In the last week or so, the Royal Institute of Chartered Surveyors, the Nationwide Building Society, Hometrack, Halifax and the Land Registry have all published their latest round of house price statistics, with the Council of Mortgage Lenders and various estate agent networks such as TEAM property also getting in on the act from time to time.

From the latest set of data, depending on particular report you read, growth in the housing market could be virtually flat or soaring ahead at breakneck speed. Clearly, if you are thinking about buying or selling a house, it's pretty important to know which way the market is moving, but with so much seemingly contradictory information out there, how on earth are you supposed to know what is going on?

Hopefully, our analysis of the different indices will provide a useful guide to the different sets of data available as well as helping you understand why they can produce such different sets of results.

Link: http://www.hbosplc.com/view/housepriceindex/housepriceindex.asp

Source: The Halifax House Price Index are based on the detailed records of the prices, physical characteristics and the regional location of all the houses on which the organisation has made a mortgage offer in each time period. The Halifax is the UK's largest mortgage lender, meaning its figures are based on at least 20,000 transactions a month.

Not all house purchases financed by the Halifax are included in the database. There are some properties that are excluded from the subsequent analyses. For example, council house sales which may have been sold at below market price. After editing, the database covers around 13,500 house purchase transactions per month.

The data refers to property valuations at the time the mortgage was approved, rather than completed, bringing the disadvantage of covering some cases that never proceed to completion.

Methodology: Like the Nationwide index, the Halifax system was originally constructed by Joe Nellis, an economist from the Cranfield School of Management. The methodology is based on the 'hedonic' approach to price measurement, which means that the valuation is associated with a particular set of physical and locational attributes possessed by the property. For instance, a property may be valued at a certain price based on qualitative factors such as the type of property and the geographical location and quantitative characteristics such as the age of the property, the number and type of rooms it has, garden space, and so on.

The index uses complex statistical methods to incorporate all of the house price data and work it back to look at the price change each month or quarter for the same imaginary set of property that has the same physical and locational attributes.

Hopefully that is clearer than a blue sky, but if not, it is really much too complicated to go into detail about the methodology here and frankly no one here wants to spend enough time reading about statistical regression to be able to write about it with any authority.

Comments: The Halifax index is far from perfect as an accurate measure of current UK property values. It is based only on a relatively small sample of properties that change hands in any given period, which is almost certain not to represent the full range of getting on for 20 million owner-occupied properties in the UK. Of the houses that change hands, it only incorporates those that are bought using a mortgage, and then only those that are bought with a mortgage from the Halifax.

However, in much the same way that the retail price index represents the change in cost of a certain basket of goods that are representative of the different things a shopper may buy, the Halifax index is a highly useful measure of the change in valuation of a range of certain property types that are representative of the different homes a buyer may purchase. It is for this reason that the Halifax index is widely used by the government, the Bank of England Monetary Policy Committee and various other influential groups, to assess the impact of policy and help determine future decisions.

Link: http://www.nationwide.co.uk/hpi/monthly.htm

Source: The Nationwide is very similar in many respects to the Halifax House Price Index. The Nationwide sample is also based on the mortgage offers it has made in any given period of time. However, it is a smaller sample than the Halifax, which you could argue makes it less representative.

The data is checked to make sure that the sample only contains properties sold at a true market price. This means that remortgages and further advances must be excluded, as are identifiable purchases below market price, e.g. Right-To-Buy and shared ownership purchases with a housing association. Properties costing £1m or over are also excluded, as are houses above 3000 sq.ft (279m2) and flats above 2000 sq.ft (186m2).

As with the Halifax, the Nationwide sample consists only of its own customers and therefore does not include completions with no mortgage. The Nationwide will not say precisely what its monthly transaction level is.

Methodology: Having originated from the same place, the methodology used to calculate the Nationwide index is pretty similar to that of the Halifax, though they do claim to have done a lot of additional work to try and make it more accurate when dealing with small sample sizes (It is not believed that this was in any way a prediction of the dramatic slump in the Nationwide's share of new mortgage business in the past 12 months!).

The Nationwide index aims to try and replicate the perfect sample - a fixed set of properties that represent a microcosm of the entire market, resold on a monthly basis - using a range of variables associated with each property. These include location (part of country & class of area), floorsize, property design (detached house, semi-detached house, terraced house, bungalow, flat, etc), tenure (freehold/leasehold/feudal), number of bathrooms, type of central heating, type of garage, number of bedrooms and whether property is new or not.

House prices are slightly seasonal - that is, prices are higher at certain times of year irrespective of the overall trend. This tends to be in spring and summer, when more buyers are in the market and hence sellers do not need to discount prices so heavily, in order to achieve a sale. Therefore both the Halifax and Nationwide seasonally adjust their indices.

What this all hopefully means is that any changes in the index are simply due to price fluctuations and not due to changes in the time of year, the size or the balance of the sample.

Comments: There is little to say about the Nationwide index that differs from the comments about the Halifax index. They use similar, but not identical methods and base their ideal sample on different criteria that should really eliminate most of the variations that would normally occur as a result of short term variations in lending patterns, or stronger levels of business in different regions of the country. However, the fact that large differences do occur shows that at least one of the indices, if not both, are vulnerable to some of the factors. Like the Halifax index, that of the Nationwide is best used as a rolling measurement of change in the housing market, as opposed to an absolute guide to house prices or as a direct comparison with other measures.

Link: http://www.landreg.gov.uk/publications/default.asp

Source: The Land Registry compiles its figures from information sent by conveyancers and solicitors on completed property sales, with every property sale in England and Wales is recorded by HM Land Registry, currently around 100,000 a month. The sales included in the report relate to the transfer for full market value of the ownership of freehold or long leasehold properties, whether or not the purchase was supported by a mortgage.

Methodology: The prices shown are actual, unadjusted averages drawn from the great majority of all residential sales in England and Wales completed during the period.

Comments: The land registry figures are by far and away the most comprehensive and most accurate for any given period of time. However, there are a number of drawbacks, especially when comparing the reports with results from a previous time period:

• Timeliness - Given that the Land Registry is measuring completions, rather than valuations, its figures actually reflect prices that are agreed anything from one to four months prior to the reporting period. Although it is far more accurate than any other data, it is also well out of date by the time it comes out.
  
  
• When looking at the breakdown into the average price for different property types at a local level, there can be large variations in price from one month to the next. This is because the values are not adjusted to take account of varying proportions of sales in different price categories, so there can be statistical anomalies due to occasional small sample sizes.
  
  
• The Land Registry figures report only on data for England and Wales, with Scotland and Northern Ireland excluded from the information. This means that average prices are generally higher than in the Halifax or Nationwide reports, as prices in those areas tend to be lower than the average price in England.

Link: http://www.hometrack.co.uk

Source: Hometrack sources its information from around 4,000 estate agents nationwide, with data updated every month.

Methodology: The hometrack index is compiled right down to postcode level, and includes information on house prices for different types of property, number of sales, growth in buyers and the number registered properties for sale.

Comments: The main downside is that the information is based on the subjective opinion of the agents in the selective sample who are not necessarily always the same from month to month. But although the hometrack index is neither the most comprehensive nor necessarily the most accurate, it is one of the best measures of up-to-date price movements at a very local level, which lets face it, is what most homebuyers and owners are interested in. It also offers some useful qualitative information that can help assess levels of supply and demand.

Link: http://www.rics.org.uk

Source: The results of the RICS Residential Housing Survey are drawn from a monthly survey of local surveyors offices up and down the country.

Methodology: The monthly report is a qualitative report that aggregates the findings of the different surveyors in ten different regions of the country. The report assesses trends within the market, including the direction (but not the value) of house price changes, the stocks of different types of property on the books of the contributing offices, the volume of completed sales and predictions for changes in the level of house prices for the forthcoming three months.

Comments: The RICS housing survey offers something a little different to the house price information that is available from other sources. The qualitative information helps build up a picture of the market and is useful for complementing data from other sources. Furthermore, the report breaks the information down to look at the different geographical areas in isolation, and also includes dozens of comments from local surveyors in each area.

However, at no point does it try to put quantitative values on the level of house prices, trying instead reporting the number of offices reporting a general increase in prices, the number reporting price falls and the net balance. This obviously means that it is highly risky to use this report as an accurate measure of the state of the housing market. It does little to assess the strength of any rises or falls - instead choosing to build a picture of how widespread the trend is - and does not seem to take into account the size individual offices in terms of the volume of properties they have on their books.

Link: http://www.cml.org.uk

Source: The council of mortgage lenders draw their information from the mortgage completion values reported in the Department of the Environment, Transport and the Regions (DTLR) and CML, Survey of Mortgage Lenders. The Survey of Mortgage Lenders (SML) is a continuous monthly survey produced jointly for the Council of Mortgage Lenders (CML) and the Department of the Environment, Transport & the Regions (DETR).

Methodology: The survey is based on 5% of mortgage completions from more than 50 lenders and is grossed up to represent the entire mortgage market. The figures are adjusted to reflect changes in the mix of properties mortgaged but are usually only displayed on a regional basis.

Comments: The CML survey is another useful measure, but the potential inaccuracies caused by grossing up, the absence of non-mortgage sales, the lack of full market coverage and the absence of any localised geographical breakdown mean that the measure is best used as a an ongoing measure of trends, rather than as an accurate measure.

Link: http://www.marketplace.co.uk

Source: All the research data is taken from the last five transactions from each of the 317 branches of Bradford & Bingley estate agents nationwide.

Methodology: The Home report is published every three months and looks at a variety of market features on a regional level including Scotland. The report opens with an overview of the state of each regional market, followed by a breakdown of the following: the average time to find a buyer by region and property type with a highlight of any local areas that are bucking regional or national trends, selling prices achieved as a proportion of asking prices in each region, the average number of viewings per sale for different property types in each different region, the average ration of buyers to properties for sale in each region under different price bands, a look at the composition of different buying groups in each area and average prices in each region paid by first time buyers.

Comments: The B & B Home report is a very useful survey for understanding key behavioural aspects of the market. It describes various trends in the market that are not available elsewhere and is particularly useful for assessing regional levels of demand, due to the nature of the data.

Link: http://www.home-sale.co.uk

Source: The Housing Market Trends reports are collated using information contributed by the 650 offices that make up the Home Sale Network.

Methodology: The Home Sale Network Housing Market Trends includes a summary of the state of play in each major geographical area, which are the same categories as used in many of the other surveys - East Anglia, East Midlands, Greater London, North West, North, South East, South West, Wales, West Midlands and Yorkshire & Humberside. The survey looks at the proportion of first time buyers among all purchasers, the average number of viewings for each property, the average selling time and the average selling price as a proportion of the asking price.

Comments: While the Home Sale Network represents a decent selection of estate agents and offers some excellent qualitative information, the last report was in July 2001, which is a huge limitation on the usefulness on this data. Comparing different indices is fraught with risk and most of them are best used to look at rolling trends. With such an infrequent level of publication, this is rather too difficult with the Home Sale Network data.

Link: http://www.teamprop.co.uk

Source: The information in the TEAM surveys is drawn from the several hundred member offices.

Methodology: The TEAM property report is a region-by-region breakdown of average "under offer" prices agreed on homes being sold through members of the TEAM network of independent estate agents.

Comments: The TEAM property report is another measure of average house prices on a county level that could be useful as a complement for data from other sources. It is published relatively frequently, though not on a monthly basis, but its lack of detail on the statistical methods used and relatively small sample base mean it is probably less accurate than other sources.

 Print

Copyright (c) 2008 TheMoveChannel. All rights reserved. Terms of Use Privacy Policy

• Developers
• Agents
• Advertisers
• Partners
• Press

Our International Property Portals: Bulgaria • Cyprus • Florida • France • Italy • Portugal • Spain • Turkey