Things that don’t add up

9 March 2010

Another thing that doesn’t quite add up is why the powers that be are extolling the virtues of narrow banking and depositor funded lending when so many building societies are being subsumed within their larger brethren.

A month or two ago it was the Chelsea going to the Yorkshire, a week ago it was the Chesham going to the Skipton and now the talk is of the Stroud & Swindon being sent to Coventry.

And as for the state owned banks, judging by the latest financial results, Royal Bank of Scotland, with its more risky broad banking structure, is likely to pay back taxpayers sooner than Lloyds Banking Group which is deemed to have the safer business model.

The perceived wisdom seems to be at odds with what’s happening in the market.

Then arguably less contentious is the house price conundrum and what in fact the seven published house price indices actually tell us.

For example, did house prices rise in 2009 by 1.1% (Halifax) or by 5.9% (Nationwide); or did they fall by 1.9% (Hometrack)?

I tried to address this indices problem several years ago in Lending Strategy with an in depth piece by the Council of Mortgage Lenders but as Acadametrics, a specialist agency in this area, points out in a new paper on the subject, a more vaulted figure was equally troubled by the same issue some 12 years ago.

The person in question was Mervyn King who was puzzled by an “unfortunate” divergence of the lender indices. The Bank of England, Acadametrics assures us, “now employs an average of the lenders results, as do many in the financial sector.”

But the current low level of business volumes can easily distort an analysis so whose figures should we trust? Acadametrics pithily points out that the government provides two official indices and three different house price measures, none of which are supported by the Bank.

Actually, we can trust all the indices, up to a point, so long as we understand their short comings and what it is that they are measuring.

Acadametrics defines these in detail but it’s useful to remember that Rightmove provides asking prices for a sample of properties offered on its website while Hometrack provides estimated values of what a panel of estate agents thinks is the achievable price for each of four property types.

Halifax and Nationwide on the other hand report mortgage offer valuations taken from a sample of properties taken on their books while the Communities & Local Government index provides mortgage completion pricesbased upon a sample of around 60% of mortgages.

This in turn is different to the Land Registry index which uses a sample of Land Registry transacted prices which includes both properties bought on mortgage and for cash. Acadametrics, as its paper implies, goes better than that and ultimately employs every transacted price from the Land Registry and drills down to local levels.

So although the seven house price indices don’t deliver the same sums at least we can understand why. The problem is that the monthly changes in the value of properties upon which Halifax and Nationwide offer mortgages tend to be mistaken for changes in the prices at which houses are being sold and specifically, as Acadametrics points out, an index based on a monthly loan book, if comprising ‘safe’ upmarket properties, may reflect price rises resulting from lack of supply, rather than price trends, nationally.