When Fitch Ratings, Moody’s and Standard & Poor’s get up and apologise unreservedly to everyone around the world for getting it so wrong for so long on rating debt, especially housing-related debt and derivatives, then I will take Fitch’s Australian banks “stress test” release on house price declines seriously.
Until that happens, we can only look askance at Fitch’s stunt and the way it was lapped up by the media yesterday afternoon and this morning.
The justification for the stunt was advanced by Ben McCarthy, managing director for Australia and head of Asia Pacific Structured Finance: “Over the last few months, Fitch has received numerous enquiries as to the sustainability of Australian residential property prices and the possible impacts of a correction.”
Yet none of the reports carried important background information, or even so much as an accurate figure for the rise in Australian home prices. The Fairfax broadsheets this morning announced: “Australia’s capital city home prices have risen 41% since June 2006, on official Australian Bureau of Statistics data. Over the same period, prices plunged in the US, Britain, Ireland and Spain.”
Well, yes the prices in the US, Britain, Ireland and Spain have fallen. All four economies have had or are still in recession, all four overbuilt houses and all four compounded the error by allowing derivatives and securitisation of those mortgages to carry dodgy credit ratings from the likes of Fitch and its mates. And their regulators (even the strong Bank of Spain) got it wrong. Governments in all four countries compounded the errors.
That 41% rise referred to is just in the eight capital cities as measured by the ABS. No account is taken of prices in regional areas, which remains the major omission in the ABS Index. For example, in the early naughties home prices on the Gold Coast jumped 35% in a year.
In fact, the more broader and more accurate RPData Rismark group (latest figures out tomorrow), has pointed out that the market is cooling, because its index measures all of the country, capital cities and regional centres. The ABS index has yet to pick up fully on that cooling.
“As a result of the housing market’s cooling in the June quarter, the year-on-year capital growth rate has fallen from 13.6% in the 12 months to March 2010 to 10.5% in the year to June 2010 (seasonally-adjusted),” the firm said in its report on the June month, quarter and year data.
And then there’s this comment:
“Despite the recent moderation in capital gains, the risk of a dramatic decline in Australian dwelling values remains remote.
“As the RBA has independently confirmed, arguments in favour of house price “bubbles” remain, in my opinion, overstated. Australia’s housing market has a structural shortage of roughly 200,000 homes, which has been substantiated by the National Housing Supply Council. While the inventory of unsold homes has risen of late, our Market Activity Index suggests that new listings activity will slow over the coming months. And although average time on market and vendor discounting have also expanded with the weaker conditions, these remain in line with historically reasonable levels…”
All this was ignored in the reporting. And there was no background as to the mortgage holdings of banks. There was this in one report in The Age: “With an estimated 60 per cent of Australian banks’ loan books secured by residential property, international investors have questioned the sustainability of house prices.” Has that changed over time? What about the level back in the housing slump in the early years of last decade, or the early 1990s crash for comparison?
The breakdown of Australian homes is as follows: around 30% of homes are owned outright, 40% have a mortgage, 30% are rented (and these homes, in many cases, would have been negatively geared and therefore have a mortgage). The level of subprime home loans is now down to around 1% at most. Around 20% of all mortgages were being securitised in 2007 (and most were prime). That has fallen to 10% at best, with government support.
This means the banks have a substantial buffer. And unlike what happened in the US for example, our banks haven’t compounded problems like the already-high leverage in their balance sheets. But they’re still borrowing more, gearing up further and punting on things like collateralised debt obligations (CDOs) and the various other symptoms of derivative hell.
And the reporting of the Fitch move was light on the differences between Australia and those other countries.
As RP Data pointed out, we have unmet demand for homes. Australia also has a strong economy, rising employment and falling unemployment. Incomes are rising at between 2.5% and 4.5% a year. Our home loan providers are all well capitalised and have low levels of bad debts in the sector, unlike banks in the US, UK, Ireland (the banking system is broke and on government life support) and Spain. We have full recourse loans, meaning you go bankrupt if you can’t pay, unlike in the US where people just walk away and challenge their banks to deal with them. We also don’t have tax concessions for owner occupiers like America does, nor do we have strange bodies like Fannie Mae and Freddie Mac which added an implied Federal Government guarantee to many of the securitised loans, allowing bigger and bigger risks to be taken.
Meanwhile, the International Monetary Fund delivered its annual report on the Australian economy overnight, but it’s odd that the only reference made in that report to the housing “bubble” was in the risk section of the final statement: “On the domestic front, a fall in house prices could hit consumer confidence and slow the recovery, or the mining boom may have a larger-than-expected impact on output and inflation.”
So the IMF, after talking to the RBA, APRA and the Federal Government (and others) found no fears or concerns from Australian housing, except as potential risk to “consumer confidence” (Which would be a good thing anyway, if the resources boom is really straining the economy and boosting inflation).
But the experts at Fitch are now so concerned they are going to do a stress test, even though their track record on housing is appalling. So will Fitch publish all the details of its tests, their basis, parameters and all the findings? And will they compare them to the stress tests that APRA uses? There is a major Trans Tasman crisis test coming up…
The IMF said this about APRA (and didn’t mention housing in its assessment of the financial system either): “In particular, we commend APRA’s approach to stress testing as it carefully assesses the vulnerabilities that banks face. The severity of the scenarios provides reassurance about the resilience of the banks to large but plausible shocks.”
And, given the appalling track record of Fitch and its peers, if Australian banks pass the test, will any of the bubbleistas here or offshore believe it?
Glenn
Another good balanced article.
The long-term trend suggests that Australian house prices have been expensive for much of the last ten years. The last time they were below the long-term trend was around 1995. Today, residential property is between 30% and 40% above the long-term average, indicating that buyers are purchasing in an expensive period. In the recent past, the aggressive relaxation of credit allowed end prices to rise. There is limited room for credit to be further eased.
However, and despite the gloomy nature of the bubbleistas (to borrow your phase) the risk of big price declines in the near term seems low. I think – and this is a good result (believe it or not) – that the real return on residential property over the next three to five years (maybe even a decade) is likely to be negative. And as a result, houses will once again be somewhere to live, not vehicles for speculation.
What may deflate the “housing bubble” faster is broad-based job losses. That doesn’t appear likely in the near term at least. In fact, jobs, at present are being created rather than lost. But the more likely and imminent triggers include the tightening of credit – interest rates are predicted to rise, not fall, in coming months – and our middle-aged (and older) loss-making landlords starting to sell. Despite strong price growth over the last ten years, the number of loss-making landlords has risen from 50% to 70%.
It is sad that a more balanced conversation is not taking place re: Aussie house values.
“All this was ignored in the reporting. And there was no background as to the mortgage holdings of banks. There was this in one report in The Age: “With an estimated 60 per cent of Australian banks’ loan books secured by residential property, international investors have questioned the sustainability of house prices.” Has that changed over time? What about the level back in the housing slump in the early years of last decade, or the early 1990s crash for comparison?”
I don’t disagree with the thrust of this piece, and the reports certainly should have provided historical context here. But it is true that residential lending is historically high as a proportion of bank assets. It was above 40 per cent for the first time in August this year, a record. In the early 1990s it was only around 20 per cent. The real shift came in the mid-1990s, though, not more recently. It rose from 20 to 35 per cent between 1992 and 1995, hovered there for the rest of the decade, then fluctuated between 32 and 40 per cent over the 2000s. Between 2005 and 2008 it looked like it might be heading back down, but it’s been rising with dwelling prices since 2008.
Data is from RBA Statistical Table B2: http://www.rba.gov.au/statistics/tables/index.html