Australia is enjoying a home building boom and new dwelling commencements will surpass the 200,000 mark this year for the first time in history.

It is far from a universal boom; many regions in Australia would disagree with the description entirely. At a national level, however, we have a great news story regarding new housing generating considerable domestic demand at a time when the Australian economy is growing below trend. How much further under the mark would we be without the multiplier impacts of new housing?

With new housing supply being very large this cycle, we have inevitably also had some sweeping statements about Australia having an oversupply of housing. Such claims are bogus and if fact damage the prospect for meaningful policy reform because they cast a spurious account of how Australia is performing in terms of housing its growing and ageing population.

Any cumulative housing oversupply (or undersupply for that matter) is dependent on the point in time when the ‘count’ starts, and with households’ consumption of housing services changing in response to market dynamics there cannot be a static oversupply.

What’s the justification for a particular starting point in time where the balance between the housing stock of dwellings and demand from households was equal? Let’s pretend somebody did get that ‘point’ right. Even if the supply/demand were in balance at the national aggregate level, this wouldn’t be the case in each state, and would be even less so in each sub-market within each state. We view attempts to quantify a housing oversupply with skepticism.

The geographical breakdown angle is important. Australia’s current new housing ‘boom’ has been driven by two strong states, one healthy state, two recovering slowly from recessionary starting points, one falling over as soon as temporary stimulus was removed, and two declining sharply. You can’t generalise about housing oversupply in the midst of that variance in conditions!

The breakdown of dwelling type is also important. As noted in the Autumn 2015 edition of the HIA's National Outlook, the detached house construction cycle had peaked well below its potential. There is a cohort of home buyers out there who would like to build a detached house but are being shut out of the market. It’s gotten to the point this cycle where households can’t carry the cost of waiting up to 14 months for titled land, or multiple months for a simple building approval, or borrow the additional amount required to cover government-imposed gold-plating of user pays infrastructure.

While the National Outlook identifies further short term momentum for the construction of semi-detached/townhouse product, this part of the market is also likely to peak without all the potential demand being realised because of the same aforementioned constraints. The detached/semi-detached market still accounts for 70 per cent of all new home building in Australia, and in the here and now of 2015, there will be insufficient supply to meet the demand.

The National Outlook highlights that there were 112,232 detached house commencements in Australia in 2014. That’s an impressive volume which is eight per cent above the long-term average. The forecast is for detached house commencements to ease by 1.4 per cent this year, but the number could have increased further in 2015 if all the potential demand could be realised.

This is an important lesson for policy makers as they (hopefully) look to the future. Housing Australia’s growing and ageing population isn’t about where we’ve been or spurious mutterings about oversupply. It’s about ensuring the supply of new housing going forward is able to respond to the needs of the community in a timely and efficient manner.

Last year, the HIA released the Housing Australia’s Future research. The research incorporates a vast range of possible scenarios for how much housing Australia may need to supply over the next 40 years. Let’s drill down to one possible outcome and look at a population growth rate over coming decades of 1.3 per cent per annum, the rate assumed in the Intergenerational Report released last month. Let’s also assume that real income growth travels along at a reasonable pace in coming decades.

Under this scenario, we would need to average over 186,000 new dwelling commencements per annum to adequately house our growing and ageing population. Over the last 10 years we have averaged 159,125 commencements. The 20-year average is 155,400, while the long-term average is 152,700.

Those figures are all a long way short of 186,000. How we bridge such a gap needs to be a primary focus, regardless of where the national new housing cycle peaks this time around.

  • Andrew Heaton
    Andrew Heaton
    Industry Journalist
    1 year, 8 months ago

    As is rightly pointed out above, the situation does seem to be somewhat mixed across different regions and sectors. Rising vacancy rates for rental property in Perth, for example, do seem to indicate a fair amount of oversupply there, especially in the multi-residential sector. In Melbourne, credible economists like BIS Shrapnel's Frank Gelber are starting to suggest that apartments could be moving toward oversupply over the next couple of years. In Tasmania, one tends to wonder about the sustainability of the recent upturn in approvals bought about by home buyer incentive programs given that state's virtually stagnant population.

    But these are isolated examples. In Sydney, construction may be booming but it is following on from a massive period of low building activity. BIS talks of a housing shortfall of 50,000 there – equal to one full year of strong construction activity in that state. Brisbane, too, appears undersupplied.

    So it does vary according to location, and any evidence of any national level oversupply would not appear to be supported by much evidence at this point in time.

  • I'm intrigued. The HIA number is 112,232. Other numbers say significantly more. One showed a MAT at 210,000+. Others show numbers between these. Which one is right?
    Who is responsible for generating the correct number that would be agreed to? I'd suggest, when a correct number, agreed to by all, is presented, then perhaps we can focus on just if, what and where any over or under supply is occuring.