Aussie housing stock is not too expensive

As we look to what's ahead for 2012, Property Observer is republishing some of our most noteworthy stories of 2011.

One of the perennial puzzles tossed around Australian dinner tables is whether our bricks and mortar is expensively priced.

Despite the steady ascent of national dwelling prices at an inflation-plus pace over the last few decades, one has never had to look far to find an “expert” declaring that the world was about to end. Predictions of catastrophic falls in house prices have always been good for garnering a few media headlines.

And, regrettably, the media is not great at holding purported experts to account. A classic example appears in the Australian Financial Review today. AMP’s Shane Oliver is reported as claiming that Sydney housing is overvalued because the representative dwelling price is an unacceptably high $634,000. Shane also contends that since his home’s capital gains have outstripped his personal income growth, this is likely true of all Sydney housing.

Unfortunately, the facts do not fit with this fiction. Shane’s estimate is actually off by $119,000 (or 23%). According to the biggest database of home sales in Australia, which covers all transactions, the median dwelling price in Sydney is currently $515,000. Furthermore, Sydney dwelling prices have only risen by a compound annual growth rate of 2.7% over the last eight years. That means Sydney dwelling prices have declined in “inflation-adjusted” terms (after a very strong run-up before 2003). If we look at Sydney detached houses – as opposed to all dwellings – the compound annual growth rate falls to just 2.3%.

In income-adjusted terms, the comparison is even more stark: the ABS reports that average disposable household incomes have advanced by 6.3% per annum since March 2003. This is a massive difference. It means that while disposable household incomes across Australia have risen by 63.4%, Sydney dwelling prices have expanded by only 24.6%. That is, incomes have outpaced Sydney dwelling prices by an extraordinary 38.8 percentage points over the period March 2003 to March 2011.

In sympathy with the boom-bust business cycle, asset prices tend to also rise and fall.

The housing market is a little different to other asset-classes insofar as you can typically obtain more debt against a property – more than 90% in some cases – and for longer terms – up to 30 years – than alternative investments. This means that there are latent risks that can build up if credit growth is unnecessarily lubricated and used to fuel house price inflation, which can stimulate more credit creation until the Ponzi pattern is undone by a shock.

The point is that policymakers need to be attuned to these interdependencies if they are to avoid the sorts of problems we saw emerge in the US.

Across the $3.5 trillion worth of private housing in Australia, the amount of debt secured against it is actually quite low, at circa 29%. For those home owners with a mortgage, the value of their loan as a proportion of their home is, on average, also pretty low, at 50 to 60%. That means house prices would have to fall by about 40% before the typical person suffered “negative equity”.

Clearly, there are also folks sitting in the tails of this distribution who are more vulnerable. But they are a small percentage of the total. The RBA estimates that only 2.5% of all borrowers have a loan-to-property-value ratio equal to or greater than 90% and are paying away more than 50% of their disposable income to service their repayments. 

One of the reasons banks have been prepared to lend so much for so long is the fact that borrowers have historically been vigilant in paying off these loans. Today there are slightly more than 30,000 borrowers who are more than three months behind on their home loan repayments, juxtaposed against a total pool of roughly 4 million to 5 million borrowers. That is, Australia’s “mortgage default rate” is a paltry 0.7%, despite our internationally high lending rates. This is less than one-10^th and one-quarter the equivalent US and UK default rates respectively.

With more than 60% of all household wealth accounted for by housing, discussions around its valuation understandably attract attention.

There are, of course, many ways to “value” an asset. One school of thought is that the best possible measure is the price at which willing buyers and sellers are prepared to transact. That is, today’s prices. This is known as the “efficient markets” view of the world.

Others like to point to the underlying demand- and supply-side fundamentals while abstracting away from prices, incomes and the cost of debt. These drivers are, among other things, the rate at which new households are formed, which is determined by population growth and the number of people willing to share their homes, and the rate of construction of new dwellings (i.e., supply).

The RBA, Treasury and the Government’s own National Housing Supply Council have all concluded that underlying housing demand in Australia looks to be running well ahead of supply.

An alternative benchmark that gets regularly referenced in the media is how house prices compare with incomes over time.

Although the calculation of a “house price-to-income ratio” is tricky, if you can lay your hands on a timely (i.e., regularly updated) proxy for dwelling prices, and contrast this with an equally frequent measure of household incomes, you can make some valuable comparisons of the change in normalised housing costs over time.

Rismark produces Australia’s most comprehensive and timely dwelling price-to-income ratio, which is released on a quarterly basis following the publication of the ABS’s National Accounts. It covers all homes sales across Australia and uses the ABS’s measure of disposable household incomes, which is the only official estimate available on a quarterly horizon.

In encouraging news for home buyers, Australia’s dwelling price-to-disposable income ratio is now back around its June 2003 level, based on Rismark’s latest analysis.

We found that Australia’s average dwelling price-to-average disposable household income ratio has fallen to just 4.2 times in March 2011 from its recent peak of 4.7 times in December 2009. (While one can quibble about the exact levels, the change in incomes and prices is more difficult to dispute.)

Assuming that the RBA hikes interest rates another 0.5 percentage points over the next six to 12 months, we believe that the gap in the growth rates attributable to incomes and prices will widen.

While it is unlikely that that Australian dwelling prices will rise much as long as the RBA wages its war against inflation, investors will continue to benefit from robust rental returns given our historically low vacancy rates. And although we will hear complaints about the multi-speed economy, solid overall wages growth of more than 4% per annum in concert with the income flowing from the capex boom should yield reasonable increases in household earnings.

The RBA will be pleased with these developments, since it is hoping that the retail and housing sectors will make room for the unprecedented volumes of private investment already baked into the capex pipeline.

Rismark’s central case is that the RBA will commence normalising interest rates at some point during 2012-13, which should drive a powerful affordability dividend. This will stimulate higher capital growth rates and the price signals necessary for builders and developers to start committing scarce capital to the production of new housing supply. Indeed, the recovery, when it eventually materialises, could give rise to significant capital gains. Through the cycle, however, we believe that the cost of housing will track disposable household earnings, as it has done in the past.

Contrary to popular myth, Australia’s house prices have actually grown more slowly than household incomes since the end of the last boom in 2003. According to the ABS’s National Accounts data, disposable incomes on a per household basis have realised a compound annual growth rate of 6.3% since March 2003. In contrast, RP Data-Rismark’s hedonic index suggests capital city dwelling values have risen by a more modest 5.7% per annum. On this basis, disposable incomes in Australia have advanced 7.5% further than capital city dwelling prices over the last eight years.

Unlike other estimates, Rismark’s national dwelling price-to-disposable household income ratio includes:

• Dwellings across all Australian regions (not just capital cities);
• All property types (not just detached houses); and
• The ABS’s quarterly National Accounts measure of average disposable household incomes (not just average weekly earnings), which captures income earned from all areas (e.g., labour and investment income) and reflects the fact that there is typically more than one income earner per household.

Our preferred ratio also compares average dwelling prices with average disposable incomes on a per household basis.

It is important to note that the income measure Rismark uses is the quarterly ABS National Accounts disposable household income proxy before interest payments and excluding unincorporated enterprises, which is then divided by the HIA’s estimate of the total number of households. It represents the total disposable wage and investment income generated by all members of the household and should not be confused with simpler estimates of average wages or median incomes.

On an annual basis, one can examine the impact of excluding the “net imputed owner-occupied rents” that the ABS includes in its National Accounts numbers. The chart below shows that this results in a small increase in the dwelling price-to-income ratio.

Christopher Joye is a leading financial economist and works with Rismark International. Rismark and RP Data provide house price analytics products, and solutions that enable investors to go long and/or short the housing market. The above article is not investment advice. You can follow Christopher on twitter at @cjoye or read his blog.