The housing industry has become ground zero for the current policy debate over what demand and supply in a market economy really means. This has been the subject of further analysis given the impact housing supply and in particular rents have had on the April CPI.
Not surprisingly, concurrent with the increase in rents, increases in house prices were reported for May with CoreLogic data showing house values increased in every capital city. In Melbourne, house prices increased 0.9% to a median of $911,000, while Sydney’s 2.1% jump took the median value to almost $1.3 million. There was a 1.5% jump in Brisbane to $792,000.
RBA Governor Philip Lowe, in responding to questions at Senate Estimates, commented that the best way to bring down rents and house prices was ‘‘supply, supply, supply’’, but he also noted people would likely need to ‘‘economise’’ on their property choices. He said that could mean bringing in an extra flatmate or moving back in with their parents.
This statement was perhaps misinterpreted as it flows from some RBA research – reported in the April issue of Stats Count – showing that during COVID, the average household size in capital cities reduced from 2.6 to 2.55 residents as people sought more space during the pandemic. In macro terms, this meant an extra 140,000 dwelling units was required during a period when there was no migration.
So effectively, this excess demand is pushing prices and rents higher. To some extent, Governor Lowe is stating the obvious when he says supply is the answer, but that of course takes time. In the short-term, average household size will need to increase. Nevertheless, the cartoonists had a field day at the Governor’s expense.
Source: Golding The Age 1 June 23
However, this provides context for the current debate about what is happening to housing supply.
Christopher Joye, writing in the AFR, was very strong in saying the current increase in house prices is illusory and a 20% drop in prices from peak to trough is still forecast by some, with the decline at the moment only 10.5% – so plenty more to come. We know from past experience, declines in existing house prices have a knock-on effect to new housing demand. Housing approvals are now at their lowest level since 2012.
The level of approvals has been interpreted in some quarters – including a recent series in the Financial Review (AFR) on the “Housing Crisis” – as the crux of the problem. Lambasting planning approvals processes for causing the chronic shortage of supply. An analysis by Dr Tony Richards covered in the AFR on the 29th May suggests the shortfall over the past 20 years is some 1.3 million dwellings. The report comments:
“…housing supply has expanded at just 4.5 per cent ahead of population growth over the 20 years to 2021, much slower than the 17 per cent above the population increase in the previous 20 years.”
Maintaining the earlier pace would have increased the housing stock by 220,000 homes a year, instead of the actual growth of 153,000 a year.
Dr Richards’ simple conclusion is that planning processes have been slowed by councils and a general community attitude of opposition to new developments or NIMBYs (not in my backyard).
The impact of not keeping ahead of population growth can be broken down by State as follows:
The logic of this may seem compelling, as many of us may have direct experience of the planning process either as a proponent or an objector, and the time required for the system to make its decisions.
However, the data does not necessarily support this view. If you look at housing activity over the past 40 years (the period under Dr Richards’ analysis) there has been no effective change in the building completion effort (the red line). So as approvals, have reached all-time record levels during the current cycle, the problem with supply would appear to be the capacity of the industry to build that number of dwellings, not approval processes.
The building activity data has consistently shown that as approvals have increased during this period, the number of houses under construction has skyrocketed. However, completion rates have not changed significantly from historic levels of between 27,000 to 30,000 detached houses per quarter. As a consequence, all that has happened is that the time it takes to build a house has become longer. As at December quarter 2022 it had increased to 9.12 months up from 7.35 in the December quarter 2021.
The question of capacity has been touched on in the AFR series but assumes that additional workers and other resources can be found. Well, if it was as easy as that, then surely during the past 18 months the system of work would have expanded to meet the demand? Part of the solution may lie with more off-site manufacture with the use of pre-fabricated and modular construction. This is a complex area requiring further research to better understand what is causing the delays and in turn how we can reduce completion times and improve the system of work.
A further challenge is that Councils have provided approvals for new sub-divisions, but the flow of these projects/lots onto the market is driven by a combination of the developer/builder/banker’s interests. Prosper Australia published a report in July 2022 looking at this issue. The evidence suggests that the release of those lots is driven by the desire for an orderly market rather than meeting demand in full. This process known as land banking is a major structural challenge for accelerating supply.
Prosper Australia’s research examined data from nine Master Plan Communities (MPC) to map supply and demand.
Average sales per year turned out to be below demand with the result that prices were maintained at lower sales levels. To reinforce the point, the constraint here is not council approval processes, as the MPCs have been approved. The constraint here is the flow of land sales which appears to be governed by limited market forces. The report authors commenting:
“The complete record of over 25,000 sales in these nine sample subdivisions was analysed, revealing that sale rates in these large approved MPCs vary significantly according to market conditions. Markedly higher sales occurred when the market was running hot (mid 2015-mid 2017) compared to when the market cooled (18-19). The rate of sales, and hence the rate of new dwelling development, contracted 48.7% between the two periods as the promise of supply-led affordability was held at bay.”
The mix of dwelling type is also important and has the potential to make greater use of land in areas where people want to live. From a town planning perspective, this increase in density, in areas with established services and transport infrastructure, makes sense.
Reflecting the need for greater density, in recent years, the mix of dwelling units has continued to evolve with more townhouses, mid-rise and multi residential projects. Intuitively, planning challenges in established suburbs may accord with the NIMBY focus of the AFR series. However, it is not accurate to say this applied across the entire housing supply spectrum where it appears construction capacity and land banking are also issues.
In fact, the conflicting opinions were reflected in an article in the AFR on the 5th June, by Tom Rabe, under the headline “Builder shortages inflame Perth’s housing crisis”. The article comments that Local governments in Perth have been accused of “dragging their heels” on approving medium-density housing prompting the state government to reform planning regulations this year. Significantly, the building industry is quoted saying:
“Freeing up these processes will only result in a genuine lift in new housing if there is enough workers to undertake projects”
However, in none of the recent policy debate has there been recognition that in a market-based economy, housing affordability is fundamental to the conversion of population growth/household formation rates from latent to real demand for housing. The recent housing approvals which are the lowest since 2012 bears this out at a time when the population and demand for housing is much greater than 10 years ago. There are a range of measures of housing affordability including:
- The ratio of dwelling values to household income
- The number of years taken to save a 20% deposit
- The proportion of income required to service a new mortgage
- The proportion of household income to service rent
Using the data which is in the public domain on the first measure shows that affordability has deteriorated some 20% since COVID and for the period back to September 2011, the change has been 48%.
The only reason there has been a slight improvement in affordability in the 12 months to March 2023 is the decline in house prices.
So how to increase supply when the product is not affordable? This is where the mix of dwelling types and locations come into play.
It is also important to note that an area where supply has been significantly curtailed has been public housing (dwellings built by the public purse). Except for the GFC stimulus money provided in the Rudd/Gillard years, the decline in dwellings built by the public sector has been in continuous decline. A change in this space is also required and a lot of work is progressing on a National Housing Plan, with the framework currently before the Federal Parliament.
So, returning to our starting point.
While Governor Lowe is talking about the need for “Supply, Supply, Supply”, the challenge is how to make that happen, given the constraints on the housing construction system of work has only resulted in the nation taking longer to build the same number of dwellings.
Economics 101 tells us that demand exceeds supply both in housing and in the broader economy. Interest rates are the blunt tool which pulls money out of the economy and thereby reduces demand. But looking ahead, with the population continuing to expand as net migration increases, we need to also expand the capacity of the housing construction system of work.
