Opinion: Is house price growth accelerating or slowing? And how on earth are we supposed to be able to tell?

By David Hargreaves

I don't mind if people don't always believe things I write.

I do, however, have cause for concern when I have difficulty myself believing something I write.

This was the problem in front of me when I reported on the latest Real Estate Institute figures for March.

The information from REINZ said that annual house price inflation had risen, just in the past month, to 9.2% from 8.2%. Now, I'm not saying that didn't happen. Because I don’t know. It's just that my gut instinct tells me there's no way it happened.

Remember this information came out in the same week that the Government valuer, Quotable Value, reported that house price inflation on an annual basis had fallen in the past month to 8.8% from 9.3%.

Confusing? You bet.

So, it was with something approaching relief that I read Westpac economists’ latest Home Truths publication, which stated with refreshing candour: "It is impossible to tell what is really going on with house prices."

However, my relief should be other people's discomfort.

Impossible to tell what is really going on with house prices? Really? Oh, dear. Is this ideal?

We seem to have placed ourselves in an interesting position here.

A juggling act

On the one hand at the moment there is a big requirement for timely house price inflation information, largely because the Reserve Bank has put in place its ‘speed limits’ on high loan-to-value lending and we need to measure what impact these are having.

But on the other hand big question marks are appearing over the veracity of some house price information, er, because the Reserve Bank’s ‘speed limits’ on high loan-to-value lending are doing really weird things to the housing market...

Yes, that’s right. We need information in order to measure the impact of LVRs - and the LVRs themselves are putting the spanner in the works regarding effective collation of such information.

Now, the REINZ might argue with me, and say that its figures are painting an accurate portrait of the current market. Only a few months more worth of data might really prove that point.

But whatever the view, there’s no question that measuring house price inflation has always been a pretty vexed issue.

Try cornflakes

If you want to measure how much the price of cornflakes is going up, buy a packet one month and then buy another packet of the same brand the next month. Then compare prices. Easy.

But houses? A similar method of pricing comparison would only really work if you sold the same house in the same street every month. Otherwise it’s never truly comparing apples with apples. The composition of house sales is never going to be the same month by month, so trying to establish some sort of coherent pricing trend is always problematic.

In New Zealand we have two main sources of house price information: REINZ and QV.

In basic terms the difference between the two is that REINZ records house sales the minute an offer goes unconditional, while QV records transactions at settlement and compares the price attained against valuation.

Based on the methodology it could be reasonably expected that the QV information would be more accurate, not least because it's based on a final outcome, IE the finished sale. The trouble is, because it records transactions only at settlement, its information tends to ‘lag’ by around two months.

When the market is really heating, as it was in Auckland last year, there are some months when the QV information really doesn’t reflect what’s happening on the ‘street’ right then and there.

Hot hands

Which is where the REINZ information comes in. It’s hot from the estate agents’ hands and it measures a sale as soon as an offer goes unconditional. Which means, of course, that it's very timely.

Personally, I’ve tended to prefer the REINZ data as offering an insight into the real market mood of the moment.

But then came the LVRs…

As if it wasn’t difficult enough to compare house sales on a month by month basis, suddenly there’s a new regulation that completely skews house buying trends.

Because so many first home buyers have currently dropped out of the market, the number of houses sold at the lower end of the house price range has dropped, sharply.

In its latest release REINZ said that while the total number of sales for March 2014 was down 10% compared with March 2013, the number of sales below $400,000 fell by 21.9%. This followed a fall in sales below $400,000 of 17.7% between February 2014 and February 2013.

What this means is that the ‘normal’ range of prices from nought to whatever has been totally skewed, because the low end has been artificially knee-capped due to a new regulation.

A nightmare on Auckland's streets

This further means that calculating the average and the median prices becomes a nightmare.

Remove the bottom end of the price range and obviously both the average and the median prices move up considerably – but does that mean that prices have really risen, when what’s really simply happened is that a whole section of the market has been taken out?

The REINZ Monthly Housing Price Index (HPI) is, in REINZ’s own words, calculated using a technique known as stratification, which provides an averaging of sales prices for common groups of houses.

“This approach is considered a more robust analysis of actual house price trends and was developed in conjunction with the Reserve Bank,” REINZ says.

Well, “more robust” or not, I agree with Westpac chief economist Dominick Stephens that “in the current circumstances the HPI may have been skewed”.

As stated earlier, another few months of sales figures will help tell whether that is the case.

Where are we?

But, disappointingly, it appears that at the moment we are struggling for timely information that tells us exactly where the housing market is right now.

Is house price growth easing as QV’s figures suggest? Or, are those QV figures picking up what happened two months ago – IE an easing – which has now been overtaken by another wave of price rises? Are the REINZ figures, despite what I may think, actually showing a genuine new surge of house price inflation?

I really don’t know. And it would be nice to know.

I wonder if the RBNZ had any idea of the difficulties its LVRs would cause in measuring house price inflation.

The situation seems truly ironic - when one of the aims (though the primary aim was financial stability) of the LVRs was to take some heat out of house price inflation. So, introduce a measure to curb house price inflation and then not be able to gauge house price inflation because of the measure you've just introduced...Interesting, to say the least.

Do you believe?

Right now probably most people are inclined to believe (if we accept QV as more accurate) that the LVRs have put the dampener on house price inflation, but we’re not absolutely certain!

So, the question is, do we have a better “more robust” way of monitoring house price inflation available? And, no, dear reader, I’m afraid the question is not rhetorical. I don't know;  it’s a genuine question.

Maybe an LVR-specific house price inflation measure can be developed. Maybe somebody should have thought of that first. Or did the RBNZ not realise the extent to which its LVRs would kill off the bottom end of the house market, however temporarily?

But it may be the RBNZ will be happy enough with the QV figures, slight time lag and all, which are suggesting an easing in house price growth.

To my mind, however, it’s not a particularly satisfactory situation.