Last week's federal budget included a special chapter on Australia's housing crisis.
It was 42 pages long, and contained over20 graphs, with Treasury officials taking time to explain the problems in our housing system.
One graph showed Australia's rental vacancy rate hit a record low recently. But what is the vacancy rate actually signifying?
If you think it's telling you about the total stock of rental properties in Australia's housing markets, that's not quite right.
'Vacancy' and 'unoccupied' are not the same thing
To visualise what we're talking about, this is the graph Treasury officials used in the budget to show the decline in rental vacancy rate.
They say thenumber of dwellings available for rent has been falling in Australia since early 2020.
They saythe rental vacancy rate has fallen below 1.5 per cent, and 3 per cent is considered the rate that reflects a "balanced" rental market (where landlords and tenants have similar bargaining power).
In his recent book,The Great Housing Hijack: The hoaxes and myths keeping prices high for renters and buyers in Australia, the economist Cameron Murray spends more time explaining how to think about that graphic, and it's really interesting.
He says two vacancy concepts often get conflated in our discussions about housing: vacant and unoccupied.
He says what's called the "rental vacancy rate" is the number of rental advertisem*nts that have been posted online for longer than three weeks in a month, divided by the estimated stock of rental properties.
So, if you read a story in the media that says the national vacancy rate has just fallen to its lowest level (say 0.8 per cent), then you can interpret that as telling you: "Of the nation's estimated 3.1 million rental homes, 24,000 of them were advertised online for more than three weeks based on website statistics over the past month."
Dr Murray says "unoccupied" properties are a different concept.
Why? Because while vacant homes might be occupied while being advertised for rent, unoccupied homes are empty.
He says the main data on unoccupied homes comes from the national census taken every five years, which measures how many homes were empty on census night. The most recent census (in 2021) found one million homes were empty on census night.
But he says that's not a particularly insightful figure, because the bulk of those properties were unoccupied because their residents were travelling temporarily on the night the census was taken, they were second homes, or they were for sale or rent.
He says the think tank Proper Australiauses water meter data from Melbourne water providers to estimate the number of homes that have sat unoccupied in Melbourne for an entire year, and that kind of information can be more insightful for the discussion at hand.
The similarity between the unemployment rate and rental vacancy rate
Next, here's a graph of the national unemployment rate that roughly covers the same time period as the rental vacancy rate graph above.
Notice how similar they look?
The rise and fall in the unemployment rate seems to loosely track the rental vacancy rate.
Dr Murray says the two are linked, and interestingly, the best way to understand vacant and unoccupied homes is to think about housing markets in the same way we think about the labour market.
For example, we know the unemployment rate doesn't actually measure the number of people without jobs — it only measures the number of people who are actively looking for jobs.
There can be millions of people who aren't working and who aren't counted as being "unemployed" because they haven't made themselves available for hire.
Dr Murray says the rental vacancy rate is a similar concept.
He says just as people can be "out of the labour force" for many reasons, dwellings can be unoccupied and not looking for a tenant for many reasons, and therefore be "out of the dwelling force."
"The unemployment rate measures the number of people currently not working but actively seeking work, while the rental vacancy rate measures the number of dwellings soon to be unoccupied and actively seeking a new tenant," he says.
"They are both measures of the state of the current market, either for jobs or for tenants."
He says in the same way that a booming labour market brings people into the labour force to look for work, a booming housing market brings unoccupied homes into the rental market.
Why? Because a low rental vacancy signals to property owners that, if they build new homes faster (or put unoccupied dwellings into occupancy), they'll be able to easily find tenants without having to reduce rental price.
But that process doesn't happen instantly, and it can lead to periods of high housing stress.
"It is common the hear that low rental vacancy causes rents to rise. But low rental vacancy rates are the mechanism by which rental prices adjust, just like low unemployment is the mechanism by which wages adjust," he says.
"Rising wages and falling unemployment are the same thing, measured differently. Rising rents and falling rental vacancy are the same thing, measured differently. They are both symptoms of the same equilibrium adjustment in the respective labour and rental markets.
"This similarity also explains why a 'rental crisis' is a predictable part of the business cycle and usually coincides with a booming labour market," he says.
"We can look, for example, back to the 1880s property boom and then the early 1890s bust in Victoria. The number of unoccupied dwellings in Melbourne rocketed to 14,000 in 1893, which is equivalent today to 150,000 unoccupied dwellings.
"But as the economy recovered in 1894 and 1895, this fell to 10,000 and lower. The vacancy rate for housing tracked the unemployment rate for people in that cycle," he says.
In short, he says, rental vacancy rates, like the unemployment rate, are useful for understanding which way the rental equilibrium is adjusting.
"But these metrics are not very useful for telling us whether there are physically too many or too few homes, or the state of unoccupied housing," he says.
How do we get more dwellings into the 'dwelling force'?
What does he mean by that last sentence?
He says there can often be plenty of already-constructed dwellings held back from rental markets, which the "vacancy rate" ignores.
"In one of Australia's first major build-to-rent estates, the Smith Collective on the Gold Coast (the former 2018 Commonwealth Games athletes village), the 1,251 already-constructed dwellings took nearly four years to be fully leased to renter households, despite low rental vacancy on the Gold Coast," he writes.
"The managers of the project explained to me in 2021 that 'the precinct has been on a staged released strategy so as to not flood the rental market'.
"They held hundreds of dwellings vacant for many years from the rental market, at a time when the vacancy rate in the area was around 2.5 per cent, near the 'natural rate.' Only in 2021 and later, when local vacancy fell to 0.7 per cent, did they accelerate renting the last buildings in the project.
"This is completely normal property market behaviour," he says.
He then talks briefly about the logic of vacancy taxes, when thinking about system-wide housing pressures.
"In Victoria, Australia, and British Columbia, Canada, vacancy taxes, or more accurately unoccupied home taxes, have been recently introduced," he writes.
"Property owners with homes held unoccupied for periods of more than, say, six months in a year, must pay for the privilege of excluding people from existing homes.
"The motives behind this policy are admirable.
"There is a commonsense logic to it, as captured by a 2022 article in the Investigate Europe magazine that noted: 'Europe has reached a point where there are too many people without homes and many homes without people'."
