Large differences between house price surveys across all the major players are often apparent, particularly at times when the market is fluctuating. Interested observers have long argued the various merits of different schemes, mostly with emphasis on time lag and whether the prices used are those paid, or those agreed before sale.

However, in a press release yesterday Hometrack discloses further effects that it says cause the big differences currently seen between indexes, and says most indexes may be misleading commentators and policy makers.

One key reason why house prices indices vary is they monitor price information at different stages in the house buying process, and they don’t necessarily take into account areas where activity has slowed considerably.

Hometrack said yesterday that it’s latest survey shows an annual price fall of 3.6% as opposed to an increase shown by other indexes such as Nationwide and Halifax. Home track monitors prices agreed.

Comparison of annual house price indexes (% annual change in average house prices in the UK)

With Hometrack monitoring house prices at point of agreed sale price, Halifax and Nationwide at point of mortgage approval and ODPM and Land Registry up to four weeks and six weeks after mortgage completion respectively, there is a significant time lag of up to six months in house price commentary. However this does not explain the major differences between the indexes, said Hometrack.

So what is the difference?

Hometrack’s monthly index monitors the overall housing stock, not just those areas that have market activity. By polling estate agents active in every market, Hometrack is able to also include those markets where there is limited activity.

The first indication of a market decline is typically a lack of activity, as sellers are slower to adjust to lowered expectations than buyers. Sellers will normally not accept an offer until they are convinced that a higher one will not be forthcoming. By tracking prices from the best estate agents around the country, Hometrack’s index is able to factor in the value of all housing stock including areas of more limited activity. Indexes reliant solely on transactions are highly volatile and are biased towards the sector of the market that is currently the most buoyant.

Another significant factor in monitoring the value of the UK housing stock is the ‘first time buyer’ effect. With the lowest number of first time buyers in over 25 years at 31% (versus 50-55% traditionally) and the levels of activity dropping by 33% from the first quarter of 2004, activity at the bottom end of the market is at an all time low. In the current market it is no surprise transaction based indexes show an increase in average property prices as they reflect a snapshot of a higher proportion of more expensive (second and third time buyers) properties.

John Wriglesworth, housing economist of Hometrack comments: "Housing surveys documenting an increase in annual growth are misleading policy decision makers and the Bank of England."

House price surveys that are dependent upon transaction prices are not taking into consideration the property prices across the whole market and are therefore distorting the figures and over-egging the market. This distortion is currently at its most marked with the extremely low levels of activity at the bottom end of the market."

Who’s to believe?

Well, they are all correct in what they do, and if used in the way they are designed to be used. So briefly, what are the major differences?

Two property website ones Hometrack and Rightmove collect data from asking prices, but they statistically work differently and don’t produce identical results. Their big failing is that they don’t source the actual price paid and whether the sale was successful. But the trends are clear and the results come earliest.

One of the issues with surveys like Halifax and Nationwide is that they are conducted by mortgage lenders and so don’t record sales made without mortgages, currently about a quarter of UK homes are sold with cash transactions.

The Land registry survey, which comes out every three months lags the market by a long way but is arguably the most exact as it records every property transaction. (Except re-possessions and transactions caused by divorce as these would skew the figures.)

The only survey to explicitly measure the price movements of first-time buyer properties is the government’s experimental survey which comes out monthly but lags most of the other surveys. Its data comes only from mortgage lenders.