Despite talk of a housing bubble in Australia, house prices have been flat in real terms over the past six years since the pre-financial crisis peak in December 2007. The weighted average of established house prices in Australia’s capital cities has just kept pace with inflation, according to the Australian Bureau of Statistics house price index.
Much discussion has focussed on short-term factors that have supported housing prices since the crisis: Chinese buyers, low interest rates, the switch from shares into property by investors, and the renewed willingness to take on debt since households and investors have repaired their balance sheets post-crisis.
But we need to bear in mind potential long-term drivers of house prices, one of which is population ageing. We’ve known about population ageing in Australia and other developed countries for a long time. It’s like the Queen Mary - we can see it coming from a long way off, but it’s difficult to turn around.
Predicting house prices
There have been some over-hyped predictions about the potential effect of population ageing on house prices in the past. The so-called “asset meltdown” hypothesis was fuelled by some influential US economists. They said that the retiring baby boom generation would sell down their assets, like housing and shares, to pay for their retirement, causing asset prices to plunge. The infamous Mankiw and Weil paper, for example, predicted that US house prices could fall by up to 40% as the baby boomers started to retire between 1990 and 2007. Suffice to say, this didn’t happen!
My research for Australia suggests such a prediction is way out of the ball park (as are many other predictions of calamitous change to be wreaked by population ageing – but that’s a topic for another day). That said, ageing will affect house prices. But it will be a slow-burn effect, occurring gradually over several decades, and the magnitude will be modest.
According to our analysis based on both econometric estimation using historic data as well as predictions from a household behavioural model, ageing will cause house prices to be lower than they otherwise would be, by somewhere between 3% and 25% over the next 35 years.
Two factors drive the effect of ageing on house prices in Australia. First is the declining population share of 35 to 59 year olds who have the highest incidence of property purchases, including both investment and owner-occupied housing. Investment housing has increased as a share of total housing purchases in the last two decades according to data from the Reserve Bank of Australia, and it is the 35 to 59 year olds who tend to be housing investors.
Second, older people may trade down in order to release housing equity to support their retirement. The latter effect has, however, become weaker as households have the ability to withdraw equity in their house without needing to sell it, through facilities such as equity redraw loans and reverse mortgages.
Housing bubbles and baby booms
The link between house prices and demographic change could also run the other way: from house prices to the population age distribution.
High house prices tend to reduce fertility rates according to the international evidence, at least among non-home owners, although it may increase fertility among existing home owners. High house prices reduce the affordability of children for non-home owners, but increases the affordability for existing home-owners by increasing their wealth. In fact given financial and emotional investment in both housing and childbirth it stands to reason that these decisions are related in quite complex ways, and depend on government policies in the areas of family assistance and housing affordability. I plan to investigate these relationships for Australia with co-authors from Macquarie University and the ANU.
One of the lessons from economics is that it’s very risky to extrapolate from the past by drawing straight lines from historic data into the long-term future, for two reasons: the economy has powerful self-correcting forces that we tend to underestimate; and unexpected shocks – Donald Rumsfeld’s “unknown unknowns” – can throw the data off course for a considerable period of time.
This was the fate of Mankiw and Weil’s prediction of a possible 40% fall in US house prices over 20 years – in fact house prices rose dramatically over that period. Compared with the slow-burning effect of population ageing, events like financial booms and crises have a much greater potential to cause swings in house prices over the short to medium term – say five to ten years. And this is the period that matters for most people.
The long term is interesting and we should keep an eye on it. But we’d do well to remember the maxim of the great economist John Maynard Keynes who said: “In the long run we are all dead”.
This is the fourth piece in our Housing 2020 series, exploring the major policy issues facing housing over the next five years. Click on the links below to read the other pieces.