sign up log in
Want to go ad-free? Find out how, here.

Alistair Helm shows how reported median house prices can rise during a structural market shift even though no individual house prices change

Alistair Helm shows how reported median house prices can rise during a structural market shift even though no individual house prices change

By Alistair Helm*

How can the median price be rising when property sale prices aren't rising?

This is not a conundrum, it is in my opinion the reality of the situation in today’s market and has lead to these fairly striking statement by some leading economists:

Westpac's Chief Economist Dominick Stephens said "It is impossible to tell what is really going on with house prices".

New Zealand Institute of Economic Research (NZIER) principal economist Shamubeel Eaqub said "trying to forecast house prices has been a mug's game".

The reason behind these statements is the conflicting data coming out from REINZ and from QV - specifically the March median price data which showed an accelerating rise in price from 8.2% year-on-year growth in February to 9.2% in March set against a fall in sales of 10% and QV reporting that the rate of price increase was easing.

In a single month REINZ reported the median price of NZ property had risen from $415,000 to $440,000.

There has even been questions as to the accuracy of the REINZ data.

This is not something I believe, or have any insight into, but such comments certainly demonstrate the concern in the market as so much value is attached to the timely and accurate indicators of the property market by so many sectors of the economy.

I have long advocated the use of the REINZ Stratified Median price index as a more accurate methodology for tracking the true indicator of the price of property sales across the country, however even this measure, long trusted for its lack of volatility has of late shown some wild fluctuations. 

Tracking these fluctuations over the past 20 years as the chart below highlights shows that on a 3 month moving average basis the recent decade has shown a normal fluctuation range of around $3,900 from month to month, this was a higher level of volatility when compared to the 90’s when the volatility was less than $1,700 month to month.

In the last 3 months this volatility has spiked with the 3 months average for the 2014 year so far showing a volatility month-to-month exceeding $11,000 - a highly volatile situation.

So why is it that we are seeing this volatility?

In my opinion the impact of the LVR restrictions are having a greater impact on the property market than is currently being acknowledged.

Let's look at the facts:

  • Sales of lower priced properties are down - the data is reported by both REINZ and in the Auckland market by Barfoot & Thompson. 
  • Overall property sales have already come off their peak and are easing - 10% down in the year to March.
  • The retail banks have demonstrated an ‘over-correction’ to the Reserve Bank imposed 10% criteria for high LVR lending, resulting in upwards of a 90% fall in the approval of 80+% LVR mortgages.

To better assist in understanding how these indicators might be contributing to the volatility in the median price I have developed a hypothetical scenario of the composition of the property market sales in a month. The by comparing this with a subsequent month where the underlying property prices remain the same year-on-year across all price sectors, where sales volumes remain unchanged across all price sectors, with the exception of the lower priced end of the market and let me show the impact.

Here is a hypothetical normal distribution of property sales in say March 2013 - 5,082 sales with a median price of $400,000 and for reference an average price of $507,000

Now let’s jump forward to March 2014 - let's reduce by 23% the sales volumes in the price ranges $225,000 to $400,000 - just these price ranges. All other sales volumes by price range remain identical to the year earlier.

The outcome of this impact (the hypothetical impact of the LVR restriction) is shown in the chart by the marginal sales reduction in red across those price ranges.

That 23% fall in sales across those lower priced properties segment leads to an overall 10% fall in total sales to 4,567.

The median price though went up by 8% to $431,000 and the average price went up by just 4% to $528,000.

This model is designed to demonstrate that what we could be experiencing is two components of the property market working in complete isolation.

The majority of the property market is plodding on unaffected with the no change in year-on-year sales volumes, and not experiencing any price appreciation.

While in those sectors directly affected by the LVR restriction the sales volumes have dropped by 23% but equally with no price change amongst those sales.

The net effect though is that the signal being sent out to the market through the median sales price is that property prices overall are rising in an inflationary manner.

A classic situation where aggregated statistics belies the true situation.


The above article was written by Alistair Helm, and is republished with his approval. The article was originally published on Properazzi here

We welcome your comments below. If you are not already registered, please register to comment.

Remember we welcome robust, respectful and insightful debate. We don't welcome abusive or defamatory comments and will de-register those repeatedly making such comments. Our current comment policy is here.


Good point and could also be first home buyers might be in he apartment maket (like my younger sister) so less of them are being bought as LVR's hit. Maybe there is some analysis on apartments in AKL CBD, have they gone down in sales? We spoke last year about there should really be a 3 bedroom house indicator (and then the rest!) Mixing apartments with houses does not give a true representation of house prices.


If you agree this abuse of statistics produces very misleading outcomes - consider the fact that household debt to incomes is averaged over all homes including those with no mortage.


Net effect is to grossly understate the extent of indebtedness of those with mortgages.


Household debt to incomes for those with mortages will be at least twice the published figures if half  don't have mortgages.


It is totally incomprehensible how the RBNZ can evaluate risk without using specific debt ratios  of those with mortgages.


I'm thinking that if you needed this spelled out, as is well covered above, then you really have not got the level of understanding required to make correct assessments in this business. Concentrating capital at the top of the market means the toffs cop it most when it all flops. Works for me.

Similarly, I was stunned at the poor grasp of the fundamentals of revolving credit when it was mentioned recently. By the comments, otherwise educated people showed their true grasp of reality (or lack) there.

Household income being maxed out as almost all participants now earn and contribute means that the saturation point of credit expansion and  the physical limits of repayment ability are close. True home buyers are having to question the old drivel and ubiquituous published untruths in the face of poor affordability and obvious debt servitude.

Even at low LVR, deposit saving from scratch is daunting. Many now rely on inheretence alone for this. If your parents were not in the market it will be tough.

And the blinkered view that this is about FHB's just irks me, speculators put in as little as possible and are quick to change their habits when the ROI numbers no longer stack up.

Despite the wails of despair I hope that this all means that a longish capital gain flat spot flushes these speculators and baby boomer property accumulators so shelter can be evenly spread again.