"CML predicts 7% house price fall"

Firstly, won't this sort of prediction become a self-fulfilling prophecy? What's the point of buying a house, paying 6% interest and suffering a 7% capital loss if you can rent a house for a typical rental yield of 4%?

Secondly, in calculating this, presumably all they did was stick a ruler on a graph and extrapolate - according to the April House Price Index published by the Halifax/HBOS (one of the CML's largest members), prices are down 5.3% since their peak in August 2007. 5.3% ÷ 8 months x 12 months = 8%.

Finally, although the recent past is usually a good guide to the near future, the recent past is not a good guide to the distant future, which is the mistake that Prof Stephen Nickell made, who a few months ago solemnly swore that house prices would rise at 1.5% a year faster than earnings for the next 18 years*. Has he resigned his cushy taxpayer-funded job with the 'National Housing and Planning Advice Unit' in shame? Nope, thought not.

* 9.3 ÷ 7.07 = 1.32, the 19th root of 1.32, =1.32^(1/19) = 1.015