China our one saving grace as Australia’s debt-driven love affair with house prices faces the chopping block
Reality came to reality TV in Australia last week, when three of the four properties in the much-hyped 'Flip that House' program The Block failed to sell at their nationally televised auction. A 400-person live audience, watched by over 3 million TV viewers, couldn’t entice more than one person to part with money rather than eyeballs.
As The Sydney Morning Herald observed: "Whatever the lure of a celebrity house, the would-be buyers in Fitzroy Town Hall were just as jittery as the would-be buyers at any other auction in recent weeks."
The remaining three properties sold in the week after the sale, but at a substantial loss compared to the initial purchase prices plus the sums expended on them by the four couples in their two months of televised renovations (and let’s not mention the advertising budget).
So is the chopping of The Block a sign that the days of ever-rising house prices are over? Not if you listen to Christopher Joye's prediction that house prices will rise by 55% in a decade. The median forecast of the “21 leading market economists” he polled was for 5 per cent growth in nominal house prices per annum for the next ten years, which Chris notes would suggest “that they will likely be 55 per cent higher in ten years’ time”.
Good luck with that. As Chris notes, my forecast wasn’t included, but it should be no surprise that I expect a fall in house prices of about 40 per cent over the same time period.
I differ with the 20 who predicted positive price growth for one simple reason: I focus on the role of debt in driving house prices. Having argued that debt drove prices up over the last fifteen years, I now expect debt to drive them down again.
The mechanism is simple – but it’s not part of conventional 'neoclassical' economics, which is why Chris and his surveyed market economists don’t consider it. Aggregate demand is the sum of income plus the change in debt, and this is spent on both goods and services and assets. There is thus a link between the change in debt and the level of asset prices (and the fraction sold, and the quantity produced, but I’ll focus on just house prices for now).
Going one step further, the change in aggregate demand is the change in income plus the acceleration of debt. There is thus a link between the acceleration of debt and the rate of change of house prices. If this relationship is strong, then rising house prices require that the rate of growth of debt rises over time.
So just how strong is the relationship? Using the RBA’s data on mortgage debt from 1992 until now (there was a break in the series in 1991) and the ABS House Price Index, the correlation between accelerating mortgage debt and the change in real house prices is 0.53 and highly significant (see figure 1).
Figure 1: The Mortgage Debt Accelerator and change in real house prices
The acceleration in mortgage debt has been volatile, but on average positive. For two decades, mortgage debt has accelerated at 0.5 per cent of GDP per annum. Can that continue for the next ten years?
No way. That sustained acceleration of debt has caused mortgage debt to rising dramatically, from less than 30 per cent of household disposable income in 1991, to a peak of 135 per cent of disposable income early in 2011 (see figure 2).
Figure 2: A 4.5 times increase in mortgage debt compared to disposable income over 2 decades
That’s a 4.5-fold increase over 20 years, compared to the 50% fall in mortgage rates across the same period.Simply paying the interest on outstanding mortgage debt now consumes over 8% of household disposable income, versus 4 per cent back in 1991 – and less than 2% in the 1970s.
Figure 3: A fourfold increase in mortgage servicing vosts since 1980
The situation is worse when debt repayment is taken into account. The cost of paying a 25-year variable rate mortgage on the average first home loan has risen from 45% of average weekly earnings in 1991 to 63% now – and it peaked at 74% of AWE before the 'unexpected' global financial crisis forced the RBA to drastically cut rates in 2008.
Figure 4: It now takes 2/3rds of the average wage to become a First Home Buyer
Chris realistically observes that household leverage can’t rise any further, but implies that this is neutral for house prices. But stabilising debt is not neutral for house prices: since debt levels have risen until now, a stable debt level in the future means decelerating debt and falling house prices. Figure 5 shows that the deceleration (on an annual basis) began in October 2010, and it has gathered pace since.
Figure 5: Mortgage debt decelerating
If rather than stabilising debt, Australian households start to reduce their debt as US households have done (see figure 6), then house prices would need to defy the gravity of decelerating debt to keep rising at the 5 per cent nominal rate (roughly a 2 per cent real rate) that Chris Joye predicts for the next decade.
Figure 6: Mortgage debt in the USA is now falling
Of course, this is Australia, where the world is upside down: maybe 'this place is different'?
It will need to be, if the US post-bubble experience is anything to go by. The relationship between mortgage debt acceleration and change in house prices has held up through the ups and the downs of the US market since 1986 (with a correlation of 0.8) – see figure 7. The US experience since 2006 shows what is likely to happen here as the debt bubble that fed the housing bubble finally comes to an end.
Figure 7: Debt acceleration determines change in US house prices
The final retort to the argument that house prices will crash here as they have elsewhere is that there hasn’t been a bubble here, so a crash can’t happen. Chris acknowledges that house prices have risen faster than disposable income per household in Australia, but attributes that to rational rather than bubble factors:
"By way of historical context, disposable income on a per household basis has averaged a healthy 5.8% per annum over the last ten years, and 4.9% per annum over the past 18 years.
"Yet for a range of reasons that I have explained many times before – including the once-off, 40% plus reduction in nominal interest rates over the 1980 to 2011 period – historical house price appreciation has consistently outperformed disposable income growth.
"For example, we estimate that between 1982 and 2011 median Australian house prices rose at a 7.7 per cent compound annual growth rate" (Property's fine forecast,August 25).
Firstly, as noted earlier, a 40 per cent fall in interest rates can’t explain the 4.5-fold increase in the household debt-to-income ratio. Secondly, the argument that debt levels have risen because interest rates have fallen can’t explain why debt levels were much lower in the 1960s when interest rates were also lower than today. If households responded rationally to the fall in rates by increasing debt levels in the 1990s, why didn’t they respond rationally to the increase in rates during the 70s by reducing debt levels?
Mortgage debt almost doubled as a percentage of household disposable income from the mid-1970s until the early 1990s, even though interest rates (adjusted for inflation) increased from minus 4% to over 10 per cent across that period. The debt ratio also increased from 75% to over 120%between 2001 and 2008, when real mortgage rates rose from 0.5% to 6.4% (see figure 8).
Figure 8: Mortgage debt rose before real interest rates fell
Rather than changes in debt levels reflecting rational, equilibrium responses to changes in interest rates, the growth in mortgage debt was the fuel for a Ponzi scheme that has propelled house prices far faster than incomes have risen.
Bubbles upon bubbles
There have been three big bubbles in Australian housing in the last 50 years, all driven by accelerating levels of private debt: the late-60s to early-70s bubble focused on Sydney; the 1988 bubble when the second incarnation of the First Home Vendors Scheme transferred speculation from the busted stock market into housing; and our current one since 1997, which has been driven by accelerating mortgage debt and government policy – by both Liberal and Labor – with the First Home Vendors Scheme being used to give the economy a sharp stimulus to avoid recession.
Figure 9: Bubbles upon bubbles in Australian housing
These bubbles have been built on each other only because the debt has continued to accelerate. But now that Australia has reached a mortgage debt to GDP ratio that exceeds the worst ever experienced in the US (see figure 10), the days of accelerating mortgage debt are over – and so are the days of prices rising faster than incomes.
Figure 10: Mortgage debt grew faster in Australia than in the USA
Even to simply eliminate the impact of the last bubble that began in 1997, prices would need to fall 40% (compared to incomes) from their current levels. Australia is now starting to experience the same process of debt deleveraging and falling house prices that America has been mired in for the last five years. The one saving grace we have is China – so long as China continues to grow and drive demand and prices for our raw materials. But as recent economic data has indicated, even China may not be enough to stop unemployment rising in Australia, now that Australia’s debt driven love affair with house prices is on the chopping block.
1. The Credit Accelerator is a concept that I’m still refining; here it is defined as the change in the change in debt over a year, divided by GDP at the midpoint. For this reason the data in figure 1 runs out in mid-2010.
2. My data comes from the RBA; Chris’s comparable figure clearly uses different data to conclude that “Australia’s household debt-to-disposable income ratio has, in fact, flat-lined since a number of years prior to the GFC”, since RBA data shows it rising from 115 per cent to 135 per cent from 2005 until now. The 4.5 times increase in debt versus the 50 per cent fall in mortgage rates also gives the lie to the argument that the rise in debt simply reflects a fall in interest rates, as Glenn Stevens noted: "The rough statistic that I have quoted many times was that the average rate of interest was about half; that meant you could service twice as big a debt. Guess what? That is exactly what occurred, and that had a very profound effect on asset values." (Glenn Stevens, Standing Committee on Economics, Finance and Public Administration, Februart 21, 2007, citing Reserve Bank of Australia annual report 2006, p. 26).
Click here for this post in PDF. Click here for the data in this post.
Steve Keen is Associate Professor of Economics & Finance at the University of Western Sydney and author of Debunking Economics and the blog Debtwatch.
This article originally appeared on Business Spectator.