Qld property market crashes !! ;)

[Les]

G'day all,

Just received my latest API mag and got a bit of a shock when checking out one of my "favourite suburbs". The capital growth showed as 4.83% for the year - where I know that actual growth is nearer 20%. (check Wellington Point)

So, I got looking at a few other suburbs (comparing last year's API mag figures with this one) - and was amazed to see some "doozies" - (please note, I don't know if API's data is presented raw from Residex, or whether it is entered by API themselves - I have alerted API to the problem....)

Suburb     Postcode   31 Dec 01   31 Dec 02   their %   my%

Windsor      4030       247000     321000      8.8%     30%
Red Hill     4059       283000     343000      4.8%     21%
Waterford W  4133       104000     138000      10.5%    32%
Helensvale   4210       175000     190000      30.6%    8.6%

And, for those that think this might be confined to Qld:-

Wool'mooloo   2011       2872000     1228000     8.77%   -58%
Concord W     2138       418000      553000      20.3%    32%

I have NO idea if it is Residex who has dropped the ball here - or API - suffice to say, DON'T take what you read as gospel. And, this is the FIRST time I have seen such huge deviances of growth figures, so I have not checked previous years !!!!

The figures I quote above are all ex API mags - all provided by Residex, relating to 31 Dec 01 and 31 Dec 02 FWIW

If a newspaper were to get hold of these figures, it could make for one helluva headline on Monday !!!

Queensland Property Market crashes
"After huge increases to Sep 02, Brisbane property values have shown only marginal increase for the full year to 31 Dec 02"

Regards,

 

[rambada]

My understanding is that the API data is Residex data and therefore filtered. What is the source of your data? I agree with your figure for Helensvale as I live on the Gold Coast - but I can only speak anecdotely and don't have access to raw or filtered data. If you aware of an alternative or better data can you please supply?

Regards

Rambada

' Data is not information, information is not knowledge, knowledge is not wisdom'

 

[bbruham]

G'day Les,

I listen to two property real estate guessing shows on radio every
Saturday. One hour on 2GB (RPD Research), 9am to 10am, them I switch to 2UE and do another hour listening to Residex.10am to 11am.
On 2GB one woman had a valuation of $350,000 for her unit.
Her reply was, why then are all offers no higher than $289,000?
RPD's answer was units are begining to SOFTEN in price.
These lot of dopes are only guessing!!!
I've done business with Residex buying current valuations and future valuations. The future vals. are just crystal ball gazing.
So when these people print out future stats, I don't believe them.
Purchase their current valuations, do your own research and if everything on a property, that fits your strategy, then call in an independent property valuer. It costs $250 a valuation.
My current API magazine is still unopen.

bbruham.

 

[Les]

G'day Rambada,

What is the source of your data?

As mentioned, this was ex API magazine - prices quoted were ex API mag April/May 2003 and April/May 2002. Whether these were verbatim from Residex, or not, I've yet to know - BUT, the %ages quoted were incorrect wrt to the figures quoted.

There is NO WAY that a median property can grow from $247k to $331k in one year and still only grow 8.8% in value - the two are mutually exclusive !!!! So watch out !!! Don't take anything at face value.

G'day Bbruham,

So when these people print out future stats, I don't believe them.

Good call - but my point was even PAST statistics can't be trusted if taken at face value (see above)

Regards,

 

[Kevin Hockey]

Hi Les,

This is no surprise to me. In all my real estate career I have seen all media completely misjudge current market trends.

To begin with they are a quarter, at least, behind the current market. I read some articles and burst out laughing during our recent boom. My recent thread regarding the waterfront property for sale was a perfect example. The growth on the waterfront has been very high, yet stats for the area (based on postcode) showed only minimal growth. By using postcode it included Bracken Ridge & Deagon which are km's away and not in the same ballpark as Sandgate waterfront. There was no comparison.

Wavell Hts has seen massive growth in Brisbane. API shows it as a red zone for Brissie growth, but not all of wavell is the same. Some streets have city views and others are much less desirable. I don't believe you can determine growth by the suburb, you can have 2 very different areas within a suburb.

I almost completely disregard the media with regard to property. I invest in areas I know and don't go outside of there unless I have a good reason. I might miss a good buy now and then but I never get caught out either.

Kev

www.nundahrealestate.com.au

 

[Les]

G'day Kev,

What you are referring to is the "areas within areas" syndrome - and I don't see that being put right without stepping outside of postcodes and into suburbs. Then, again, there are streets within suburbs........

My point (which was not presented well because of the vagaries of websites - but now fixed, thanks to Mike ) was that a capital growth from $247k to $321k in 12 months is quoted as an 8.8% growth !!!!!!

My "alarm bell" was Wellington Point (pc 4160) which has grown about 20% (even if some valuers have their heads in the sand, eh Asy ) - it showed median of $207k in Dec 01 and median of $245k in Dec 02 - and that was represented in API by a growth of 4.83%!!!!!!! Someone needs a new calculator !!!

Getting deeper, postcode 4160 includes Ormiston as well - but that's not the point. The point is simply that the growth % does NOT MATCH the stated $ growth between 01 and 02 - and that is across the board (not just Wgtn Pt) and this is all from the same source !!!! And, yes, I also read the API article that mentioned median $$ values, and how they could be different - but that's not the point either..... How can growth of $74k on a start base of $247k equate to 8.8% growth?????? Answer - it is just plain WRONG!!!!!!

For those newer souls, I repeat - don't just accept what is printed - in the the words of "Porgy and Bess" it ain't necessarily so !!!

Regards,

 

[XBenX]

is that -58% for wool'mloo ?

 

[Les]

G'day XBenX,

Yeah, mate, I reckon it is -58%. Here is a reprint of the API mags figures (hopefully a little better presented than in the original post)....

Suburb:- Woolloomooloo
Postcode: 2011
Median as at 31Dec01:- $2,872,000
Median as at 31Dec02:- $1,228,000
Their growth %:- 8.77%
I reckon:- -58%


By the way, if anyone knows how to present data such that it produces columns on the page (as first attempted), I'd be interested to know HOW ......[Use the PHP function and adjust spacing until columns line up - Mike]

PS - I am making a large assumption here, and that is that it is their growth % figure that is wrong. Could it be their median figures that are wrong, while growth% is CORRECT?? (In the case of Wgtn Point, I'd say "No!" - but W'mooloo? ....????)
Regards,

 

[see_change]

Les

Is it possible that they are taking their % increase on resales of the same properties, rather than increase of average/ median price.

Certainly in Wolloomooloo , there are some very expensive properties on the market which didn't exist before and these would have shifted the average, but don't represent an increase in price on existing houses.

Even then , when they taking into account resale of existing houses, that house may have been extensively renovated.

see change

 

[Aceyducey]

Why not tell them about the error.

I've been pointing out errors in Allhomes for awhile (mostly mis-typed sales prices, but can really alter those medians).

I will leverage my rep to get some extra info out of them when I really need it.

Cheers,

Aceyducey

 

[sbe]

I thought API had an article in the latest edition expaining all that.....

They use a repeat sales index, which tracks the same property's value.

See P18 current issue.

Cheers,

Simon.

 

[JohnE]

Hi All, Simon is correct. But to help a shortened version of some of the info.

The Capital growth rates are not calculated using the change in the median value for a suburb over the last 12 months. The median value is not useful for calculating growth rates, it simply gives an indication of the middle price for the market. The median can be skewed by a number of factors, including the number of sales over the given period, changed buying patterns
and releases of prestigious real estate, for example. The skewness of the median is measured by the Typicalness Rating (TR) given in the end column in the tables in the magazine.

Instead, Residex capital growth rates are calculated by finding the growth rates for individual properties. This means we compare the previous sale price with the latest sale price for each property, and then the growth rate is averaged for the suburb. Finding growth rates on an individual property basis eliminates the skewness in medians discussed above.

Hope the above helps. One other thing: yes, predicting the future is always going to be a guessing game but providing you do some detailed analysis you can gain an understanding of the most likely outcome and this is what Residex does. Within the next few weeks we are going to release some free software that will allow you to model the future and hopefully get a better understanding of the future. If you wish you will also be able to purchase a set of indicies that will allow you to answer the question: “How likely is the future I’m modelling? What are my chances of getting the return I'm expecting?” Anyway, go to residex.com.au and register for our free news letter and we will then be able to notify you of it becoming available.

Regards, John E CEO Residex

 

[XBenX]

Thanks John for the clarification, welcome to the board it I hope you have sometime to contribute.

Your original show on 2GB ??? was one of my catalysts for starting property investment.

For those of you who dont know Residex and what they do check them out.... I think Ive used all of their products at sometime in the past. The quarterly report is a good exec summary of the market as well as lots of associated data that is useful for your own projections etc.

Without API in front of me… Id hate to be the person/persons who made W'mloo have -58% growth ! that is some serious $$$ down the drain

Everyone should subscribe to the newsletter - there is even a mental challenge each month

 

[Les]

G'day JohnE,

Thanks for weighing into this one - I can really do with some help. First off, you mentioned "typicalness rating" - and it sounds like the lower the number, the less chance that median changes are "actual" - is that right?

I've been in touch with API, and am told that they faithfully reproduce what Residex provides. So, can you confirm that the figures I posted above are as provided by Residex?

What I'm now trying to come to grips with, is :- Which figure is likely to be the most reasonable to use? The growth %, or the median change? From your comments, and on reading the API article (p18 onward), it seems the growth % "should" be more accurate.




BUT (didn't you know there was gunna be one of them )

HOW can Wellington Point gain (over a twelve month period) at the rate of 17.2% (to Mar 02), 27.3% (to Jun 02), 23.7% (to Sep 02) and 4.8% (to Dec 02).

At the same time, median prices showed growth at 16.9% (to Mar 02), 21.9% (to Jun 02), 25.5% (to Sep 02), then 18.4% (to Dec 02)

I see a certain similarity between the growth and median increase figures - until this Dec quarter when they diverge wildly.


So, to repeat, what is the more realistic figure to use to get "an idea" of suburb growth - the growth %, or the median change?

I am aware of the vagaries of "medians" - what I'm not yet aware of are the vagaries of your posted "growth %" figure. Can it be unrealistic? If so, how so? What changes would lead to an unreasonable growth% figure?


I'd noticed Wgtn Point "slowing down" - but from 23.7% to 4.8%? That's a helluva quick downturn.......

I'm interested in gaining some more knowledge re this "growth" figure, John, if you'd be so kind ........

I'm finding it hard to understand how "sales of the same property" can be more accurate than medians (considering forced sales, emotive buyers, etc.)

Regards,

PS By the way, I just heard back today from API mag - they affirmed that they faithfully reproduce the figures you provide - and the respondent also spent a bit of time explaining in a way similar to yourself just how these "median vs growth" figures can be "out of whack". For myself, I can appreciate how DIFFERENT firms can produce different results - but my figures are ONLY comparing API/Residex figures with API/Residex figures (a year before).......

Regards,

 

[XBenX]

Les,

basically its because medians arent a very good measure of what is happening in the property market - to explain

Let say we look at Somersoft postcode 2003

in month 1 there is a new land release in somersoft, the only thing selling is this land it sells for 100K 101K 99K so the median is 100K

in month2 the stock of land is gone and some different extablished houses come onto the market the sell for 200K 190K and 210K the median is 200K

so the median has gone from 100K to 200K in a month WOW thats some change.... see why medians suck

so to get around this residex only uses resales of the same property to calculate growth rates..... a much more accurate measure.... but still it doesnt take into account the $$$ spent of the property (reno/rebuild/additions/etc) so it probably overstates the growth slightly (unless they adjust for this ?)

its a good method - if you call them up they also try to take out the $$$ spent problem by asking if the house has changed since it was last traded... oh it also assumes market value was paid in the first place.

that was clear as mud right ?

 

[JohnE]

Hi, Basically you should not read too much into medians. The only time they are safe is when the properties being sold are representative of the total market. It is difficult to know when this is the case. The method we use is Internationally recognised as the most definitive measure. We developed it about 12 years ago and it has been recognised by many governments as the best. In Australia the Reserve Bank uses our numbers following some extensive research into the best measure. The Median is easy to calculate so that is why most do. We would love to see the market stop using it as a measure.

Regards, John E

 

[Kevmeister]

Perhaps I have misunderstood this. John says the growth is based on the growth in sale price of the same property.

XBenX gives a good example why basing growth on medians sucks. Different house sales produce different medians month-to-month. Fair enough.

Since these numbers are published quarterly and, as far as I'm aware most people don't buy and sell a house in the same quarter, what gives?

 

[Rickardo]

I would assume that they would only use data of all the houses that were sold in that quarter. These would most likely not be the same houses sold in the previous quarter since a house doesn't generally get sold that often. I still believe that this is a way better indicator than the median, i dont really understand why this is used to often in the first place.

 

[Les]

G'day XBenX,

And thanks for adding extra value -

so to get around this residex only uses resales of the same property to calculate growth rates..... a much more accurate measure.... but still it doesnt take into account the $$$ spent of the property (reno/rebuild/additions/etc) so it probably overstates the growth slightly (unless they adjust for this ?)

Ok, HOW MUCH MORE accurate is reporting growth ONLY on resales? Couldn't the same "vagaries" affect these figures? When they work this out ONLY on resales, are they ONLY resales made within the last 12 months (i.e. a property is bought AND sold within the previous 12 month period .....) that would be pretty limiting, I would've thought....

Or, do they source EVERY house sold within the last 12 months, find WHEN it last sold, and then calculate "growth" as an even growth over a number of years?

As I said, I'm aware of the vagaries of medians - what I'm NOT aware of is any vagaries that exist with the "growth %" figure.

e.g. let's say someone buys a house in 1996 in wgtn Pt for $120k - it was an emotional buy (it's gunna be "the first family home" and they just WANT IT !!!! They could've bought a similar quality home for $100k.

Anyway, time moves on, hubby and wifey have some probs, can't resolve them, and they split - and the house is sold to "divvy up" the gains..... The year is now 2003, and (because it's one of the 3 D's) it fetches $170k. If they had been a little less "pressed, it could have easily fetched $200k.

SO, over 7 years, this property has advanced in value by $50k over 7 years (on an original base of $120k) - so the "average growth per year" has been 5.9% - is THIS the kind of figure that makes up "growth %"??? If the buy and later resell was more "normal", then growth would've been nearer 14%.......


Spill the beans, XBenX - just HOW does this growth figure get calculated ??????? The current trend seems to be to say it is "more accurate" than medians - fair enough, but HOW MUCH more accurate?

Can you help?

Regards,

 

[Bill.L]

Hi All

Lies, dammed lies and statistics.
Les you are embarrasing the professional statisticians by proving that you can't compare an apple to an orange to a.. er..lemon
I am a person who loves numbers and statistics but realises how limiting they can be. Property investment is more than just numbers, however massaged. It is an art in knowing an area and having your finger on the pulse of what is happening.
Les, you know your areas, and would think that the statistics should show this, from this thread the truth emerges.

bye