In comparative terms, Australia has only been mildly affected by the economic crisis. Whatever’s enabled us to sidestep the worst of it so far, it’s still fair to wonder whether this good fortune can last.
A recent research update by two University of Chicago professors (hat-tip Calculated Risk) might have some relevance to this question. In a brief article at Planet Money (Lessons From The Fall), Atif Mian and Amir Sufi discuss the study:
The central lesson we as economists have learned from the crisis is that an unsustainable increase in household debt is one of the most serious threats to the U.S. economy. Going forward, policy-makers, regulators and researchers must recognize the central role of household financial health in causing economic turbulence if we are to avoid repeating this painful episode. This is not the first time excessive household leverage preceded a severe economic downturn, and the effect of household debt on the economy is not unique to the United States.
Rather than simply looking at US statistics in aggregate, they compared household leveraging and subsequent economic performance across counties:
A cross-sectional analysis of U.S. counties shows that areas with modest increases in leverage from 2002 to 2006 have experienced only a minor economic downturn, whereas counties with large increases in household leverage from 2002 to 2006 have experienced a severe recession. Our findings suggest that the process of household de-leveraging is likely to be the major headwind facing the economy going forward.
The consequences have been dramatic; the most highly leveraged decile, for example, saw their unemployment rate (to the end of 2008) rise by 4%, while for the least leveraged it rose 1.3%. More generally: “Default rates for low credit-quality homeowners rose by more than 12 percentage points in places where housing was scarcest and prices had risen most. In elastic cities, by contrast, the increase was less than four percentage points.”
Does any of this, interesting though it may be, have any relevance for us?
I think yes, but then again I’ve thought for a long time we were vulnerable and been surprised at how we have to date sailed through relatively unscathed. I’ll be interested to see what others think about why, and how, we’ve managed to do this.
First, though, a few of the figures that underlie my doubts:
- Our residential real estate/GDP ratio was 306.7% as at the end of March this year. At the peak in the US housing market, their equivalent ratio was about 165%.
- In terms of household indebtedness, as at the end of the first quarter ours was 119.5% of GDP, the US about 98%.
- Net external indebtedness is $633 billion ($726 billion if you include the equity deficit).
- Our current account deficit crept up to 4.5% in the June quarter after briefly getting down to 2.6% and 2.1% in December and March quarters as imports plummeted while iron ore and coal exports were still at the old contract levels.*
(These latter two sets of figures also suggest our dependence on the China story turning out well).
One thing the Mian and Sufi report reinforced was the tight correlation in the downturn of almost everything in the US (and elsewhere) to the fall in residential real estate. Not exactly news, but this dynamic may go some way towards explaining why we’ve escaped a more serious downturn.
What it doesn’t tell us, of course, is whether the relative buoyancy in our housing market will continue. According to the RBA figures, the value of dwellings fell by 3.8% in the year to March 2009 but things have apparently ticked up a bit in recent months under the influence of some combination of the FHBG, low rates and a growing belief that we may have seen the worst.
For a bit of perspective, this chart (from the National Housing Supply Council State of Supply Report) provides a long-term overview of real house prices, rents, construction costs and household income. Quite a remarkable picture.
Will it continue? Can it continue? And if housing does finally turn down in a more serious fashion, will the broader economy follow as it has elsewhere?
* (all Australian figures from RBA and ABS / US from the Fed Z.1 Flow of Funds report)
I probably should read the paper before commenting, but two point:
1 unsustainable levels of debt in any significant sector of the economy are a danger. The trick is identifying when they are unsustainable. With reference to home mortgage debt in Oz, how relevant are the figures when the economy grows and the amount of money available for discretionary spending increases? The comparisons to the past need to be adjusted for that reason, assuming the adjustment is possible.
2 is the difference between the US counties partly an effect of their starting point, poor counties could well have less leverage because they are poor and less of a fall in the recession for the same reason.
So, an extended thread about Krugman and one comment so far about real estate. What are we to make of that, Pedro? Seems a bit counterintuitive.
On your points, no argument; “unsustainable” is easier to specify with hindsight. Still, I think one can say that certain trends are unsustainable and that it’s therefore merely a matter of when they will end rather than whether. I guess this crisis should in any case help us to fine up our knowledge a little on where the boundaries may lie.
Part of the problem, of course (indeed, in my view by far the largest part) is that the outer edges of unsustainable were greatly extended this time around by government actions. The various safety nets, and more generally the growing sense that governments “had our back” so quieted traditional perceptions of risk that the tree did almost grow to the sky. Indeed, I’m still not wholely convinced that faith has been adequately shaken; the apparent “success” of various fiscal and monetary efforts may, if anything, have boosted belief. A scary thought.
I think “adjustments” as you’re using the word are already built-in to the various debt ratios. By comparing debt to GDP, income, assets and so on, we take account of both real economic growth and nominal changes. Or have I misunderstood you somehow?
What I mean is that the affordability of housing as a percent of income now is not the same as 30 years ago when a greater percentage had to be spent on, say, food.
You see the same argument about health care spending. Part of the reason it grows as a percent of GDP is that we can afford it. Better health is, in a way, a luxury good. A bigger or better located house is also a luxury good. I’m not sure if that is factored into housing affordability comparisons which usually seem to be simply based on median household income or whatever. Same goes for comparisons between here and the US. I think our housing is more expensive for supply cost and constraint reasons, but that does not mean we cannot afford to buy houses at the higher prices, although other consumption will likely be foregone.
Right, I see what you’re getting at. I guess it has to be a factor, particularly when things are being viewed over the very long haul.
Conrad raised a somewhat similar point in an earlier thread. As I said then, the size of the deviation over the last decade (as per the chart above) probably tends to dwarf such considerations in this instance. Unless, I suppose, one were to view it as a one-time adjustment to previously irrational underinvestment in real estate. Given the extent of the recent mania, and the sizeable shift in price/rent ratios, I struggle with that notion.
Some researchers have been thinking along similar lines to you and built in an adjustment to long-term price series in an attempt to account for changes in quality. This was the case, for example, in the long term chart (1880 – present) included in Steve’s Keen’s essay “Lies, Damned Lies, and Housing Statistics” where Stapleton deflated the housing prices by 0.6% p.a.
I expect you’re correct that increase in available income can’t explain all of the housing cost increase. Perhaps part of the recent jump is also explicable by the fall in interest rates. Given the near ubiquity of borrowing, house prices and house costs are not the same thing.
What has gone up the most appears to be the cost of land, though it could be that the construction price line is also hiding renovation effects. At any time the vast majority of the residential property being sold is already built, but a good percentage of that will be subject to ongoing improvements that might not be reflected in the growth of the cost of new housing (I hope I’ve expressed that idea clearly).
Having spent the last few days looking for arguments about head works charges that have more than doubled in a few years I’m not surprised at land price increases. New land is now heavily loaded with fees and development costs and existing (and often preferred) land gets pushed right up as a consequence.
“Unless, I suppose, one were to view it as a one-time adjustment to previously irrational underinvestment in real estate. Given the extent of the recent mania, and the sizeable shift in price/rent ratios, I struggle with that notion.”
Yes, though perhaps it would be wrong to characterise the past as an underinvestment. Prices are set by supply and demand and there definitely seem to be issues on the supply side. Perhaps people would have paid more in the past if they needed to, but luckily didn’t.
Sorry, but I just thought of something else. It is tempting to look at the past and see a bubble. I’ve worried like that myself. But a bubble exists if prices are at risk of a sudden large decrease because of the realisation that the prices cannot now be sustained or increase. The fact that people paid less for a house in the past is not really very good evidence that a bubble exists now. I think you would need to see both an oversupply of houses and other things that indicate prices are being pushed up unsustainably, such as by stupid financing. Not sure either factor can be found here.
I do think that a change in regulatory approach to land development would see a long term reduction in real house prices, but the supply of housing is too slow to make that precipitous and the time horizon for the more attractive property is even longer. First house on the estate has a way different environment than house #1000. I expect the profit margin on an S class merc is very different from a commodore.
Your points seem reasonable, Pedro, although I don’t know enough about the housing market to comment in any detail. A few general thoughts follow.
Most analysts and commentators seem to agree that the market here is quite tight (unlike the situation in the US, for example, where they ended up overbuilding badly). Quite how tight is disputed. Nevertheless, almost 10% of dwellings are unoccupied (according to that National Housing Supply Council State of Supply Report) so I guess it very much depends on the nature of those that stand empty (and why) and whether that’s likely to change.
So much depends on economic conditions: if they remain relatively good, then the declining average number of occupants per dwelling may continue; should they take a turn for the worse, that trend could very quickly reverse with dramatic results on the whole supply/demand equation.
There’s a reasonably interesting paper at the RBA site containing a speech (with charts) given by their Head of Economic Analysis at the 4th Annual Housing Congress in March this year. He’s relatively sanguine, but what you may find most interesting are some of the charts. One, for example, shows net housing equity injection from 1980 to present. For about six years from 2001 to 2007, there was net equity withdrawal averaging probably about 2.5 to 3% of disposable income. Nothing like as extreme as in the US, but it does give an indication that they were definitely some real boom attitudes here as well.
Very interesting Ingolf thanks. Glad to say that we took a bullish line early in the 00s and then sold out all but out home by mid 2007. Having decided this year to start looking again, hoping for bargains, we quickly found that there aren’t any in out area in the inner west in Brisbane. Rats.