ABC News – 16 April, 2015
By Michael Janda
The Australian housing shortage is a gospel truth for everyone from Treasury and Reserve Bank policymakers down to your average Joe chatting with mates at a weekend barbeque.
It is one of the rare areas where analysts with vested interests (such as those employed by property developers, real estate agents and banks) can agree with most independent analysts (such as academics) free from such interests.
However, despite this overwhelming consensus I have a nagging doubt that it is true, and I am not alone.
On Monday night’s The Business, one-time business analyst and author of Australia Boom to Bust Lindsay David said Australia had built plenty of homes – about one for every 1.9 people that had been added to the population.
The same segment featured a graph from this story that I published late last year, which showed that Australia had actually built more homes than England over the past decade, even though England has twice the population and also had a bigger numerical increase in population.
Admittedly, all that shows is that England has a bigger recent housing shortage than Australia. It doesn’t answer the question of whether Australia has a housing shortage in the first place.
But a tweet from property buyers’ agent Pete Wargent inspired me to dig deeper.
Our debate on Twitter was about whether Sydney has a housing shortage, with the vast bulk of analysts attributing a recent price surge in that city to a lack of dwelling supply amidst strong population growth.
And there’s no doubt that Pete Wargent’s graph, based on Bureau of Statistics data, backed that up … over the past decade.
But this again touches on a common problem with arguments against Australia potentially having a property bubble – they use data that only goes back a decade.
The reality is that the big damage in all the metrics that suggest a possible bubble – house price-to-income ratios, household debt-to-income ratios, house prices themselves – was done in the late 1990s to early 2000s, outside this window.
So I went back to the ABS data to see how NSW dwelling completions compared to population growth over a longer time frame – almost 30 years from 1985 to late 2014 (with that range limited only by the availability of ABS data).
The graph indeed highlights strong population growth and weak dwelling construction over the latter half of last decade and the early part of this decade.
But go back to the early 2000s and there were some years where NSW actually had more new dwellings completed than there were extra people to live in them.
Supply was also very strong all through the 1990s, meaning the state (around two-thirds of which is greater Sydney in terms of population) saw at least a decade and a half of very strong housing supply before the more recent population surge.
Comparing dwelling completions to population is also quite misleading, as the Bureau of Statistics census shows an average of at least 2.5 people per dwelling.
I have adjusted the numbers in the following graph so that it shows how many people the new dwellings would actually be expected to house versus population growth and, to be conservative, have used RBA and census vacancy estimates to adjust the dwelling completions downwards to account for demolitions and intentionally vacant properties, such as holiday houses.
As you can see from the graph, there are plenty of years where new housing capacity exceeded population growth in NSW.
If you do the numbers, adjusted for 15 per cent of completions replacing demolitions and 10 per cent intentionally vacant, NSW built a new dwelling for every 2.4 extra people between 1985 and September 2014.
Given the ABS census finds an average 2.5 people per dwelling, that would suggest NSW housing construction is more or less in line with fundamental demand generated by population growth, if not a little ahead.
After all, most of the houses and units built in 1985 are still standing and being lived in, so it doesn’t make sense to limit an analysis of housing supply and demand to the last decade.
And the story is very similar for Australia as a whole.
Yes, dwelling completions versus population growth look very scary over recent years but go back further and again it seems that Australia has built plenty of homes to house both net births and net immigrants.
The graph that looks at how many people these new dwellings should be able to house illustrates this nicely, with a net result that Australia built a new dwelling for every 2.35 extra people, even adjusted for demolitions and intentionally empty properties.
None of these graphs prove that there is no housing shortage anywhere in Australia – the ABS excel tables only come with state, not regional figures.
And, given that the data only goes back to 1985, it is quite possible that Australia may have had some underlying housing shortage going into this period.
However, these data do put into question claims that Australia has a massive housing shortage and that it is this shortfall above all else that explains the substantial 68 per cent rise in capital city home prices over the past decade and meteoric 125 per cent rise over the decade before that.
Incidentally, that 125 per cent property price surge occurred precisely during the period when Australia was building lots of homes relative to its population growth.
Perhaps the spotlight needs to go back on non-residential sources of real estate demand – speculative investment based on favourable tax policies and the expectation of capital gains, and foreign investment in a perceived safe haven asset.
It is also worth considering the role that cheap and easily available credit has played in pushing prices higher for any given level of housing demand.
Finally, it is also worth remembering that housing demand is not totally inelastic, to use a piece of economic jargon. That means that people can and do adjust their housing consumption patterns based on the price of housing.
Household sizes shrank for all of the 20th century in Australia, but that trend has begun to very gradually unwind early in the 21st as unaffordable housing sees younger people stay at home longer and share houses come back into vogue, even for 20 or 30-something professionals.
So, even if we didn’t have quite enough homes at one price point, as prices keep climbing higher we may eventually find we have too many for an Australian population seeing low, or no, income growth and becoming more frugal with the space they need.