Australian housing prices are driven, almost exclusively it seems, by the level of interest rates. Latest data, from the RBA point to that, and models developed by an investment firm confirm the finding that, under a low interest rate environment, the average price of houses will lift by around 25% to the end of 2023.
This piece of information, one of many, does not tell a complete story of the housing market, but it is a good beginning and has received solid attention recently.
Using underpinning information from the RBA, Coolabah Capital Investments (CCI) has predicted house prices will rise by 25% by the end of 2023, with rises of an average 8% in 2021, an average 9% in 2022 and around 8% in 2023. The range within the predictions is an aggregate of between 14% and 33% by the end of 2023.
The model is more complex than just a linear relationship between interest rates and prices. As Matthew Cranston identified, in the Australian Financial Review, the model is not adapted for the pandemic and that could alter the predictions.
But behind the private research by CCI, sits a ‘ground-breaking’ housing market model developed by the RBA. The model, writes CCI, “…allows for feedback between housing construction, rents, vacancies and prices…”.
One point of comparison, in the CCI paper, shows that almost all advanced economies experienced house price increases during the pandemic, and Australia is by no means experiencing the heights of some other economies– as the chart shows.
Source: Coolabah Capital Investments
From an Australian perspective, as CCI points out, there are a few drivers for the large rises in prices that cannot be ignored:
· A job retention and welfare scheme
· Record low mortgage rates
· Mortgage payment deferrals, moratoria on rent increases and evictions
· Government subsidies to encourage construction
So, with that context in mind, CCI deployed the RBA model – known as the Saunders-Tulip model (ST) – of the Australian housing market in order to undertake its own price analysis.
The CCI says of the ST model:
The ST model is used by the RBA to analyse the relationships between interest rates, residential investment, rents and house prices. CCI has used it to calculate internally-consistent forecasts that allow for feedback between quantities and prices. The model contrasts with the usual academic approach of estimating house prices using a single equation and provides richer detail on the housing market than the RBA’s MARTIN macroeconomic model.
At its very simplest, the ST model uses multiple factors to predict what will happen in the housing market. When applied to price analysis, CCI found that the ST pointed to:
· The strong house price growth described above
· Stronger residential property investment
· Weaker rents
The last point is important because it seems counter-intuitive. That is, if the income yield (the rent) from an asset decreases, you would expect the value of the asset (reflected in the house price), to decline. CCI points out that the model seems ‘pessimistic’, in this respect. Perhaps so, and maybe all that this means is that the bottom of the price growth range (around 14% on average by the end of 2023) is more likely than the upper limit (33%).
Coming at the ST model from a different direction, Greg Jericho wrote in The Guardian that “… [interest] rate cuts have far outweighed the impact of higher population growth.” When it comes to house prices, but that the situation is reversed when it comes to rental prices.
He also points out that the ST model proves what we have long understood intuitively: that housing activity and building works are highly sensitive to interest rate movements and other stimulus. Jericho writes:
“Indeed, they [Saunders & Tulip] found that the impact of a cut in rates is quite immediate – a one percentage point cut leads to a surge in building approvals for houses within nine months, and for residential apartments within a year and half:”
If nothing else, as the chart shows, the ST model delivers proof that housing is the right place to land stimulus for what is around 25% of the Australian economy.
Commenting on the often-made assertion that housing affordability is impacted by population growth, Jericho also uses the ST model to assert that:
“…the research suggests the impact of 12 years of higher population growth on housing affordability has been greatly outweighed by the impact of the rate cuts.”
To be explicit, that is house prices, not the volume of demand that is created by a higher population. That is, a growing population increases demand but does not have as big an impact on house prices, as do interest rates.
These are valuable insights!
In the endless search for insights into the future, everyone wants a private fortune teller with a crystal ball that provides an instant answer. But most of us have to rely on data, models and forecasts to get the answers we need. It is not as interesting as a crystal ball, but it is more reliable, for the most part. The ST model seems about as close to a crystal ball as we can hope to get.
However they may be assembled, the models like the RBA’s ST model and the outputs that CCI and Greg Jericho have gleaned from it, help to explain the housing market and can assist us to predict what will happen next. That is sufficiently valuable for industry wide efforts to unlock the future.
