Here's a chart based on Table ML2 from the Council of Mortgage Lenders statistics page, which shows the years 1974 to 2007. The red line is 'interest payments as a % of income' (left-hand axis) and the blue line is 'loan as multiple of income' (right-hand axis). The typical LTV was between 85% and 90% for most years, so that does not seem overly relevant.
So, all this chat about 'affordability' is a smokescreen - the loan-to-income ratio is at an all-time high of 3.36, so if average interest rates for FTB's have increased by 1.5% from a very low figure of 5.75% in late 2007 to about 7.25% as at April 2008, this will push the interest-to-income ratio from 19.4%* up to 24.4%, which is precisely the pain threshold at which the market turned in 1989.
Assuming average interest rates for first time buyers with a 10% deposit stay constant, prices would have to fall by about half (relative to wages) before they reached the lows of 1977 or the early 1990s, where a natural floor was reached, i.e. it is far cheaper buying than renting.
* Update, as Matthew points out, the spreadsheet has been updated and the figures for February 2008 are loan-to-income ratio 3.33 (i.e. down by 1%) and the interest-to-income ratio, which had risen to 20.7, is now at 20.1, so things are slightly worse than I thought.
A grim document
6 hours ago
7 comments:
Surely the other factor is confidence. People may still be able to afford and demand homes, but if they believe prices are about to drop then they may be holding back for fear of losing money. Thus if people feel the shit is very much hitting the fan then they will precipitate a crash despite no reaching the pain threshold you suggest.
Exactly. Land values = speculative values. The whole thing is a huge confidence trick, in the most literal sense.
I think you've missed some of the data - it goes to Feb 08, when in fact the mortgage payments as a % of total fell slightly compared with January, and anyway are 20.1%.
In 1990 they peaked at 28.1%.
Hmm. Where did you get those figures from?
ML2, here
http://www.cml.org.uk/cml/statistics
Monthly figures are only included for the last few years, so I used the annual figures that go all the way back to 1974.
Yes, but when you were speculating about what the ratio was in 2008 surely it's better to use the 2008 data? Quarterly data goes back a long way, btw.
Post a Comment