Australia’s Mixed Diagnosis

by Ingolf Eide

Credit booms generate economic and financial imbalances. The longer a boom lasts, and the more extreme the expansion, the greater these will be.

The results show up in balance sheets and income statements. Whether it’s an individual household, economic sector or the external position of a country, that entity’s financial statements tell the tale.

In Australia’s case, sectoral debt figures show remarkably divergent trends in credit growth.[1]

At 73.1%, non-financial business debt as a percentage of GDP is only 7% higher than in 1988. The high point in the intervening years was 85.6% in Sept 2007. General government debt (federal and state combined) actually declined, from 42.1% in 1988 to 22.7% now. Prior to the crisis, it hit a low of 13.4%.

The real action was in the household sector, where debt to GDP started at 42% in 1988 and is now 109.7%.

In international terms, Australia’s non-financial business debt is middle-of-the-pack, our government debt perhaps the lowest in the developed world but our households rank near the top. There are countries with higher levels (Switzerland, for example, at 118% in 2007), but there aren’t many of them.

The growth in household liabilities far outpaced income. Debt to disposable income started at 62.8% and ended last year at 169.7%. And, despite historically low interest rates, 10.6% of disposable income still went to paying interest. In September 2008, that figure was 15.3%.

Concerns about household debt are generally countered by pointing to the other side of the ledger. After all, at 19.1% of total assets and 29.4% of housing assets, household debt doesn’t seem so bad.

It looks even more reassuring compared to the US where housing debt before the downturn was just over 40% of housing assets.[2] It’s a disparity that seems odd given household debt in the US is in relative terms lower than ours. The answer is in the value of dwellings relative to GDP.

Over there, even at the peak of their boom this ratio never exceeded 165%; here, it was 308% in December 2007 (up from 178% in 1988). So, homes here are almost twice as expensive in relation to the total economy as they were in the US at its market peak. Surprising as that may seem, it does at least fit with one recent survey in which six Australian cities figured amongst the top 10 most expensive residential real estate markets in the world.

♦  ♦  ♦

After Australia’s banking system was deregulated in the mid-1980s, a change in lending patterns was immediately obvious, although it took time for the train to gather speed.

Real estate prices closely paralleled the resulting growth in mortgage debt. It’s been (indeed still is) a classic asset based lending boom where rising prices (powered by expanding mortgage credit) in turn seem to justify ever more lending. These are sweet and seductive cycles that can seemingly spin their way to heaven, and in the years leading up to the crisis they played out in much of the developed world. Unlike everywhere else, though (except for Canada), after a brief hiccup when the crisis was at its most intense the cycle here resumed. So much so that residential real estate prices hit all-time highs in the last year.

Justifications are offered, of course, and there’s truth in at least some of them. For one thing, Australia never overbuilt like the US. Indeed, quite the contrary according to all those who lay the blame (or the credit, depending on your point of view) for persistently high prices at the door of a chronic supply deficit.[3]

Perhaps. Certainly, the rental market here is for the most part tight. Still, even those parts of the US that were similarly constrained by geography and other restrictions (like San Francisco[4]) were hit hard once the cycle turned. Prices are always made at the margin, and demand can undergo dramatic shifts when unemployment takes hold and broader economic conditions deteriorate.

♦  ♦  ♦

A recent study provides strong evidence that the degree of household leverage is tightly correlated with subsequent economic performance. Mian and Sufi (University of Chicago and NBER) analysed the debt growth of US households, not only in aggregate but also in detail across 450 US counties.

“[T]he recession both began earlier and became more severe in high leverage growth counties relative to low leverage growth counties. The top 10% leverage growth counties experienced an increase in the household default rate of 12 percentage points and a decline in house prices of 40% from the second quarter of 2006 through the second quarter of 2009. In contrast, the bottom 10% leverage growth counties experienced a modest increase of 3 percentage points in the default rate and a 10% increase in house prices.”

Ditto for durables consumption, residential investment, and rates of unemployment. For all these measures of economic activity, they found “the correlation between leverage growth and the severity of the recession is robust to county-level control variables for demographics, cyclicality, and industrial composition.”

The study is a useful reminder that credit is in no sense neutral; leverage always has real-world consequences, some of them quite severe.

Only time will tell if things eventually play out in a similar fashion here. One thing, though, is certain. Household balance sheets are incomparably more stretched than they’ve ever been before and we’re therefore singularly ill-prepared for tough times. Or for higher interest rates. At these levels of leverage, the impact of a serious rise in rates hardly bears thinking about.

♦  ♦  ♦

Ploughing through this local data threw up some interesting trends. The one that most surprised me was the long-term shift in household disposable income in relation to GDP.

Over the last 50 years, it’s fallen from just under 80% to its current level of 62.9%. As can be seen from the following chart, the slide accelerated post-1983 and in the early years of the last decade.

Throughout, consumption stayed relatively constant with the most recent level of 55.8% of GDP only slightly below the starting figure of 58.1%. Clearly, household savings had to suffer. Until 1983, they were usually over 10% of GDP but from then on it was all downhill with negative prints for quite a few quarters between 2002-5. The crisis (such as it was in Australia) briefly lifted household savings to 4.6% but by the March quarter this year they were back down to 1.6%.

We don’t fare all that well by international savings standards although the UK has in recent years done distinctly worse. Only Germany (in common with most other European nations) managed to sustain household savings at historical levels. It’s curious that Anglo-Saxon countries have been so uniformly prone to debt accumulation in the last few decades. Perhaps it flows at least in part from their approach to financial deregulation.

As we saw from the earlier chart, while household disposable income and savings rates steadily fell in relation to GDP, the gross operating surplus of private and financial corporations rose from about 15% to 22.3%.

In simple structural terms, therefore, households have been the big losers in relative income and in absolute savings over the last 50 years. The corporate sector picked up some of what they lost but net national savings still fell substantially. In the early 1960s, they ranged from 13-16% of GDP whereas for most of the last decade they’ve been only 6-10%. Post crisis, burdened by government deficit spending (and more recently by the renewed drop in household savings) national net savings hit a new all time low of 4.65% in the December 2009 quarter.

No mystery, then, about why our current account deficit is so tiresomely chronic. With these savings rates and a continuing high rate of investment[5], no other outcome was possible. Net external liabilities of $757 billion (or 59.5% of GDP) as of the end of March this year are the cumulative result.

Hardly in the Greece or Portugal class but still up by almost 20% of GDP in the last 20 years. The need to refinance these obligations as they fall due (while also funding the ongoing current account deficit) does make us much more vulnerable. Particularly in a world where ongoing capital flows can no longer be taken for granted.

♦  ♦  ♦

In “Diagnosing the Disease” I suggested the failure to accept that much of recent economic growth was inherently unsustainable badly distorted policy responses to the crisis.

The ready availability of credit (both internal and external) over recent decades enabled investment in Australia to grow while spending on consumption held at historical levels. It also helped push asset prices far higher in real terms. So much so that some, like residential real estate, are now radically out of line with the underlying economy.

This process feels good while it lasts, of course, and for us it hasn’t quite ended. Unlike much of the rest of the world, our bill has yet to be presented; we’re still lingering over coffee and cognac.

Whether the investments we’re making bear fruit remains to be seen; so much is directly or indirectly dependent on the China story. However that may turn out (and I do think the resources boom is probably due for a rude, albeit temporary, interruption), at least this spending stands a decent chance of producing adequate returns.

Not so for all the consumption done with borrowed money, where the households who succumbed have a long and painful road ahead of them. Nor for much of the more extravagant real estate investment; it no doubt brings great pleasure to its owners but it adds nothing of note to our productivity. In essence, it too is consumption. While we may indeed have a housing shortfall, much of what we do have far exceeds its primary role as accommodation. Given much of this extravagance has also been put on the tab, it’ll weigh us down for many years to come.

♦  ♦  ♦

A few basic conclusions seem inescapable.

Households can’t continue on their current path for much longer. No one knows exactly how much debt is too much, but it would be foolhardy to argue that we aren’t already pushing the boundaries. In any case, to the extent the borrowing didn’t go into productive investments, those households who took it on will have no choice but to consume (and invest) less in the future.

Our government should accept that household debt needs to be wound back. It can then facilitate the necessary adjustments rather than trying to impede them by encouraging more consumption and borrowing. It can also work towards bringing housing prices back into a more sustainable relationship with the underlying economy rather than pumping them up with grants and subsidies. And, if indeed we do have a real housing shortage, focus on sensibly boosting supply.

Given the sheer extent of the excesses and imbalances both here and more particularly internationally, the real crisis almost certainly still lies ahead. Squandering resources on makework schemes and grand infrastructure projects of questionable value therefore seems shortsighted at best. Yes, we’d all like to avoid pain entirely, but the history of our actions makes this impossible. To pretend the problems are minor, and to as often as not ignore the most pressing ones, merely guarantees greater difficulties down the track.

Far more useful, surely, for the government to work out how best to respond if (or in my view when) the next storm hits. To prepare plans and administrative capacity so that when it does the most vulnerable can be adequately protected and the way smoothed for the rest of us to make the necessary structural adjustments. Education and retraining capacity, support networks, potential investment projects that could generate real returns while also boosting employment, carefully targeted supplementary welfare schemes; these are the sorts of things they could usefully plan for. And, while we wait to see how many of them may be needed, fiscal capacity should be tended carefully rather than splashed about.

Despite all the challenges that lie ahead, we needn’t be unduly gloomy. With a bit of good sense and decent preparation, they’re  entirely manageable. Indeed, by world standards we’re in an enviable position. Whatever we end up suffering is likely to be a fraction of what others must endure.

For that we must be grateful to accident, geography and abundant natural resources. And, at least in part, our own efforts.


1 Unless otherwise noted, all data is from the ABS or the RBA.

2 The sharply declining real estate prices since then lifted this ratio to almost 60%, a timely reminder that debt (until it’s either paid back, re-negotiated or defaulted on) is a fixed amount while the asset prices it funds are anything but.

3This argument is so ubiquitous it’s hard to believe there isn’t a good deal of truth in it. From a macro point of view, though, it is slightly puzzling. Dwelling investment over the last 50 years ranged between 3.8-6.8% of GDP. During the peak baby boomer household formation period it averaged around 5.5% as against 6.1% over the last decade and yet we still apparently suffer from chronic underinvestment in residential real estate. Perhaps I’m missing something obvious. For some perspective the Federal Reserve Flow of Funds figures show residential investment in the US during the boom years of 2004/5 was 5.75% of GDP.

4 Having roughly tripled in the decade leading up to the peak in mid-2006, prices in San Francisco very nearly halved in the next three years.

5 I was a bit surprised at how high Australia sits in the international rankings of gross fixed capital formation as a percentage of GDP. In 2008 (the most recent data from the OECD Factbook 2010), only two countries ranked higher, China at 41.1% and India 34.5%. Our 29.4% for that year is up by 4-6% from the levels of the 1990s and the first few years of the noughties. As an aside, the fact that investment here didn’t falter at all during the crisis was a critical factor in keeping our downturn so shallow and so brief. In the US, by contrast, gross investment fell by about 3.5% of GDP between 2008 and 2009.