Media Misleads On Monthly Data

Posted on 13/01/2018  
Media Misleads On Monthly Data

By Terry Ryder, founder,

Headlines such as “Melbourne home values drop” appeared in major newspapers recently, based on weak evidence.

The claim that Melbourne values were falling was false, in my view, and the inference this marked the end of the property boom was unreasonable and not supported by the facts.

It was yet another case of non-experts misinterpreting minor month-to-month changes in house price indexes.

The researchers at CoreLogic, who produce this data, know that small fluctuations in monthly price indexes are irrelevant, but they publish them anyway because it’s free and easy publicity.

The figures they rushed out to media at the end of December claimed that the index for Melbourne had dropped 0.2% for the month. Does this have any real significance? In my view, No, it’s a very small change and probably a statistical aberration.

It’s significant only if the next two or three months also show declines and the figures are confirmed by data from other sources. Then we may have a trend worthy of comment.

It’s worth recalling that CoreLogic recorded a 1.3% drop in the Melbourne price index in November 2016. At the time, media declared the end of the Melbourne boom.

But the following month the index rose 3.3%, at which point media declared the boom was back on again. Melbourne prices have kept rising since then and that premature declaration of the end of Melbourne’s boom was forgotten.

In Adelaide, the monthly changes have been up-down up-down month after month for the past 18 months. No two consecutive months have had the same direction, with the price index up one month and then down the next.

But the overall trend has been up, at the rate of 4-5% per year.

In Perth, there have been several occasions in the past 18 months when the monthly index has risen, according to CoreLogic, but the overall trend has been decidedly downwards.

I’ll say it one more time: month-to-month changes are misleading and irrelevant - and no self-respecting researcher, or journalist, would publish them.

But they do.

We see month-to-month statistics published on all manner of subjects, with economists, journalists and other commentators making sweeping interpretations of the impact of these figures, when in reality they often mean nothing at all.

They’re just statistical aberrations, because one month is too short a time frame to show a trend in the figures.

You need to take a long-term view, looking at quarterly and annual changes to determine the underlying noise in a market.

There has been major excitement in media over the latest data on dwelling approvals from the Australian Bureau of Statistics.

This showed a 12% monthly rise in November, led by a 38% increase for apartment approvals, which was described on one major newspaper as “stunning”.

One big publication shouted in a headline that the building industry was “back with a vengeance”. Economists and other commentators expressed surprise, shock and amazement.

But it was all much ado about nothing. It was just a statistical aberration, just one month’s figures, all caused by a number of large projects being approved in inner-city Melbourne which continues to have a building frenzy in the apartment market.

The figures didn’t say anything about the situation with the residential building sector nationwide.

And it’s highly likely that the next month’s figures will show a decline – at which time, no doubt, media will scream about the end of the boom.

As Shakespeare commented, “ it is a tale, told by an idiot, full of sound and fury, signifying nothing.”

The nation is full of economists and other commentators desperately trying to attract attention, and media outlets desperately trying to attract readers – all willing to grandstand about the importance of data that really has no importance at all.

As consumers, you can safely ignore all of them.

« Back