House prices: What the forecasters are saying

We read between the lines in the property figures and forecasts from Halifax, Land Registry, Nationwide, PwC and Capital Economics.

As always, the media is making as much as possible of the various monthly and quarterly reports on house prices, so I'm going to try to put them altogether and bring the whole show back to Earth.

I'll start with Halifax’s latest monthly report, because this is producing the most exciting news stories.

Halifax has reported a fall for the 'third successive month', or a fall in four of the past five months. Looking back at Halifax data, four falls in five months has happened six times since the beginning of 1983 (disregarding interlocking periods).

Three of those times we went on to have lower house prices one year later, while the other three times witnessed higher prices. Hence, by itself it doesn't seem to mean that we can expect lower house prices than today in a year’s time.

Also, in this report, Halifax used its seasonally adjusted data, which I find is normally more deceptive than useful. On Halifax's non-seasonally adjusted data – i.e. based on what the prices of properties mortgaged by Halifax actually seem to be selling for – prices have not fallen in four out of the last five months; they've just fallen in the past two months, taking us back to February/March 2010 prices.

I can count 21 times since the Halifax index began in January 1983 when house prices have fallen at least two months in a row. Ten of those times house prices have ended up lower one year later and 11 times they've been higher. Hence, the recent falls in Halifax data doesn't help us to forecast the future.

What about other data?

Let's move on. Nationwide data shows house prices still rising although the rate is slowing. This year house prices are up 3% and in June they were up just 0.1%. The increase over the 12 months ending June 2010 is 8.7%, whereas the annual increase at the end of May 2010 was 9.8%.

The rate is slowing because the increases over the past few months have been lower than a year earlier. However, I find that a slowing rate also has no statistical significance, which means we can't glean anything about the future with it.

The Land Registry has today announced its latest figures for England and Wales, which go up to May 2010. It says house prices were up that month by 0.1%, taking house prices to the same level as in summer 2006, at £166,000.

I'd like to make a brief comment about that. House prices being at £166,000 in 2006 were more at risk of a crash than house prices being at the same level four years later, in 2010, thanks largely to inflation. This is true whether or not house prices are still too high today, so it is certainly not to suggest that house prices can't crash again.

How about some analysis?

All that data really doesn't help us, so let's turn to forecasts.

Nationwide appears to be extrapolating on the past six months by suggesting house price inflation will slow for the rest of this year. That's not a bad prediction, because house prices rose extremely quickly in the past year or so, so it's more likely that house-price inflation will slow.

If we were to actually end up with house price deflation in December then Nationwide can at least say it wasn't too far out in its ideas. (It wouldn't want to predict falling house prices in any case, as that may help start a panic that would kickstart falls, and that would cost Nationwide money.) Therefore its prediction of slowing inflation is a safe call, and it's what I'd say if I was working for a mortgage lender and told I had to stick my neck out into the shabby world of forecasting.

Halifax's prediction is rather more daringly precise, predicting 'broadly unchanged' prices since the start of 2010, meaning prices in December should be pretty much what they were at the beginning of this year.

Yet these organisations have a hideous record of calling property prices, so let's turn to two more reports.

You may have read some of the big headlines made of the forecasts from PwC and Capital Economics. Both of the reports have been taken up by the press as signals that the property market is going to fall. However, Capital Economics' property forecasts have also been atrocious, at least this decade.

Furthermore, PwC's report was badly interpreted, as it was actually only potentially negative for property investors, with its central forecast actually being pretty good for both homeowners and wannabe homeowners alike, and overall rather neutral.

Apart from Capital Economics, none of the predictions really suggest big falls or big rises, and they're all pretty much steady as she goes or just safe, reputation-friendly forecasts. As usual, no one really convinces me that they know what's going to happen.

More: Machine gun your mortgage | The future of house prices