Riding the great debt elevator

Australian homeowners are once again starting to lever up, which explains the recent uptick in house prices. But the debt ceiling has almost been reached and house price bulls may yet be proved wrong.

Last week I described the United States Flow of Funds as the "gold standard for national financial reporting”. That’s in part because it was designed by a gold standard statistician-economist -- Morris Copeland -- during the Golden Age of business cycle studies, before the moronic nonsense of "Real Business Cycle” theory decimated the area. Copeland’s objectives were to help understand the dynamics of the American economy by documenting "who paid for what and why” using a system of double-entry tables that divided the American economy into 11 sectors, and to explain the crucial role of banks in the economic process.

He was also utterly comfortable with the proposition that the banking sector is far more than just a mere intermediary between savers and borrowers—the Loanable Funds fallacy that Ross Gittins regurgitated this week when reporting a speech by RBA Deputy Governor Guy Debelle. This simplistic model of money gives banks no driving role in the economy: they simply facilitate exchanges of already existing money between savers and borrowers, and they only exist because they can do better research than individuals can into the credit-worthiness of potential borrowers. As DeBelle puts it:
    Why do financial intermediaries exist? Why don't I as a saver lend directly to you the company?
    One of the main reasons is because of the presence of what economists refer to as asymmetric information. Asymmetric information arises when I know more about my own situation than you do. It would be difficult for individual savers to know whether they were lending to a business that was going to put the funds borrowed to a productive purpose and be in a position to repay the loan when it became due.

Copeland’s analysis of the role of banks was far more dynamic: banks created credit during an expansion and eliminated it during a contraction. Banks were not mere warehouses of pre-existing money—as in the Loanable Funds fantasy—but producers of credit who played a vital role in the business cycle. New loans during an expansion constituted a flow into the money supply, while the contraction of debt during a slump was a deduction from that supply:
    The banking sector influences other transactors … through its participation in financial moneyflows. This participation is two sided. Its main form during an expansion is a concomitant increase in bank credit (portfolios) and in currency and deposit liabilities… The concomitant increases in bank credit and currency and deposits constitute a financial moneyflow from the first (more inclusive) group to the second through the banking sector. There is a converse financial moneyflow during a contraction.

Figure 1: Copeland's vision of monetary flows in the economy as a circuit


Copeland’s labour and vision informed the construction of the US Flow of Funds, making it the most comprehensive and well-organised source on the financial dynamics of an economy on the planet. Unfortunately, this vision wasn’t replicated in the rest of the world, so economic data on the monetary system is patchy in the rest of the world—including Australia.

That said, the RBA does do a good job of collecting monetary data and making it available through its Statistical Tables, and its data collection is monthly rather than quarterly as with the Flow of Funds. But there are omissions, one of which is crucial.

The Flow of Funds breaks lending by banks down into lending to 5 domestic non-financial sectors and 12 financial sectors (see Figure 2).

Figure 2: L.1, the key aggregate debt table in the Flow of Funds


The RBA, on the other hand, publishes data on bank lending to households (with that broken down into owner-occupier and speculator mortgages plus personal debt), businesses and government, but doesn’t indicate whether the lending to businesses is to non-financial businesses only, or financial plus non-financial businesses (see Figure 3).

Figure 3: An extract from RBA Table D02


On the RBA’s published data, Australia has a much lower level of private debt that the USA. But I don’t know whether that’s because the debt is lower, or because the RBA isn’t publishing bank lending to non-bank financial institutions in that data. So though I can try to compare one to the other, I can’t know whether the comparison is valid, and I can’t rule out the possibility—as shown in this figure from Morgan Stanley & Haver Analytics—that Australia’s debt level is simply under-reported.

Figure 4: Morgan Stanley international debt comparisons


That said, the actual dynamics of Australia’s private debt go a long way to explaining the different economic fates of the two countries since the crisis began -- and only the possible fact that Australia’s aggregate private debt level is lower than the USA’s is flattering to Australia.

Australia’s private debt to GDP ratio is falling, though obviously far more slowly than the USA’s, and this is because the rate of growth of debt is now below that of nominal GDP -- not because debt is actually falling as in the USA. The primary reason why Australia hasn’t had a debt-deflationary experience is that it is one of the few countries in the Western world where private debt is, in the aggregate, still rising.

Figure 5: Australian and US aggregate private debt


Australia’s private debt to GDP ratio is falling, though obviously far more slowly than the USA’s, and this is because the rate of growth of debt is now below that of nominal GDP -- not because debt is actually falling as in the USA. The primary reason why Australia hasn’t had a debt-deflationary experience is that it is one of the few countries in the Western world where private debt is, in the aggregate, still rising.

Figure 6: Australia's level, change and acceleration of private debt


The contrast with the USA is quite striking: whereas US private debt plunged from growing at $4.3 trillion per annum to falling at $2.5 trillion p.a. over two and a half years (Figure 7), Australia had just one quarter where aggregate debt levels fell -- December 2009 (see Figure 6).

Figure 7: USA's level, change and acceleration of private debt


Consequently, there has been no depressing effect on aggregate demand in Australia from falling debt: we almost hit the zero point in 2010, but then bounced—especially when rising government debt is also considered. Aggregate demand fell -- from A$1.4 trillion toA$1.3 trillion -- but it didn’t plunge, and by 2011 Australia had exceeded the pre-crisis level of nominal aggregate demand.

Figure 8: Aggregate demand in Australia

On the other hand, even Obama’s stimulus program couldn’t prevent a US$6 trillion turnaround in aggregate demand in America, from $19 trillion in 2008 to $13 trillion in 2010.

Figure 9: Aggregate demand in the USA


So Australia learnt a valuable lesson from the crisis (irony alert): you can stay out of debt trouble if you just keep on borrowing.

The difference in the Credit Accelerators for the two countries is also marked.

Credit plunged so rapidly in America that the acceleration of private debt was for a while equivalent to minus 25 per cent of GDP. However a slowdown in the rate of reduction in debt then led to a significant rebound in debt acceleration -- even though debt was still falling. Since accelerating debt adds to the change in GDP as a source of rising demand, America -- or rather Wall Street -- got a significant boost in 2011. However that boost is now petering out, and I expect whoever is in the White House next year will find that growth stubbornly remains anemic.

In Australia, the decline stopped at about 12 per cent of GDP, and the rebound was limited too -- so we avoided a deep slump, but the stimulus from credit since has been minimal during the Mining Non-Boom.

Figure 10: Credit accelerator


The main sectoral difference in debt growth between the two countries has been the behaviour of households. All segments of US society have delivered significantly, including households. In Australia, in contrast, the household sector delivered only briefly -- before the First Home Vendors Boost weaved its magic -- and then levered up. Now that the FHVB has worn out, mortgage debt began to decline, but it is now growing slightly faster than nominal GDP.

Figure 11


These differing mortgage dynamics also explain the huge difference in how property prices have behaved. Mortgage debt decelerated massively in the USA -- starting in mid-2006 -- and hit an unprecedented level of minus 6 per cent of GDP in 2009. Since then the deceleration has declined -- even though mortgage debt is still falling -- and it has recently turned positive, pushing house prices up as a result.

Figure 12


Figure 13


Australia’s mortgage debt deceleration was stopped in its tracks by the First Home Vendors Boost, leading to a post-GFC house price boom that drove real prices up 18 per cent over 2009-2010, leading to a peak in June 2010. They fell about 10 per cent in real terms but are rising now -- in part because, though mortgage debt is rising at its lowest rate since records began, it has recently accelerated.

Figure 14: Mortgage debt accelerating even though its rate of growth continues to decline


It’s hardly a secret that I’m a property bear. I remain so despite the recent uptick, because the credit acceleration explanation I have for price changes is confirmed rather than contradicted by the price rise: from my analysis, rising house prices require accelerating mortgage debt, and that’s precisely what we have recently experienced.

Figure 15


It’s also hardly a secret that the property bulls are crowing (now there’s an interesting mixed metaphor) over the recent price rise. But the Dirty Harry question for them is, if I’m right and rising prices require accelerating mortgage debt, then just how many bullets does the Magnum 44 of the Australian mortgage market have left? That gun requires a lot of headroom to work reliably, and it was there in abundance when mortgage debt was just 20 per cent of GDP back in 1990. But now that it is 85 percent, the room for sustained acceleration simply isn’t there.

However, one way in which I am quite happy to concede that "Australia is different” is that our floating rate mortgages enable the RBA to take pressure of mortgagors -- and hence encourage the growth of mortgage debt -- by cutting the cash rate. So even though I think the cash rate should be lower -- much lower -- than it is, I know that has the potential to set off the mortgage acceleration dynamic once more.

The combination of an enticement into debt by RBA rate cuts, and a massive mortgage debt overhang that puts Australian households amongst the most indebted in the world, may deliver rising prices for a while in a Suckers Rally before falling interest rates bump into Max Headroom.

    Related Articles