. 

With employment rising and unemployment falling, with interest rates staying low and economic growth overall pretty good, with the vast  majority of investors managing the transition from interest-only loans to principal and interest, with the credit crunch already having happened and APRA, the RBA and government wanting credit to be available, with continuing strong population growth and a cultural desire to own a home when being a permanent renter is a great deal less than convenient, when building approvals and commencements are turning down, it strikes me as unlikely the current healthy correction of an overheated market will turn into a “crash”.

But why let such factors get in the way of an eyeball-catching headline? 

One indicator worth watching

So, with all those caveats out of the way, here’s a little indicator worm that’s turned: The Sydney/Brisbane house price ratio has returned to normal.

As previously reported, Sydney house prices tend to be 1.7 times more expensive than Brisbane house prices. The ratio gets out of kilter from time to time, but then an arbitrage of one kind or another tends to bring it back. 

At the June quarter peak last year, it was 2.16 when the median Sydney house sold for $1.178 million and the Brisbane median was $546,043. As suggested here in April, the 1.7 ratio was likely to be restored by one means or another, most likely a 10 per cent fall in Sydney and a little rise in Brisbane.

The November CoreLogic figures show that has nearly happened. CoreLogic says the median Sydney house was $935,713 last month. The median Brisbane house was $542,273 – a ratio of 1.726. Getting close. 

The November Sydney total dwellings median price was $821,438. Brisbane was $493,041 – a ratio of 1.666. 

Markets have a tendency to overshoot a bit in both directions. That consensus is for further falls in Sydney while Brisbane pretty much holds its own. The Sydney/Brisbane house price ratio of 1.7 beckons. 

But that looks a bit like a forecast – and forecasting is a mug’s game.

Comments
View Comments