Lies, Damn Lies and Housing Statistics
Posted on 10 February 2009 by
In its “News & Views” publication today the CML rightly makes the point that the sharp percentage increase in net mortgage lending in December (up from November’s £0.8bn to £1.9bn) shown by The Bank of England data should not be interpreted as showing a significant change of direction. They add that mortgage lending is likely to remain subdued “in the coming month” and highlight the dangers of reading too much into a single month’s figures.
The CML makes the point that “relatively small movements in monthly net lending totals produce an exaggerated effect in percentage terms when net lending is close to zero.” This of course is correct and it is probable that for most months in 2009 there will be very large percentage changes on a month by month basis in the figures, sometimes an increase and sometimes a decrease. At least this will give some of the tabloids the opportunity for plenty of eye catching headlines this year.
A point the CML didn’t make, except perhaps by implication, but which I think is worth spelling out, is that when the net lending figure is small its volatility is magnified because it is the difference between 2 far larger figures, gross lending and repayments. Gross lending in December was around £14bn and repayments £12bn odd. A relatively small movement in either or both of these figures could make a massive difference in percentage terms to the net lending figure. In fact net lending would either go negative or more than double with a change of less than 10% in both figures, if one was higher and the other lower.
One even more important reason why the figures should be treated with extreme caution is that the Bank doctors them. It would of course not use that word – it prefers the technical term much loved by economists: “seasonally adjusted,” although when the figures are announced little, if any, mention is made of the fact they are doctored and hence they get reported as gospel, which in itself causes some of the misleading interpretation by commentators.
In the current property market I maintain that it is impossible to sensibly seasonally adjust figures. I am not saying that the seasons are irrelevant but we are in unchartered territory and the degree to which the seasons impact on activity will be very different to a more normal market. Furthermore, although seasonality will always have some relevance, other factors increasingly impact on activity in the property market, particularly sharp movements in interest rates.
Therefore I would like to see these Bank of England figures, and others, and the various House Price Indices, announced with the real figures given at least equal prominence. Most commentators are intelligent enough to make their own judgement on which set of figures are more helpful and the provider of the figures will always be in a position to provide the spin they want to put on them. It is not difficult to compare the figures for one year with the figures for the same months of the previous year to get an indication of whether the monthly movements are significantly different from year to year.
The net lending figures are one of the worst (or best, depending on which way you look at it) examples of the huge impact seasonal adjustments sometimes make. For the 12 months of 2008, starting with January, the adjustments The Bank has made from the real figures to produce the seasonally adjusted figures were: -10%, + 47%, + 5%, + 31%, -23%, -32%, -37%, -40%, -25%, -31%, + 78%, + 53%. These changes are much larger than those made in 2007 and indeed bear no resemblance to the changes made for that year, which were +10%, +14%, +33%, -7%, +28%, -10%, -17%, -4%, -16%, -4%, +8%, -9%, +19%.
As long as The Bank continues to announce only the doctored figures for net lending the figures need to be treated with extreme caution. The least The Bank should do is to ensure the real figures are available on its web site when the press release is issued. At the moment they are not available until sometime later. Better still they should publish both sets of figures in the press release.
Categories: Bank of England, Mortgages
The blog postings on this site solely reflect the personal views of the authors and do not neccessarily represent the views, positions, strategies or opinions of Charcol Limited. All comments are made in good faith, and neither Charcol Limited nor Ray Boulger will accept liability for them.
Post a Comment
Please keep your comments relevant. Charcol reserves the right to edit or delete comments.