Monday 12 March 2018 by Guest Contributor Opinion

A housing bust in Australia? Not so fast

Guest economist, Saul Eslake says the forecasting of an imminent collapse in Aussie housing prices from foreign and local observers ... is a bust. Compared to other countries, he suggests a 'housing bust' in Australia is an unlikely scenario


By Saul Eslake

Over the past two decades, one feature of Australian economic commentary has been the regular flow of dire warnings from foreign observers forecasting an imminent collapse in Australian housing prices.

The widely read and respected magazine, The Economist, has been a regular source.  From as early as 2003 when it forecast a 20% fall in Australian property prices over the following four years, up to as recently as May last year.

Citigroup’s Willem Buiter, SocGen’s Albert Edwards, GMO’s Jeremy Grantham, American based demographer Harry Dent and perennial uber-bear Marc Faber, among others, have at different times over the past decade drawn attention to the ‘over valued’ Australian housing market. Each forecast imminent or predicting eventual sharp declines, often with serious consequences for the stability of the Australian banking system and for the broader Australian economy.

They’ve often been joined by locals, including Steve Keen, Lindsay David, Variant Perception’s Jonathan Tepper and Bronte Capital’s John Hempton.

None of these dire predictions have, as yet, been accurate.

Although, that doesn’t mean that they have been entirely without foundation. There’s no denying that residential property prices in Australia’s largest cities, particularly Sydney and Melbourne, have risen dramatically since the 1990s, and that they are now very high by both historical and international standards (Figure 1).

Residential property prices – Australia and selected other countries

Sources: CoreLogic; Bank for International Settlements
Figure 1

Nor is there any denying that the inflated Australian house prices have been accompanied by a similarly striking increase in household debt – an increase that has continued, in contrast to other countries, almost unabated over the past 25 years (Figure 2). 

Household debt – Australia and selected other countries

Source: Bank for International Settlements
Figure 2

The substantial increase in Australian household debt has been both a cause and a consequence of the rise in property prices over the past quarter of a century. 

Lower interest rates, more readily available credit, and distortions in Australia’s tax system have enabled and encouraged households to take on more debt in order to acquire property. This combined with the strong ‘underlying’ demand for housing pumped by a rapidly growing population has driven property prices inexorably higher. As a result, those wanting to acquire property – whether as owner-occupiers or as investors – have been obliged to take on more debt in order to fulfil their aspirations.

The ongoing inflation of Australian residential property prices is by no means the unalloyed ‘good thing’ that politicians, real estate agents and many others typically claim. Yes, it has been a source of increased wealth for many Australians. But it has also been a source of greater inequality, especially between those who own property and the increasing proportion of Australians who don’t. The most recent Census home ownership rate in August 2016, was lower than any Census since 1954. One of the core assumptions on which Australia’s retirement income system rests – that the vast majority of retired people will own their homes outright, and as a result have very low housing costs – is becoming increasingly invalid. 

Against that background, I think it is a good thing that in most of Australia’s capital cities, housing prices appear to have peaked, and in some cases have begun to decline.

However, none of this makes it likely, probable, or inevitable that Australia will sooner or later experience a housing ‘crash’ of the sort repeatedly foretold by commentators.

Housing prices aren’t subject to the law of gravity – there’s nothing in either theory or practical experience which says that, when it comes to property markets, ‘what goes up must come down’.

Those countries that have experienced a housing price ‘crash’– such as the US, Ireland and Spain (see Figure 1 above) – tells us that there are two ‘pre-requisites’ which have to be satisfied in order for something like this to take place:

1.   Forced sellers

There has to be a sufficiently large number of ‘forced sellers’. That is, property owners who have to sell their properties at whatever prices buyers are willing to pay, because they can no longer keep up the financial obligations. That usually happens when, during a boom, a large cohort of marginal borrowers manage to acquire properties, but become unable to keep up mortgage repayments because interest rates or unemployment (or both) rise sharply. 

That was certainly the case in the United States, where one of the defining characteristics of the pre-crash housing ‘bubble’ was the proliferation in ‘sub-prime’ lending. That allowed many people who had hitherto been unable to access mortgage finance to become home owners, thus leading to a rise in the home ownership rate – something which was widely acclaimed in the US at the time (Figure 3). 

Home ownership rates – United States and Australia

Sources: ABS; US Department of Commerce
Figure 3

But it also meant that, when the concessional introductory terms began to expire, and when the general level of US interest rates started to rise, a large number of these new home owners, and others who had been pure speculators, began defaulting on their loans. It was made easier in the US by the widespread use of ‘non-recourse mortgages’ where borrowers could simply walk away from the property and were not responsible for the debt.  This lead to a significant increase in the number of ‘forced sales’, putting substantial downward pressure on prices (Figure 4). 

Mortgage delinquency rates – United States and Australia

Sources: APRA and RBA; Mortgage Bankers' Association of America
Note: Series for Australia is banks' 'past due' and (after September 2003) 'impaired' mortgages; for US is all mortgages 'seriously delinquent'. 
Figure 4

By contrast, in Australia as noted earlier, the home ownership rate has been declining since the early 1990s (Figure 3); while mortgage delinquencies have always been much lower than in the US (Figure 4). This held true even during the years immediately before the onset of the financial crisis when the standard variable mortgage rate peaked at 9.6%, far higher than mortgage rates ever rose in the US during this period. 

Australia’s legal system makes default a much less ‘easy’ option for borrowers than it is in the US. More importantly, although they’ve been by no means faultless, Australian mortgage lenders have been much more prudent in their lending practices than their American counterparts. Australian regulators have also been much more effective in regulating lending practices than their American peers. 

In recent speech by Assistant Governor Michelle Bullock, a large proportion of Australian borrowers used the recent period of record low interest rates to build up ‘buffers’ against future interest rate increases. 

2.   Excess supply 

There has to be significant ‘excess supply’, as a result of new dwelling completions exceeding growth in ‘underlying’ demand for housing by a wide margin, over an extended period.

That was certainly the case in the US during the years leading up to its housing ‘bust’ – and it was an even more prominent characteristic of the booms in Spain and Ireland (Figure 5).

Housing completions – Australia and other selected countries

Sources: ABS; US Commerce Department; UK Office for National Statistics; Instituto Nacional de Estadística (Spanish Statistical Office); Ireland Central Statistics Office
Note: 2017 data for Australia, UK and Ireland are first three quarters at annualised rates
Figure 5

But it didn’t happen in the UK, which hasn’t so far had a ‘housing bust’. And it hasn’t happened in Australia, notwithstanding the record level of dwelling completions in recent years, on anything like the scale that it did in Ireland and Spain, or even the US. Moreover, the increase in dwelling supply that has occurred in Australia in recent years is overwhelmingly concentrated in apartments, whereas completions of detached dwellings (which still accounts for the bulk of the housing stock in Australia) have seen little change from historical norms.

Of course, if mortgage interest rates were to rise sharply in Australia at some point in the future, then it’s conceivable that a sufficiently large number of property owners could become ‘forced sellers’. However, it’s in part precisely for that reason that the Reserve Bank is unlikely to engineer a sharp rise in interest rates – as would be required for mortgage rates to rise sharply enough to prompt a wave of ‘forced’ selling.

If Australia’s immigration intake was lowered sharply – significantly reducing the ‘underlying’ demand for housing – then that could lead to the emergence of ‘excess supply’. If it were also accompanied by a significant volume of sales, then it may exert meaningful downward pressure on property prices. However, lower immigration doesn’t appear to have attracted widespread support.

In short, while it would be foolish to say that there is no chance of a ‘housing bust’ in Australia – and I’m not saying that – it does seem a low probability scenario. It’s certainly not one on which bond investors should be basing their portfolios, anticipating a renewed fall in interest rates in a desperate attempt to stave off, or mitigate, the consequences of a ‘housing bust’.