Spin

Freddio

According to the government via RTE

http://m.rte.ie/news/touch/business/2014/0625/626388-residential-property-prices/

However according to Daftdrop.com we are exactly where we were a year ago and more prices continue to fall than rise.

If they are going to do Seany for propping up the share price, surely they can do the same to Enda and his banks.

barneystinson

Freddio wrote: »

According to the government via RTE

http://m.rte.ie/news/touch/business/2014/0625/626388-residential-property-prices/

However according to Daftdrop.com we are exactly where we were a year ago and more prices continue to fall than rise.

If they are going to do Seany for propping up the share price, surely they can do the same to Enda and his banks.

Ummmm, what?

Are you going to show us your in depth analysis of the CSO versus Daftdrop figures to support your assertions?

I'd have thought CSO were operating out of a bigger, better dataset, but since you haven't bothered your arse to find out before you started firing accusations, and I'm not inclined to do it for you, we'll have to wait for a grown-up to arrive...

andrew

Freddio wrote: »

According to the government via RTE

http://m.rte.ie/news/touch/business/2014/0625/626388-residential-property-prices/

However according to Daftdrop.com we are exactly where we were a year ago and more prices continue to fall than rise.

If they are going to do Seany for propping up the share price, surely they can do the same to Enda and his banks.

If you look at the 'average house price by month' graph on that site you'll see that it also shows property prices rising, so I don't see where you're coming from when you say prices are falling.

Freddio

Yes - It shows they are rising since the winter just gone but it also shows that the prices are the same as what they were a year ago, where as the RTE report is reporting increases.

There is another graph there showing that the number of price decreases continues to out strip the number of increases. This graph does show imminent converging that would be a better indicator that the market is finding its own level shortly rather than the CSO saying its been increasing for a year.

I cant imagine the CSO having a bigger dataset on Irish prices than the daft database. They may have an alternative dataset based on stamp duty statistics from revenue, which would be a significantly smaller dataset.

As a lecturer told us in our Maths degrees over two decades ago. "You can say anything with statistics", which was my point about the "defacto" article in RTE

andrew

Freddio wrote: »

Yes - It shows they are rising since the winter just gone but it also shows that the prices are the same as what they were a year ago, where as the RTE report is reporting increases.

There is another graph there showing that the number of price decreases continues to out strip the number of increases. This graph does show imminent converging that would be a better indicator that the market is finding its own level shortly rather than the CSO saying its been increasing for a year.

I cant imagine the CSO having a bigger dataset on Irish prices than the daft database. They may have an alternative dataset based on stamp duty statistics from revenue, which would be a significantly smaller dataset.


As a lecturer told us in our Maths degrees over two decades ago. "You can say anything with statistics", which was my point about the "defacto" article in RTE

If you've done maths/statistics, then you'll know that the sample size you need in order to obtain an accurate measure isn't particularly high, so long as it's random. Seeing as the CSO obtain their data directly from the 8 main lenders, might there be some sort of selection bias in operation with daftdrop? What methodology does daftdrop use?

Freddio

As far as I can see they scrape the raw data in asking prices from Daft. But if the CSO are getting the data from the banks then my point about them massaging the figures stands.

andrew

Freddio wrote: »

As far as I can see they scrape the raw data in asking prices from Daft. But if the CSO are getting the data from the banks then my point about them massaging the figures stands.

I can think of two reasons the daft data might be biased then, if that's how it works. Either a) Daft don't have a random sample of houses on their website (which seems a bit unlikely) Or b) Using asking prices understates the actual value for which the property is sold. This seems pretty plausible; my understanding is that houses rarely go for at or below their asking price. The CSO would account for this seeing as it uses data about the actual mortgages acquired. Also, there could be a fault withthe daftdrop website.

You're saying that all the banks are falsifying data to the CSO? All of them? I doubt it, seeing as a) It'd be difficult to do b) Pretty easy to detect (Benfords law) c) Punished pretty harshly. The CSO have a pretty strong incentive not to use data which they think has been, or could be falsified. It also begs the question as to why they've just started falsifying data now, and why the banks themselves would bother. It doesn't really stand up to a cost/benefit analysis.

antoobrien

andrew wrote: »

This seems pretty plausible; my understanding is that houses rarely go for at or below their asking price.

For a few years the opposite of that was true, houses were selling for 20% (or more) below the asking price.

Last year something changed and it seems to be approach used by agents (at least in galway) - they started lowballing prices. This ignited bidding wars as there was much more interest than there would have been otherwise (I saw it happen with several houses in Gawlay).

How you also have the property actions that are putting in minimum values.In a few of the auctions in Galway the selling price was 50% above the AMV.

Geuze

barneystinson wrote: »

I'd have thought CSO were operating out of a bigger, better dataset, but since you haven't bothered your arse to find out before you started firing accusations, and I'm not inclined to do it for you, we'll have to wait for a grown-up to arrive...

I'm here now..........

Background Notes

Definition
The Residential Property Price Index (RPPI) is designed to measure the change in the average level of prices paid for residential properties sold in Ireland. The index is mix-adjusted to allow for the fact that different types of property are sold in different periods.

Data Source
The RPPI is compiled using data on mortgage drawdowns provided on a monthly basis by 8 of the main Mortgage Lending Institutions under Section 13 of the Housing Act (2002). This data provides details on the characteristics of properties bought (such as building type and size) as well as the price paid. It is transactions based; meaning that prices are recorded only where a sale occurs. Not all residential property transactions are funded by a mortgage (i.e. they are cash based) and these transactions are excluded from the scope of the index.

The CSO is currently examining Stamp Duty returns to the Revenue Commissioners, made via the Revenue Online Service (ROS), with a view to assessing both the extent of cash-based full market price transactions and any potential bias in the RPPI that might accrue from their exclusion.

Geuze

Most recent CSO RPPI here:

http://www.cso.ie/en/releasesandpublications/er/rppi/residentialpropertypriceindexmay2014/#.U61N1PldWSo

andrew

antoobrien wrote: »

For a few years the opposite of that was true, houses were selling for 20% (or more) below the asking price.

Last year something changed and it seems to be approach used by agents (at least in galway) - they started lowballing prices.

I suppose when prices are falling you'd expect the actual price to be less than the asking price, and when prices are rising you'd expect the actual price to be greater than the asking price.

Zubeneschamali

andrew wrote: »

I suppose when prices are falling you'd expect the actual price to be less than the asking price, and when prices are rising you'd expect the actual price to be greater than the asking price.

Which also means that when lots of properties fail to sell, people think "Values are dropping", while if agents are lowballing prices, and everything sells for more than the estimate, people think "Oh no, gotta get on that ladder now!".

Freddio

Geuze wrote: »

I'm here now..........

Data Source
The RPPI is compiled using data on mortgage drawdowns provided on a monthly basis by 8 of the main Mortgage Lending Institutions under Section 13 of the Housing Act (2002). This data provides details on the characteristics of properties bought (such as building type and size) as well as the price paid. It is transactions based; meaning that prices are recorded only where a sale occurs. Not all residential property transactions are funded by a mortgage (i.e. they are cash based) and these transactions are excluded from the scope of the index.

So what in fact these statistics are telling us is that the Banks are giving on average bigger mortgages which people are using to buy bigger houses.

EDIT: Bigger than they were a year ago.

andrew

Freddio wrote: »

So what in fact these statistics are telling us is that the Banks are giving on average bigger mortgages which people are using to buy bigger houses.

EDIT: Bigger than they were a year ago.

Bigger mortgages yes (because prices are going up), but not necessarily because the houses are bigger - how did you make that interpretation?

Which also means that when lots of properties fail to sell, people think "Values are dropping", while if agents are lowballing prices, and everything sells for more than the estimate, people think "Oh no, gotta get on that ladder now!".

If lots of properties do fail to sell, it's because the price is too high. So people rightly think 'values are dropping' because values have to drop to clear the market. Similarly, if everything sells for higher than (even low-balled) asking prices, then people rightly think that at the very least asking prices will rise, because clearly asking prices are too low.

Freddio

how did you make that interpretation?

daftdrop.com

Freddio

how did you make that interpretation?

A bank will grant a mortgage on an application - I dont see many applications being re applied for because the price of that unit has gone up.

andrew

Freddio wrote: »

daftdrop.com

I don't see anything in relation to changes in the size of houses being sold on the site?

Freddio

andrew wrote: »

I don't see anything in relation to changes in the size of houses being sold on the site?

There is a chart showing number of descreased prices versus number of increased prices on a monthly basis

andrew

Freddio wrote: »

There is a chart showing number of descreased prices versus number of increased prices on a monthly basis

But you said people were buying 'bigger' houses - that graph doesn't say anything about house size, right?

Freddio

andrew wrote: »

But you said people were buying 'bigger' houses - that graph doesn't say anything about house size, right?

A single price increase would be recorded on the specific listing on the database as indeed would a decrease

EDIT: which means more units are still coming down in asking price than going up. Thus if there was an increase in house prices based on sales, many more purchases of increased asking prices than decreased asking prices would be needed.

The market would always gravitate to the sales of the best possible within a budget. i.e Bigger houses

andrew

Freddio wrote: »

A single price increase would be recorded on the specific listing on the database as indeed would a decrease

EDIT: which means more units are still coming down in asking price than going up

Daftdrop takes a look at asking prices. So, for example, there's an apartment on Dorset street for which daftdrop has 7 prices. In May 2014 it was offered for €189,000. By June this was down to €163,000. But the downward revision of this price, doesn't mean house prices in general are falling. It might be that a year ago, that apartment was only worth €120,000, in which case the value of that apartment has actually risen, even though daftdrop is showing asking price drops.

More generally, the number of asking price decreases/increases is only a good measure of how accurate asking prices are. The fewer changes to asking prices, the closer the asking price is to the market price.

Also, you would expect there to me more downward asking price revisions than upward ones. Asking prices tend not to be revised upward - they're just sold for a higher price. However, prices will be revised downward until they're sold, and these revisions will all be recorded on daftdrop, even if the initial price was wildly optimistic.

The market would always gravitate to the sales of the best possible within a budget. i.e Bigger houses

I'm still not following your logic there to be honest.

Freddio

andrew wrote: »

I'm still not following your logic there to be honest.

To make these CSO / Bank statistics valid they would need to assess units that would have sold twice in this year on year period.

andrew

Freddio wrote: »

To make these CSO / Bank statistics valid they would need to assess units that would have sold twice in this year on year period.

I'm not sure how they do it. I get the impression that they take a sample of the mortgages given, and take a look at whether the average (or some other measure) has increased or decreased. A statistically significant random sample will be able to say, in general, whether house prices are increasing or decreasing

Edit: After controlling for hosue size that is, seeing as they collect that data too.

aliveandkicking

For those interested in this topic I'd reccomend the following threads on the property pin. Some excellent data analysis done by a couple of posters over there:

http://www.thepropertypin.com/viewtopic.php?p=771021#p771021

http://www.thepropertypin.com/viewtopic.php?f=10&t=62721&sid=b16004a099446a9995c671fb84e3e273

Godge

Freddio wrote: »

daftdrop.com

This is funny. You talk about bias but if a website is called daftdrop, they have a pretty big incentive to report prices only going one way

Freddio

Godge wrote: »

This is funny. You talk about bias but if a website is called daftdrop, they have a pretty big incentive to report prices only going one way

I think they set up to report the market status to prospective house buyers and give them some transparency to be honest.

everdead.ie

In Galway at least Houses in sought after areas are moving up apartments are much slower to move and less desirable areas are still stagnating.