The property industry's game of dodgy economics figures: Catherine Cashmore
The REIV have released their June quarterly statistics which show a seasonally adjusted 2.4% rise in the median house price with the comment
“The release of the REIV's June quarter median prices should help encourage a few more sellers into the market over spring as it shows the price falls recorded in 2011 have been recovered.”
This takes Melbourne’s median house price from a ‘revised’ $549,000 in the March quarter of this year which – (as Terry Ryder also noted Tuesday last week and myself on Monday last week) - is a ‘not so insignificant’ drop of - 2.23% decrease to the figure the REIV initially published in April of $561,500 – to the now ‘seasonally adjusted’ June median of $562,000.
In other words, had the numbers been taken on face value, without subsequent revision, Melbourne would have simply been treading water.
All of this means little to the average buyer who simply reads the headlines and has little time to dig into the detail – however at the time, the REIV heralded the first quarter of 2013 to be the ‘strongest March quarter in a decade’ which was somewhat surprising considering the ‘unadjusted’ March median actually dropped -0.9%.
Of course, the only reason it could be stated as such, was because their December 2012 median had also been ‘seasonally adjusted’ & revised from $555,000 to $534,000 – a drop of -3.7%.
The REIV only started seasonally adjusting figures at the start of this year, therefore the December adjustment was made ‘post’ the initial release in which the REIV claimed a 7.8% jump in the median house price under the headline: "Growth returns to Melbourne housing market."
I remember this well, because it prompted a joyful remark I overheard one agent relay to his ‘investor’ client, that this would equate to more than “30% capital growth for Melbourne over the following 12 month period if it maintained pace” - thankfully a comment that will remain pure fiction.
As a general rule, seasonal adjustments are calculated by looking at the ‘average’ seasonal shift during the same quarter over a lengthy time scale.
The information is then used to calculate the amount to either add, or subtract, from the raw data.
Whilst it can be useful from a trend perspective (smoothing out the bumps), there is scant information on the REIV website to explain how a drop of -0.9% can be converted into in to a somewhat robust ‘rise’ of 5.1% in a market which, although improving, was certainly not doing as well these numbers would indicate.
The reason figures are revised, is because all data providers suffer from a lag in reported results.
These typically filter in over the 3 month period ‘post’ the quarterly release from valuer general data – the most authoritative and comprehensive source of sales information we have.
Therefore, the median price home buyers and investors are reading for this quarter, and the emphasis that is put on it as newspapers go to town with boom or bust headlines, is likely to alter significantly when the next release is due. Therefore, it is arguably a misleading ‘quarterly’ indicator of true market movements.
In other words – any ‘newly’ released quarterly data needs to be taken with a pinch of salt.
With this in mind, it makes the daily index – which is neither seasonally adjusted and would also miss a wide proportion of lagging results - laughable.
Investors are often fixated on median prices.
The major investment magazines have extensive tables of figures and percentages covering the back pages. Agents use them to their own ends with comments such as "prices have increased 10% year on year over the past 5/10 years for such and such a suburb."
However whilst the median figure may have increased – the ‘middle’ figure of all cited sales – individual property prices, and the changes a property may go through in terms of renovation and extension, which would therefore warrant a higher capital price outside of natural increases, is not always represented in the information provided.
It should also be noted that each provider uses a slightly different methodology when collating their statistics.
For example, the REIV record a ‘seasonally adjusted’ 8.4% rise to the median from this time last year which seems to suggest Melbourne is coming on leaps and bounds.
However, the ABS show an increase of just 1.1% (to March 2013) - APM: 6.1% (to June 2013) and RP Data: 3.3% (to June 2013) making the REIV’s median figures higher that of every other data provider.
But, the REIV are not the only culprits when it comes to publishing data misnomers.
As economist Leith Van Onselen pointed out to me in a conversation we had regarding APM’s results:
“In its March quarter release, APM reported that Melbourne house prices led the nation, rising by 3.6% over the quarter to $538,922. What was not mentioned in their commentary, however, was that some of Melbourne’s reported strong price growth was caused by a -1.1% downward revision to the December quarter......And in their June quarter results, APM once again reported that Melbourne led the nation, recording 5.0% growth over the quarter and 6.1% growth over the year. However, part of this strong quarterly rise was caused by a downward revision to March's median house price to $527,245, without which Melbourne would have recorded 2.7% growth."As Leith points out, without the subsequent revisions “ Melbourne's annual price growth would have come in at a more moderate 4.2%.”
One indication towards the stark contrast in REIV statistics is because unlike the other indexes, they do not stratify their median figures to reflect different aspects of each housing type outside of the broad description of units and houses.
This creates significant problems for those using the information as a source of market analysis.
In REIV terms, a unit could be a small villa on a subdivided block of typically six to eight free standing or attached dwellings - a one or two bedroom apartment or flat, in either a low rise or high rise block , a high spec townhouse on a ‘side by side’ subdivision, or a bedsit.
Obviously, this can create distortions when assessing the information.
The same is the case for housing, median house prices cover detached houses, terraced houses, semi-detached houses, residential warehouse conversions, holiday houses and duplexes. Because the REIV don’t distinctly classify their results, we also see big differences in individual suburb changes.
For example, according to the June stats, Hawthorn’s unit median has increased 20% from the last quarter.
However, there should be no confusion here – individual unit prices in Hawthorn have not increased 20% – this is median data, and without stratified statistics, the median price can easily be boosted with different property types being lumped under the ‘unit’ banner.
To publicize these figures, with full knowledge of the reaction it will create, should be an area of concern, yet one which is ignored in the mainstream press.
Albeit, from an anecdotal perspective, property prices in Melbourne have increased throughout the course of 2013.
As an example, a unit which based on comparable data from late 2012 and early 2013, would be worth anything in the region of $420,000-$440,000 – under current competition, often ends up selling 2-3% higher.
This is an effect which is primarily noted in the inner and middle ring established suburbs of the city where auction sales predominate and an intense level of investor and speculative activity is evident.
Most results are staying broadly within these parameters; however, I’ve seen some crazy activity of late as mini bidding wars in various pockets have rippled across all price brackets
This can have quite a dramatic impact on both vendor expectation (as owners see neighbouring properties sell above their pre-estimate of value) and buyer physiology. Property shoppers realise they need to ‘up the budget’ to exceed what is perceived to be some kind of boom time terrain, producing gains that cannot be sustained over the resulting period – and in the current economic environment – will most likely eventuate to be nothing much more than a short term rally.
As most sales across Melbourne are conducted private treaty, the effects on the median data across the board would be minimal, and really only feed into inner city figures where auctions are the preferred method of sale.
In truth – from a macro perspective, as the latest RPData report highlights, “In five and a half years, growth in capital city home values has not increased at a rate higher than inflation”.
It should be noted that Australia is facing some significant headwinds which will have an impact on the property market in the months and years ahead.
Full time jobs growth is weak and both hours worked along with levels of underutilisation are deteriorating.
As the mining construction boom unwinds, non mining industries underperform, and we pull in our belts with fiscal consolidation, it’s likely we’ll see little change over the months ahead. The short term data is volatile and somewhat unreliable albeit, the long term trend is clear.
We’re entering an era of slow growth and it’s yet to be seen if it will be stable growth.
In this respect, when you lift the blanket on unemployment data the trends are concerning and certainly not as good as the headline rate suggests.
Outside of keeping interest rates low, or offering grants for new homes to enable buyers to take on a larger proportion of capital debt, there are no long term sustainable solutions offered from either side of politics.
The speculative behaviour we’re currently seeing in various pockets of capital city markets as investors lead the widely spruiked ‘housing recovery,’ neatly packaged under the words ‘typical market cycle’ as different states go through their own ‘wax and wane’ periods of supply/demand activity.
This will have far wider impacts for our younger and future population of buyers than is fully recognised across the broader population that already own a home, or are paying off a mortgage.
Catherine Cashmore is a market analyst with extensive experience in all aspects relating to property acquisition.