[Plonk]

I put this to black_dragon on CC, but it's worthwhile putting it here, too- I'll quote it). Check out the following. It makes no sense:

Quote

The auction results are getting ridiculous. The numbers by REIV and APM for Victoria have absolutely no bearing to each other. REIV says there were 869 auctions and APM says there were 243 auctions. REIV says there was a 73% clearance rate and APM says there was a 67% clearance rate. The disparate number, though, in the auctions, is the key point. How can one report say 869 auctions and the other say 243 auctions- for the same weekend?

REIV: http://reiv.com.au/h...&...26&nav2=162

APM: http://www.homeprice...urne_domain.pdf

Thoughts?

Is one of them- either REIV or APM- going to have to make a revision? The numbers are crazy.

[zaph]

Plonk, on 30 May 2010 - 01:14 AM, said:

I put this to black_dragon on CC, but it's worthwhile putting it here, too- I'll quote it). Check out the following. It makes no sense:



Is one of them- either REIV or APM- going to have to make a revision? The numbers are crazy.

the thing is the revision is too late. the Sunday papers have been printed.

clearance data is dodgy at best. but the question for me is, is it consistently dodgy? if it's roughly consistent then i don't mind much because it's about trends.

i suspect it is anything but consistent.

[hamish]

The truth is out there, but I doubt it's contained in anything that the REIV or APM publishes...

A perfect example of the lack of independant, transparent, and accurate figures that Black Dragon talks about. If these companies had good methodologies, you would expect their figures to be very close given that they are measuring the same thing.

Imagine the stock exchange trying to run on such conflicting data, there'd be outrage, yet the RE industry expects people to make major investment decisions based on this crap.

[Plonk]

Quote

REIV says there were 869 auctions and APM says there were 243 auctions.

I reckon REIV would have to be closer to the mark on this one, though, eh? APM says there were more auctions in Sydney than Melbourne this weekend (@ 259 in Sydney).

http://www.homeprice...dney_domain.pdf

Melbourne always has more auctions than Sydney. Maybe the agents aren't reporting to APMM and APM is only going on the results they're getting. If that were the case, they're only reporting on <1/3 actual auctions.

I've never seen the auction numbers so far apart. I think APM got it wrong.

[staringclown]

When it comes to auction results I wouldn't trust either APM or REIV.

I guess most will have seen these pieces but they're worth posting again.

Dirty Little Secrets - mediawatch

"Bad" real estate agent stories pulled after complaints from the REIV.

REIV caught cheating again - Neil Jenman

Quote

1. Ignore failed sales. When properties are passed-in at auction (meaning they are not sold), these properties will often not be included in the statistics. By leaving out the 'failed' sales, the official 'sold' sales percentage appears higher.

2. Include after-auction sales. When properties are passed-in at auction, they are often sold at a later time. But, no matter when those sales take place, they are often reported as 'auction sales'. Of course, the only sales that can truthfully be quoted as a 'sale at auction' are those sales that actually occur at the auction - that's when the auctioneer says 'Sold'. But truthfulness is seldom part of the auction reporting system.

3. Include before-auction sales. If properties are sold before an auction it means they were not sold at the auction. But, again, these properties are often reported as auction sales.

4. False highest-bid prices. Often, when a property is passed-in, the price at which it was passed-in is also reported. For example, the result in the newspaper may say, "77 Smith Street - P/I at $600,000".

But, often, the passed-in price was a vendor (dummy) bid. The true highest price may have been $500,000. But the property is reported as passed-in at a higher figure which deceives the buyers into believing the property is worth the amount of the highest bid.

5. Satisfaction ignored. The real estate industry measures "success" as a "sale". That's basically the sole yardstick. Never mind that the poor sellers may have been "crunched" and bullied by the high pressure auction process into accepting an amount far below what they were first quoted by the agent; if the property is sold, that's considered a success. If the auction system was measured by client happiness, it's percentage level would be lucky to make double figures (10%).

6. Undersold. One thing that the auction results never tell you - and that's the price that the successful bidder was truly willing to pay.

Agents know (or should know; in fairness, many aren't too bright) that the highest bidders at auctions seldom bid to their highest prices.

For example, if a bidder intends to pay up to $600,000 for a property and if the reserve price on that property is $500,000 and there is only one other bidder (whose highest price is $550,000) then the bidder who is prepared to pay $600,000 will buy the property for tens of thousands of dollars below what he was prepared to pay. He will probably pay about $560,000, at best which means he saves $40,000 and seller has undersold by $40.000. In auctions, this happens all the time.

The auction system is riddled with deceit and false information. Normally, by the time the sellers discover the horrible truth about auctions, it's too late. They have been duped.

And the duping starts, of course, with the fudged figures reported in the newspapers.

For example, last weekend, according to the on-line news site, crikey.com.au, "Of the 332 Melbourne properties listed for auction on Saturday 6 December, 144 were sold - a clearance rate of less than 44%. A stark contrast to the reported clearance rate of 55%".

A raft of articles about REIV dodginess from Peter Mericka - A victorian conveyancing lawyer. He really doesn't like Enzo.

[tom]

If a house sells pre auction or even immediately after an auction by negotiation with teh highest bidder as many do I think it speaks to the strength of the property market.

I cannot see why these should not be included? Makes more sense to include them then exclude them to me? If all the houses sold pre auction you would hardly say it is a laggard of a market would you?

As zaph says above basically as long as they are consistantly dodgy we know we have had 40% clearance rates in Melbourne / Sydney before so we can get them again. We know on their "dodgy" figures at below 50% clearance rates in Melb and Syd prices start to falter. Personally I dont care how they are derived, it is better that the methodology does not change now or if it was to change the old and new would have to be calibrated so the new rates could be reflected on in reference to our expected price outcomes at various clearance rates.

[gibber_blot]

zaph, on 30 May 2010 - 01:24 AM, said:

clearance data is dodgy at best. but the question for me is, is it consistently dodgy? if it's roughly consistent then i don't mind much because it's about trends.

It depends. If we give the REIV the benefit of the doubt and assume that use a completely objective metric for calculating the clearance rates (which I don't think they do), it can still distort trends. Consider, for example, if they always remove auctions that were cancelled due to lack of interest; i.e. no registered bidders, so the auction is cancelled. This can be applied objectively, however, if in one month, 10% of all auctions are cancelled, and then the next, 50% are cancelled, but the clearance rate remains about the same, I would say that the market is tanking, despite what the trend says.

Any worthwhile clearance rate model should include cancelled auctions, because attracting people to your marketplace is absolutely part of the process. The formula should be the number of scheduled auctions divided by the number of sales, including pre-sales and sales made within a fixed time limit (e.g. the same weekend).

Unfortunately, many bodies like the REIV use data reported by real estate agents. If some agents are not reporting cancelled auctions, a single agent's motivation to do so is reduced, because their clearance rates will look worse than the rest.

[tom]

gibber_blot, on 31 May 2010 - 02:44 AM, said:

It depends. If we give the REIV the benefit of the doubt and assume that use a completely objective metric for calculating the clearance rates (which I don't think they do), it can still distort trends. Consider, for example, if they always remove auctions that were cancelled due to lack of interest; i.e. no registered bidders, so the auction is cancelled. This can be applied objectively, however, if in one month, 10% of all auctions are cancelled, and then the next, 50% are cancelled, but the clearance rate remains about the same, I would say that the market is tanking, despite what the trend says.

Any worthwhile clearance rate model should include cancelled auctions, because attracting people to your marketplace is absolutely part of the process. The formula should be the number of scheduled auctions divided by the number of sales, including pre-sales and sales made within a fixed time limit (e.g. the same weekend).

Unfortunately, many bodies like the REIV use data reported by real estate agents. If some agents are not reporting cancelled auctions, a single agent's motivation to do so is reduced, because their clearance rates will look worse than the rest.

I agree.

It would likely be more accurate if you randomly picked say 50 planned auctions at random and followed through on the 50 auctions to get an accurate picture on 50 rather than a wrong picture on what is meant to be the entire market when it is not but a selected list as opposed to a random smaller list

[staringclown]

staringclown, on 30 May 2010 - 03:55 AM, said:

1. Ignore failed sales. When properties are passed-in at auction (meaning they are not sold), these properties will often not be included in the statistics. By leaving out the 'failed' sales, the official 'sold' sales percentage appears higher.

I'm with gibber here, I just don't know how dodgy this guy can be. These guys aren't publishing an index. They push auctions as the mode of selling apart from sheer numbers and that's where the exaggeration really bites. The failed reporting of sales to boost the results is really what annoys me. I want transparency. I don't think the guys who have a vested interest should be publishing the figures.
Edit: spelling

This post has been edited by staringclown: 31 May 2010 - 01:08 PM

[booboo]

staringclown, on 31 May 2010 - 01:07 PM, said:

I'm with gibber here, I just don't know how dodgy this guy can be. These guys aren't publishing an index. They push auctions as the mode of selling apart from sheer numbers and that's where the exaggeration really bites. The failed reporting of sales to boost the results is really what annoys me. I want transparency. I don't think the guys who have a vested interest should be publishing the figures.
Edit: spelling

Yes, agree. I would also love to see the total number of auctions for the weekend and reported number to base conclusions on - for example, they could release two figures: reported auction clearance rate, plus reported % (reported * 100 / total).

When lurking on CC, I noticed some bulls argued that that even though the clearance rate is fudged, you can still draw conclusions as it will have been continually fudged. Unfortunately, I doubt that's the case as it requires the level of "fudging", or misreporting, to be a constant. For example:

- in the "good times", an REA has 12 auctions and reports 10 (say, 8 sold reported, 2 unsold reported, plus 2 unsold unreported) - for a reported clearance rate of 80% and real clearance rate 66%;
- in the "bad times", an REA has 12 auctions and reports 5 (say, 4 reported sold, 1 reported unsold, plus 7 unsold unreported) - for a reported clearance rate of 80% and real clearance rate of 41%.

In the above (heavily contrived) example the clearance rates may even look identical (80%). Yes, it may be a contrived scenario, but shows that fudged clearance rates are, unfortunately, meaningless.

I understand that past clearance rates are often revised down in the future, but the revised clearance rates rarely seem to be mentioned. Is it possible to see past clearance rates after the auditor general's office release data? Or is this not done?

[Plonk]

If APM are reporting on < 1/3 auctions for last weekend, then I would think their report becomes less than useful. Isn't that when they should be writing "SNR" (statistically non-reliable)?

[tom]

Plonk, on 01 June 2010 - 12:55 AM, said:

If APM are reporting on < 1/3 auctions for last weekend, then I would think their report becomes less than useful. Isn't that when they should be writing "SNR" (statistically non-reliable)?

Random sampling is potentially more accurate then attempting to cover the entire field unsuccessfully.

If you picked say 50 proposed auctions at random and then rigorously followed through whether they were sold or not and when this is likely to be better than selecting auctions based on results. Clearly a random sample is better than a non random attempt at the whole population.

I actually agree with the bulls though about all this, these auction figures have been 40% before for Melbourne so I assume they can be again. When the clearance rate is 40% we know prices start dropping.

They only get one shot at dodgying them up because if we start getting price falls at auction clearance rates of say 60% then this becomes the new benchmark for falling prices. It really serves little purpose for them except as a once off. For RE agents it is a different storey of course if they can demonstrate they are better than the rest of the population.

[zaph]

i think one of the problems with APM (and perhaps the rei's) is the speed at which they report. if i was a rea the last thing i'd want to do after working all saturday is start reporting, i'd be off to the pub or home to my family and leave the 'paperwork' for monday.

[Plonk]

tom, on 01 June 2010 - 01:17 AM, said:

I actually agree with the bulls though about all this, these auction figures have been 40% before for Melbourne so I assume they can be again. When the clearance rate is 40% we know prices start dropping.

tom, I don't know about <50% being the indicator. I read somewhere that falling auction levels have a real correlation to future house prices. Apparently, auction completions are some kind of leading indicator. What I have noticed is that bulls have lowered the threshold, as figures have fallen. Now, it's "oh, well, its not below 50%." I just wonder what 50% means. Victoria seems to have auctions as a preferred methof, Sydney less so, other states even less so. Perhaps trend means more than nominal, in the case of auctions.

Re APM, I'd be more comfortable if they had a footnote in their results, saying that the results are not comprehensive, and are based on less than 1/3 of actual Melbourne auctions. At least Enzo has the greater numbers reported.

[Plonk]

zaph, on 01 June 2010 - 01:19 AM, said:

i think one of the problems with APM (and perhaps the rei's) is the speed at which they report. if i was a rea the last thing i'd want to do after working all saturday is start reporting, i'd be off to the pub or home to my family and leave the 'paperwork' for monday.

Lol! And you would be part of the problem, zaph! Alll they have to do is ring Enzo, have a chat tell him the news and be home for beers later.

I wish ABS would collect data on auctions.

[Sherlock]

booboo, on 01 June 2010 - 12:36 AM, said:

- in the "good times", an REA has 12 auctions and reports 10 (say, 8 sold reported, 2 unsold reported, plus 2 unsold unreported) - for a reported clearance rate of 80% and real clearance rate 66%;
- in the "bad times", an REA has 12 auctions and reports 5 (say, 4 reported sold, 1 reported unsold, plus 7 unsold unreported) - for a reported clearance rate of 80% and real clearance rate of 41%.

Brisbane, and Adelaide, agents must not know, about this cunning plan -- they've been reporting 30-40 percent clearance rates, for many months!! Likewise Sydney, and Melbourne, in 2008, when clearance rates tumbled. In short -- there is no evidence, of agents falsely inflating figures, during "bad times", because these figures do drop, very low, during bad times................

[Plonk]

Sherlock, on 01 June 2010 - 02:24 AM, said:

Brisbane, and Adelaide, agents must not know, about this cunning plan -- they've been reporting 30-40 percent clearance rates, for many months!!

Those places barely have auctions. Auctions are Victorian, and to a lesser extent Sydney.

Quote

Likewise Sydney, and Melbourne, in 2008, when clearance rates tumbled. In short -- there is no evidence, of agents falsely inflating figures, during "bad times", because these figures do drop, very low, during bad times................

sherlock, what do you think about this topic- and APM reporting on 1/3 of the auctions that REIV reported on?

[booboo]

Sherlock, on 01 June 2010 - 02:24 AM, said:

Brisbane, and Adelaide, agents must not know, about this cunning plan -- they've been reporting 30-40 percent clearance rates, for many months!! Likewise Sydney, and Melbourne, in 2008, when clearance rates tumbled. In short -- there is no evidence, of agents falsely inflating figures, during "bad times", because these figures do drop, very low, during bad times................

It was black_dragon and Neil Jenman who have complained about REAs under reporting failed auction results. No offence, but I will believe them over your claim of "no evidence", and both of those guys are in the industry, have worked with REAs and understand what goes on. Unless you are an REA in disguise and wish to prove as much, I will take their word over yours. The door is open, however, to show your credentials. I freely admit I have none apart from being an interested watcher and reading what I can.

As much as I'd love to say that the auction clearance rates in Brisbane are indicative of much, they have the same problem Sydney had during the GFC: the sample size was/is too low. And as Plonk mentioned, Brisbane and Adelaide clearance rates are traditionally lower as auctions are not the preferred method to sell in those cities - different markets / expectations from buyers I guess (and hence the lower sample size). In comparison, auctions in Sydney and Melbourne are more the norm and have significant sample size to draw conclusions from - or they would, if more than 1/3 of the auctions were reported.

For me personally, when I have auctions I am interested in it is a source of frustration to see them not reported by APM (mainly as I like to monitor prices in the areas I am interested in, or to see how places I am interested perform). I will freely admit that there are a significant number of auctions unreported that do sell (eg I see removed from Domain/RE.com.au after the auctions), although from my anecdotal evidence of watching for auctions more of the unreported auctions tend to have not sold (eg they are not removed from the relevant website).

As a final digression, it was with much mirth that I saw that Enzo blamed high auction numbers for lowering clearance rates. If demand really was as high as claimed at the current high prices and supply so low, then either 1) it would be unlikely to see such "high" numbers of auctions, or 2) the clearance rates would not fall (if the auctions replaced, say, normal listings).

This post has been edited by booboo: 01 June 2010 - 03:50 AM

[zaph]

booboo, on 01 June 2010 - 03:49 AM, said:

If demand really was as high as claimed at the current high prices and supply so low, then either 1) it would be unlikely to see such "high" numbers of auctions, or 2) the clearance rates would not fall (if the auctions replaced, say, normal listings).

i guess the clearance data has to be read in conjunction with sales price. it's possible to have a falling market with a high clearance rate, although as you said there would probably be less auctions and it would change.

[Plonk]

Article by Tim Lawless about the auction results.

http://www.smartcomp...ance-rates.html

RPData say they collect 80% of auction results. They found clearance rates this weeked gone to be:

Melbourne: 69.4%
Sydney: 63%

This graph is interesting- it shows which capitals do auctions- it's obviously just a culture that's developed over time: