Statistics tell some home truths on the plunge in property values
Thursday November 12 2009
THERE are a number of reasons why the perception of what is happening to house prices does not match what is recorded by statistical measures.
Firstly, people may be influenced by the price a nearby house has sold for. However, the price will reflect the characteristics of that house and could differ quite significantly from their own house.
Secondly, house price changes differ across geographic markets. For example, the Dublin market, which is widely reported, has experienced much greater house price falls than much of the rest of the country. Thirdly, price changes differ across sub-markets -- for example, the price of expensive properties has fallen more in recent times and these attract disproportionate media coverage, creating the impression that the rate of decline is the same for all dwellings.
Finally, coverage of the housing market tends to concentrate on large price changes, which are more newsworthy. Smaller changes in house prices tend not to get as much coverage.
The widespread belief that published measures, such as the permanent tsb/ESRI index, are somehow underestimating the true extent of the house price declines may also reflect a change in "base", when switching from calculating a price increase to calculating a price fall.
It is possible that this change in base leads people to misinterpret the current scale of decline relative to the increases reported in previous years. A move in prices from €400,000 to €500,000 represents an increase of 25pc. However, if prices were to fall from €500,000 to €400,000, then this change, which is of equal monetary value, represents a decline of 20pc.
As with many economic statistics, the measurement of house prices is challenging. No two dwellings are identical. It can be difficult to reliably measure a price unless the dwelling is actually sold. Generally, houses are only sold infrequently and so in any given time period we do not observe prices for most houses.
House sales usually are the result of negotiation, and the eventual sale price may differ considerably from the advertised price -- this is particularly the case in times of rapidly rising or falling market prices.
However, it is important to measure house price change as accurately as we can. The value of housing assets is a major component of individual wealth and, as such, would be expected to influence consumer confidence and spending.
House prices have a key role to play in the measurement of the affordability of home ownership, and consequently the standard of living. House price changes also influence the decision to build new houses (supply) as well as the decision to either trade up or down or to decide to become a homeowner (demand).
The permanent tsb/ESRI index is constructed using the internationally recognised approach. The basic premise is that the price of a house is related to its characteristics. It is based on a large database of mortgages related to house sales for the whole country -- not just those reported for the Dublin area, which dominate media coverage.
THE actual index value published is a three-month moving average; this is good statistical practice when seeking to capture a trend. The measure of prices is taken at the time the mortgage is drawn down, ie, when the sale is actually closed. As a result, if prices are changing rapidly, this index lags any measure based on the date of the sale itself.
It is possible that this lag has lengthened as the market has slowed, resulting in the index lagging market changes to a greater extent than previously. Therefore, recovery in the housing market, when it does happen, is likely to pre-date any upturn in the index. This is a feature of the way in which the index is constructed. However, the index continues to be a comprehensive measure based on actual sale prices.
One alternative approach is to use data related to the loan approval stage as the basis for an index, as it would reduce this lag. However, such an approach is not without its own difficulties because loan approvals may not proceed to completion, or may ultimately proceed with another lender. In other words, while approval may be sought for a particular property, the final transaction may be on a different property.
Thus, there are a variety of reasons why perceptions of house price changes differ from the recorded change in surveys. Of course, public acceptance of statistical measures of price change is not helped by the existence of a number of such measures in the Irish market.
Some of the differences between the measures reflect the fact that they are constructed in different ways, use different data sets and measure the price at a different point in the transaction cycle. The permanent tsb/ESRI index does lag the market and so any upturn will not be immediately reflected in the index.
However, the index does have the advantage that it uses the actual price paid for a property to measure changing prices and it takes account systematically of the differences in characteristics of properties transacted in each period.
- David Duffy
Irish Independent
