Money Property Why a soft landing for housing is entirely feasible
Updated:

Why a soft landing for housing is entirely feasible

house price soft landing
While some hope for a property 'crash', a soft landing will be much better for the economy as a whole. Photo: AAP
Share
Tweet Share Reddit Pin Email Comment

With the latest Bureau of Statistics data showing property price falls across four cities, debate is intensifying as to whether we’re seeing a ‘crash’ or a ‘soft landing’ for the overheated housing market.

More than ever it’s important to reframe that question as ‘which areas will crash?’, because the data continues to show very different conditions in different cities.

The falls seem to be gathering pace in Sydney (-1.4 per cent in September); resource state capitals Perth (-1.0 per cent) and Darwin (-2.6 per cent) are predictably weak; and even home of the federal government gravy train, Canberra (-0.2), looks shaky.

So will those falls spread to other capitals, and if so will it be a cooling or a big freeze?

The charts below give some clues – and note that if you hover over a coloured dot in the legend, it’s easier to see how individual cities have performed.

The first thing to note is that the history of price growth across the country has been far from smooth in the past 15 years.

Sydney suffered some pretty shocking price stagnation, especially in its outer western and south-western suburbs, between 2004 and 2009 – a reminder that our most populous city can do one thing, while the rest of the nation does another.

That’s a crucial point, because when we talk about the ‘bursting of the Australian housing bubble’, the language used by media outlets can help stoke panic selling, or calm things down.

It may be far more useful in months ahead to talk about a Sydney slump and a nationwide soft landing.

On Tuesday, economists at HSBC said they thought the market as a whole would have a soft landing, returning to annual price gains of between 3 and 6 per cent.

Capital Economics, which has a strong record of forecasting price moves, foresees a 10 per cent overall price fall in the second half of 2019.

But again, it’s important to remember that Sydney’s disproportionate size, and staggeringly high house prices, will account for a very large part of that fall if it happens.

And the really big question is not whether a few over-leveraged investors lose their shirts, but whether the real economy will be ravaged and family breadwinners thrown out of work.

I have been arguing for some months that we should all hope for a soft landing – yes, it’s galling that the credit bubble got this large, but the entire nation has an interest in seeing it deflate in an orderly way.

Signs of hope

The chart below shows why this is not a forlorn hope. Essentially it shows which cities have surged ahead of income growth and which have not.

For the sake of argument, I have assumed that a couple, both earning the average total pay for their state, want to buy a home – and note that ‘total’ pay is what they are actually taking home, rather than the often misleading ‘full-time weekly earnings’ figure.

It is not a perfect calculation, as incomes do vary between regions and the capital cities in each state, but as a guide it shows the cities in which wages have actually done a pretty good job of keeping up with property price gains.

Hobart, Perth, Brisbane and Adelaide all fit that category, with median house prices hovering in the range of 3.25 to 4.25 times the hypothetical couple’s incomes.

Before the madness of the post-financial crisis credit boom, 3.5 times gross income was considered a normal, conservative upper limit for lending, so they’re still looking pretty reasonable.

Sydney and Melbourne, not so much. The average couples there would be trying to borrow 7.5 and 5.7 times their incomes, respectively.

So commentators around the world who are predicting an Australian crash may be disappointed.

When you get beyond the headlines, the ‘bursting bubble’ may end up being a lot more localised than the aggregate data suggests.

Comments
View Comments