Business Spectator Home Page

You often hear commentators make flawed comparisons* between the Japanese and Australian housing markets, which are then used to underpin claims that local residential property prices are about to enter into a death-spiral.

Setting aside the obvious differences between Japan’s banking system in the early 1990s — which effectively collapsed contributing to the so-called ‘lost decade’ wherein GDP growth was no more than 1.5 per cent per annum—and Australia’s relatively healthy major banks, there is another more fundamental distinction between Japan and Australia that renders any such comparisons redundant: population growth.

Population growth is widely accepted to be one of the most critical determinants of long-term housing demand. The figure below shows recent Commonwealth Treasury forecasts for Australian housing demand out to 2009-10. The blue line represents Treasury’s housing supply projection while the sum of the solid segments is their forecast for total housing demand.

You can see that aside from the fact that the Treasury expects to see a large and growing gap between housing demand and supply in Australia, by far the biggest components of demand are organic population growth (viz., fertility or ‘fecundity’) and net migration. (There is another major and rarely commented on curve-ball here that demographers have yet to get their heads around: longevity risk… What happens if we start living longer? Current population projections are based on relatively crude life expectancy estimates that do not accommodate significant non-linear changes in longevity that could be caused by medical breakthroughs. And the probability of such events is presumably far higher since we started decoding the human genome.)

Source: Commonwealth Treasury

If a nation’s population growth rate slows, or even — heaven forbid — slips into negative territory (ie, its population shrinks), the growth in housing demand will, ceteris paribus, also fall, in turn placing pressure on prices.

With this in mind, one would be hard pressed to find a better case study in population contrasts between Australia today and Japan at the start of the last decade.

Whereas Japan’s population growth rate began an inexorable decline in 1975 (when it peaked at a reasonable 1.35 per cent annualised rate) slumping to just 0.42 per cent by 1990, and continued falling until it finally turned negative in 2005, Australia has experienced a population boom the likes of which has not been evidenced for 20 years. And Australia’s high rates of growth are expected by the ABS to persist for decades to come.

On the 2nd of December 2008, the ABS announced that Australia’s population growth rate had risen to 1.7 per cent, which was its strongest rate since 1989. Australia’s total fertility rate has also increased to over 1.9 births per woman, a rate of pace that we have not experienced since way back in 1981. Importantly, this is a significantly higher fecundity rate than that which is assumed in the ABS’s most recent population projections.

By way of striking contrast, Japan’s Ministry of Internal Affairs and Communication (MIAC) informs us that “since the 1980s, [Japan's population growth rate] has declined sharply. Japan’s total population started declining [ie, realising negative growth]…at the end of 2004”.

The MIAC forecasts that Japan will continue to be characterised negative population growth rates right through to 2050, which is the end of its projection horizon. More specifically, they estimate that Japan’s total population will have declined to just 95.1m persons in 2050 from its peak of 127.8m persons in 2004. Put differently, Japan is forecast to lose an incredible 32.7m persons — or nearly 27 per cent of its 2004 population — over the next 40-50 years.

The de-facto failure of Japan’s banking system, which brought about extreme credit rationing, combined with very low to negative population growth rates, and anemic economic growth during the so-called ‘lost decade’, were ultimately responsible for residential prices in Tokyo, Osaka and Nagoya falling by 30-40 per cent between the early 1990s and mid 2000s (these losses were, incidentally, far less than those suffered in the commercial property sector). More recently, banking crises, capital market failures, and credit rationing in the US and UK have led to similar consequences albeit without the albatross of negative population growth.

Yet Australia benefits from a relatively secure banking system that has thus far avoided any crises, near-record population growth rates that are projected to continue out to at least 2056, mortgage default rates that are a fraction of equivalent metrics in the US and UK, a housing finance industry that is bereft of any real sub-prime characteristics, little-to-no government debt, a mortgage culture in which 85 per cent of borrowers have variable rate loans that get immediate relief from central bank-induced rate cuts (in contrast to the US, and to a lesser extent the UK, where fixed rate loans dominate), and a consensus amongst most mainstream economists that we have credible long-term GDP growth prospects in part because of the ample ability of both monetary and fiscal policy to further stimulate the economy (ie, once we navigate our way through the current downturn).

This latter point raises a final contrast of note: while the Japanese have for a long time struggled with the spectre of a ‘liquidity trap’ with the official discount rate having been less than 1 per cent since 1995, Australia’s cash rate of 4.25 per cent affords the RBA a great deal of room to move.

These and other intrinsic differences between the two countries suggest that it is foolhardy to use Japan’s lost decade as a guide to Australia’s economic future.

While we’re on the subject of population growth, it is worthwhile taking a closer look at Australia’s demographic outlook (especially given its crucial role in driving long-term housing demand and hence prices).

Every four years the ABS produces new population ‘projections’ for Australia. The ABS computes three main series of projections — Series A, B and C — which are selected from a possible 72 individual combinations of various assumptions. The ABS comments, “Series B largely reflects current trends in fertility, life expectancy at birth, net overseas migration and net interstate migration, whereas Series A and Series C are based on high and low assumptions for each of these variables respectively.”

The ABS’s latest set of Series B projections show a big increase over their 2004 estimates. The ABS now projects that Australia’s population will increase by more than 65 per cent from 20.7 million to 34.2 million persons by June 2051. The last time the ABS released population projections — in June 2004 — it forecast that the population would be 28.2 million persons by June 2051 (see the first table below). That is, the ABS modelling implies that Australia will have six million more residents in 2051 than it previously projected.

The principal explanation for this change is a large increase in the ABS’s assumption around net overseas migration. It has boosted this from 110,000 persons per annum (in 2004) to their 2008 estimate of 180,000 persons per annum (note that immigration has of late been running closer to 200k per annum). The ABS has also slightly increased the assumed fertility rate from 1.7 to 1.8 babies per woman — again, notably less than the current rate.

The dramatic lift in the ABS’s immigration forecast reflects something that we have been arguing for a long time now: with an ageing population, Australia will have no choice but to actively embrace stronger immigration (or higher fertility) in order to compensate for the reduction in its labour force.

This point is particularly germane given our low unemployment rate (4.4 per cent), which suggests that even in the current environment there are serious labour shortages (ie, before the baby boomer population bubble moves into full retirement).

In fact, insofar as the Series B estimates simply extrapolate out from recent empirical trends, we believe that there is potentially a case for one to play closer attention to the Series A outcomes, which utilise more aggressive assumptions for variables such as net overseas migration.

In response to the ABS’s new Series B projections, the widely referenced demographer, Bernard Salt, has concluded, “The bottom line is that there has been a paradigm shift in the way demographers view Australia’s future and the key difference is immigration.” Salt believes that these new projections, “change the ball game with regard to metropolitan planning…The new projections… are so vastly different that it is now necessary to rethink the scale and form of our largest cities…At the very least, it requires all existing planning documents to be recast to accommodate growth on a hitherto unparalleled scale. We need to think boldly about how we manage our burgeoning cities and regions in the future."

* In this ABC interview Steve Keen claims house prices in Tokyo fell by 75 per cent. This appears to be quite inaccurate. According to academic research, between 1993 and 2003 residential prices in Tokyo, Osaka and Nagoya fell between 29 per cent and 43 per cent. Keen may have been confusing residential prices with commercial prices, which fell between 64 per cent and 74 per cent.