Peter 14.7
04-09-2004, 04:55 PM
Lies, damn lies
By Alan Kohler
September 4, 2004[/B]
This artcle outlines the number of different approaches taken to measuring the property market movements. It goes a good way to explaining the soofs and making an overall asessment which is more gloom to come.
For those who wish to skim, I will bold the bits of note.
Peter 147
Yesterday's headlines were pretty clear about it: "House prices decline sharply" and "Bubble deflating as house prices fall for the first time in four years."
Those stories were based on the Australian Bureau Statistics house price indices for the June quarter. Unfortunately, for the mental stability of those who watch these things, two days earlier the Real Estate Institute of Australia reported something entirely different.
In places where the ABS says prices went up in the June quarter, the REIA said they went down. And where the ABS said they had fallen, the REIA said they had gone up.
So, on Thursday, the ABS reported that house prices in Sydney had slumped 5.4 per cent in the June quarter; on Tuesday, the REIA had reported that Sydney prices were steady. The ABS said Melbourne prices had gone up 1 per cent; the REIA said they were down 2.7 per cent. The ABS said Perth prices had gone up 2.1 per cent; the REIA said they were down 1 per cent.
The funniest thing about all this nonsense is that the ABS and the REIA get their data from exactly the same source - state governments. Why are they different? That's a bit hard to fathom, although the ABS figures are an index, whatever that means, and the REIA figures are median prices. Maybe that's it. Yes, say the researchers and statisticians involved, that must be it.
So the two questions still to be answered about the Australian property market are: is the collapse on? And when are all these statisticians going to get their act together?
Answers: yes and soon (they've even had a meeting to talk about getting their act together).
I think property prices are definitely falling because the research house with the best data - Australian Property Monitors - says so. APM is the only outfit that sends staff to auctions with notepads to take down the prices, and it's the only one that uses contract of sale dates rather settlement dates, which are an average of 60 days later.
Most of the sales included by both the ABS and REIA for the June quarter actually took place in the March quarter. APM includes sales only in the period when the buyer and seller actually shook hands. That, as everyone knows, is when the real deal took place; the settlement date is irrelevant.
On APM's data, so far unreported, Sydney values fell 7.8 per cent in the June quarter, Melbourne fell 2.6 per cent, Perth fell 0.7 per cent, Canberra fell 1.2 per cent, Adelaide rose 3.6 per cent and Brisbane rose 1 per cent.
Most economists reacted to the ABS data by opining that the property market was in an "orderly decline", but a fall of nearly 8 per cent per quarter, if sustained, is anything but orderly. It is an absolute rout.
APM also collects other data worth knowing about. Auction clearance rates around the country are now 50 per cent or less, compared with about 70 per cent a year ago; the average number of days it takes to sell a house is 69 days in Sydney, up from 49 days last year and 36 days in 2002; the average discount off the advertised price that it takes to sell a house in Sydney and Melbourne has gone up to nearly 8 per cent, compared to the long-term average of about 4.8 per cent.
A month ago, in response to a series of articles by your columnist on this subject, the research staff from the RBA, Treasury, the ABS, APM, Residex and all of the state real estate institutes met in Canberra to discuss what to do.
They all agreed that using settlement data was not a good idea and that the ABS and the REIA and the state institutes would work towards basing their figures on contract dates. Separately, the ABS is working on some statistical adjustments to the data to move away from median prices and adjust house prices for the size and type of houses, their location and, importantly, renovations.
So tick two issues as being tackled at least: out of date data and the use of median prices.
Two more issues remain to be tackled: the fact that we're using quarterly figures and the fact that buyers don't have free, useful, independent data on sales in the areas in which they are looking for houses.
Everyone I spoke to yesterday moaned about the difficulty of getting real estate agents to report their sales more frequently to a public database.
They should just be told to do it - on the day the sale is made. The problem is not that they can't - they don't want to. Knowledge is power or, in this case, money.
By Alan Kohler
September 4, 2004[/B]
This artcle outlines the number of different approaches taken to measuring the property market movements. It goes a good way to explaining the soofs and making an overall asessment which is more gloom to come.
For those who wish to skim, I will bold the bits of note.
Peter 147
Yesterday's headlines were pretty clear about it: "House prices decline sharply" and "Bubble deflating as house prices fall for the first time in four years."
Those stories were based on the Australian Bureau Statistics house price indices for the June quarter. Unfortunately, for the mental stability of those who watch these things, two days earlier the Real Estate Institute of Australia reported something entirely different.
In places where the ABS says prices went up in the June quarter, the REIA said they went down. And where the ABS said they had fallen, the REIA said they had gone up.
So, on Thursday, the ABS reported that house prices in Sydney had slumped 5.4 per cent in the June quarter; on Tuesday, the REIA had reported that Sydney prices were steady. The ABS said Melbourne prices had gone up 1 per cent; the REIA said they were down 2.7 per cent. The ABS said Perth prices had gone up 2.1 per cent; the REIA said they were down 1 per cent.
The funniest thing about all this nonsense is that the ABS and the REIA get their data from exactly the same source - state governments. Why are they different? That's a bit hard to fathom, although the ABS figures are an index, whatever that means, and the REIA figures are median prices. Maybe that's it. Yes, say the researchers and statisticians involved, that must be it.
So the two questions still to be answered about the Australian property market are: is the collapse on? And when are all these statisticians going to get their act together?
Answers: yes and soon (they've even had a meeting to talk about getting their act together).
I think property prices are definitely falling because the research house with the best data - Australian Property Monitors - says so. APM is the only outfit that sends staff to auctions with notepads to take down the prices, and it's the only one that uses contract of sale dates rather settlement dates, which are an average of 60 days later.
Most of the sales included by both the ABS and REIA for the June quarter actually took place in the March quarter. APM includes sales only in the period when the buyer and seller actually shook hands. That, as everyone knows, is when the real deal took place; the settlement date is irrelevant.
On APM's data, so far unreported, Sydney values fell 7.8 per cent in the June quarter, Melbourne fell 2.6 per cent, Perth fell 0.7 per cent, Canberra fell 1.2 per cent, Adelaide rose 3.6 per cent and Brisbane rose 1 per cent.
Most economists reacted to the ABS data by opining that the property market was in an "orderly decline", but a fall of nearly 8 per cent per quarter, if sustained, is anything but orderly. It is an absolute rout.
APM also collects other data worth knowing about. Auction clearance rates around the country are now 50 per cent or less, compared with about 70 per cent a year ago; the average number of days it takes to sell a house is 69 days in Sydney, up from 49 days last year and 36 days in 2002; the average discount off the advertised price that it takes to sell a house in Sydney and Melbourne has gone up to nearly 8 per cent, compared to the long-term average of about 4.8 per cent.
A month ago, in response to a series of articles by your columnist on this subject, the research staff from the RBA, Treasury, the ABS, APM, Residex and all of the state real estate institutes met in Canberra to discuss what to do.
They all agreed that using settlement data was not a good idea and that the ABS and the REIA and the state institutes would work towards basing their figures on contract dates. Separately, the ABS is working on some statistical adjustments to the data to move away from median prices and adjust house prices for the size and type of houses, their location and, importantly, renovations.
So tick two issues as being tackled at least: out of date data and the use of median prices.
Two more issues remain to be tackled: the fact that we're using quarterly figures and the fact that buyers don't have free, useful, independent data on sales in the areas in which they are looking for houses.
Everyone I spoke to yesterday moaned about the difficulty of getting real estate agents to report their sales more frequently to a public database.
They should just be told to do it - on the day the sale is made. The problem is not that they can't - they don't want to. Knowledge is power or, in this case, money.