There a little bit of residential construction going on in Australia right now. Well, more than a little. A lot.
In reality it’s an unprecedented building boom, with 220,000 new residential dwellings in the pipeline worth a record $35 billion.
So yeah, A LOT.
Most of the dwellings under construction are high-density, high-rise apartments, predominately in Australia’s southeastern capitals, Sydney, Melbourne and Brisbane.
That means that there’s going to be a lot of new housing stock hitting these markets over the next few years, raising concerns among some parties that there may actually be too many dwellings being built, often in the same locations.
There are many more.
For the moment, the most acute concerns surround the outlook for Brisbane and Melbourne’s inner city apartment market, fanning fears over a potential supply glut forming.
As this chart from ANZ shows, an incredible increase in apartment supply will hit these cities, and to a lesser degree Sydney, in the years to come.
Taking building approvals figures from the past five years, ANZ has calculated the likely percentage increase in housing and apartment stock based off the levels reported in the 2011 census.
Perhaps appropriately, the bars for the expected increase in apartment stock are as high as the actual towers being built: a 46% increase in Brisbane, 37% in Melbourne and 26% in Sydney.
Big increases in anyone’s language.
It’s little wonder than some fear that such an abundance of supply could lead to sharp price declines, heightening financial stability risks in these cities that could potentially spread across the country.
They are, after all, Australia’s three most populous cities.
Unlike the saying “what happens in Vegas stays in Vegas”, what happens in these markets won’t just stay in these markets. They’ll have broader impacts on the Australian economy given their sheer size.
But by how much?
That’s what David Plank, Giulia Lavinia Specchia and Daniel Gradwell, economists at ANZ, have been investigating, creating a model to determine what impact an apartment price drop of 10% in any one of these cities could have on the others housing and labour markets.
Here’s what its modelling came up with:
Spill-over effects to other cities are the largest for Sydney, with a price correction in unit prices in Sydney leading to a shock in unit prices in both Melbourne and Brisbane. Specifically, following a negative 10% shock in Sydney’s unit prices, apartment prices in both Melbourne and Brisbane fall by roughly 6% in the first quarter after the initial shock.
We also find evidence of spill-over effects to other cities for Melbourne – although their extent is more limited. According to our model, a 10% shock to Melbourne’s unit prices leads to a decline in Sydney’s apartment prices by roughly 6% in the first quarter after the shock. The impact of Melbourne’s shock on Sydney’s prices is short-lived, however. Further, Melbourne seems to have no significant spill-over effects on Brisbane.
We find no evidence of spill-over effects to other cities from a unit price shock in Brisbane.
A sharp drop in unit prices has a direct negative impact on local house prices in all three cities, with the negative effect lasting around four to five quarters.The impact on housing prices is roughly of the same magnitude as the decline in unit prices in Sydney and Melbourne – with house prices falling by about 9% in both cities over the three months after the shock. Brisbane’s housing market seems to be less affected, with house prices falling by roughly 4% in the three months after the shock.
While the unemployment rate tends to edge higher following a shock to unit prices, confidence intervals are wide. This implies that in all three cities the impact on local labour market is not statistically different from zero.
Given its sheer size, a 10% decline in Sydney apartment prices tends to lead to price declines in Melbourne and Brisbane of around 6%, according to the modelling. And a 10% drop in Melbourne prices tends to weaken Sydney prices, but not those in Brisbane. And Brisbane does not impact either the Sydney or Melbourne markets, an outcome that fits with the recent trend where prices have fallen in Brisbane but not Sydney or Melbourne.
And it is not expected to impact unemployment levels.
Perhaps the most interesting finding is that apartment price declines in all three cities are predicted to lead to falls in house prices in those cities too — a risk that few have cited in the current debate over the high-rise construction boom.
Plank, Specchia and Gradwell agree.
“This research suggests that unit prices cannot be thought of as in a vacuum,” they say. “It doesn’t matter where the price falls originate — both sectors of the market head in the same direction.”
“The net upshot of this is that even owners, builders or developers of detached houses should all be watching developments in the apartment sector very closely, they add.
Like the RBA, they also have stronger concerns over the possibility of apartment oversupply and corresponding price declines in the Brisbane market.
“While Melbourne and Sydney are also experiencing significant additions to supply, the smaller scale of the construction relative to existing supply, higher geographical dispersion, and stronger population growth are all likely to lessen — although not eliminate — the risk of price declines in these markets.
“As such, while the risk of price declines in the Brisbane apartment market remains elevated, it is reassuring that this is unlikely to impact other cities.”
In absolute terms, they think that the results of the modelling are “somewhat encouraging from a systemic risk perspective”.
Plank, Specchia and Gradwell also point out that the assumptions are for modelling purposes only, and are not official forecasts.