AFR Article: 6-7 November 2021 page 26
When you write about bricks and mortar, it inevitably attracts a lot of hyperbolic attention: you tend to be typecast as a preternatural “bull” or “bear”. To be clear, I am neither: my task is to try to accurately anticipate what will unfold.
After explaining that Aussie house prices would have to correct by 15-25 per cent if (heaven forbid) the RBA ever lifted its target cash rate by 100 basis points or more, some readers responded that they had never seen us predict price falls before. While I will address this misperception later, let’s first deal with the future direction of the $9 trillion residential real estate market.
More than a decade ago, we argued that the community should expect much more volatility from residential property because of the huge increase in the household debt-to-income ratio, which had made borrowers far more sensitive to interest rate changes. At the time, we asserted that this would generate a more frequent boom-bust cycle in prices as a result of variations in borrowing rates.
If you look at the chart of CoreLogic’s eight capital city hedonic index, you can see that substantial drawdowns in prices were relatively rare between 1980 and 2003. And yet since 2003, there have been six distinct episodes in which prices have declined with what appears to be increasing severity. It might come as a surprise that the single biggest fall in Aussie house prices over the past 40 years was the innocuous episode between September 2017 and June 2019 when capital city values dropped by a record 10.2 per cent care of APRA’s macro-prudential constraints on lending. The losses at this juncture were, in fact, much worse than those experienced during either the GFC or the pandemic-induced recession.
Since the end of the 2017 correction, capital city home values have climbed by a robust 30 per cent. The capital gains after the much milder COVID-19 retrenchment have been 21 per cent. Going back to the end of the 2010-2011 downturn, we find that homeowners have profited from a 72 per cent increase in the value of their most important asset. That means that dwelling values have appreciated at a circa 6 per cent annualised pace over the last decade. And that is at the overall asset (or property) level, assuming no gearing. Accounting for the use of significant amounts of leverage, the actual tax-free return on equity homeowners have captured has been much higher again.
The RBA has made it abundantly clear that it is going to be highly resistant to lifting its cash rate until it observes consistent annual wage growth of 3-4 per cent coupled with core inflation sustainably sitting at or above the mid-point of its target 2-3 per cent band. This implies that it will not touch rates until sometime between late 2022 and mid-2023.
We are still forecasting ongoing house price appreciation until the RBA hikes and/ or banks materially lift mortgage rates. More specifically, home values should climb by another 5-10 percentage points. So there is some upside left in this trade. Yet if and when the RBA does seek to normalise the cash rate, prices should fall, as night follows day. And if the RBA is able to lift rates by 100 basis points or more, it will likely be the largest correction on record.
Assuming rates increase relatively promptly over, say, a 12-month period, we would expect national home values to decline by 15-25 per cent. It is possible that the adjustment is smaller if the RBA moves more slowly and the value of residential real estate mean reverts partly via household income growth over time. But our central case would be a circa 20 per cent decline after the first 100 basis points of hikes.
It’s worth noting that if we apply the RBA’s internal housing valuation model to this question, we get somewhat larger numbers. The model developed by former RBA economists Peter Tulip and Trent Saunders, which we have replicated and refined, suggests dwelling values could drop by about 33 per cent following 100 basis points of hikes. While renters might embrace this prospect, homeowners would obviously rather avoid it.
Bull v bear backstory
I want to conclude with some comments on the relentless “bull” versus “bear” stereotyping. We are neither: we are just trying to divine the direction of the market.
By way of background, we were the first to call a 10 per cent correction in Aussie house prices in 2017, which is what transpired between September 2017 and June 2019. We were also the first to anticipate a 10 per cent rise in prices in April 2019, which is what materialised before the COVID-19 shock.
To the best of our knowledge, we were the only forecasters to predict both a modest 0-5 per cent drop in home values between March and September 2020 (they fell 2.7 per cent across the capital cities) and a subsequent 10-20 per cent increase in prices starting in September that year.
Since September 2020, capital city dwelling values have appreciated 21 per cent. Our 10-20 per cent forecast for future capital gains following a modest dip between March and September 2020 was predicated on the assumption of 100 basis points of rate cuts. Accounting for the steeper fixed-rate mortgage reductions that ensued care of the RBA lending $188 billion of ultra-cheap, three-year money to banks, we adjusted our expectation for the price rise to 20-30 per cent, which we are on track to achieve.
Going back further, we forecast that prices would soften in late 2010 after a series of aggressive RBA rate hikes, which they did (capital city prices declined by 6 per cent between late 2010 through to the end of 2011). Yet in 2010, doomsayer Jeremy Grantham (co-founder of global fund manager GMO) had other ideas, claiming that Aussie house prices would plunge 42 per cent. We bet Grantham $100 million against the CoreLogic index that prices would be higher, not lower, in three years’ time even though we were a little bearish on the immediate-term outlook. Over the period covered by this proposed wager, dwelling values did indeed climb by 5.8 per cent.
In early 2012, we got into a debate about whether the housing market was starting to recover: our data suggested it was, whereas others felt prices were still falling. We now know prices began appreciating in January that year.
In 2013 we argued that the RBA’s rate cuts would trigger a housing boom and years of double-digit price growth, which would eventually morph into a bubble. The RBA panned the proposition at the time. Yet that’s precisely what we got between 2013 and 2017, which eventually compelled APRA to aggressively intervene.
Finally, way back in 2008 we argued that the national housing correction wrought by the GFC would be modest, regularly debating the likes of Australian economist Steve Keen and others who predicted much more calamitous 30-40 per cent price falls. In practice, dwelling values retrenched by just 6.4 per cent in 2008 and promptly rebounded by 12.2 per cent in 2009.
So we have a bit of history with housing. My introduction to the topic was a result of co-authoring the 2003 Prime Minister’s Home Ownership Task Force report on the demand- and supply-sides of the market. I also co-founded a business that developed the daily hedonic house price indices that CoreLogic now publishes and the associated automated property valuation models that leverage off the same technology.
Christopher Joye is a portfolio manager with Coolabah Capital, which invests in fixed-income securities including those discussed in his column.
