House price guides prove more like lottery

If you are lucky enough to be the person who reportedly paid £2.5 million for a semidetached Victorian house in south Dublin last December, you might well be wondering how much your investment has appreciated so far this year.

The good news is that it rose in value by £210,000 in the three months to the end of March, according to Sherry FitzGerald, the estate agents, and the bad news is that it fell by £115,000 in the same period, according to First Active, the mortgage lender. Irish Permanent believes it is worth £140,000 more than you paid for it, while the Government - in the form of the Department of the Environment - says it is worth £67,000 less.

Confused? You should be.

The figures for the next three months do not help much. Your semi-detached Victorian house rose by £147,000 between April and June, according to First Active, which published its new index figure this week, but is still only £32,870 more than it cost you.

Sherry FitzGerald's index is a bit more encouraging. The estate agent claims your house has increased by £355,000 since the start of the year. The Institute of Professional Auctioneers and Valuers calculate your gains since January at a more modest £200,000.

The four published house price indices, which started out as a clever marketing tools for mortgage companies and estate agents, are in danger of turning into something of an embarrassment for their sponsors. The failure of all four to agree with one another and the huge gulf between them and the official Government figures has served to undermine them all and forced the Government to re-think the way it tracks the price of houses.

The Department of the Environment publishes a Housing Statistics Bulletin every three months. There is generally a time-lag of about three months between the end of the quarter and publication of the report.

The Department's statistics are based on returns from lending institutions who are required to submit detail of all mortgages. The Department breaks down the figures between new and secondhand homes and along regional lines. It then publishes a simple average figure for all of these categories.

In the years when house prices were more or less static and the increases currently being experiences were merely the stuff of estate agents' dreams, the Government's figures were adequate. A book running to about 50 pages of tables was published every quarter and that was that. The mortgage lenders did their own sums but kept the results to themselves.

All changed in 1998 when Irish Permanent launched its House Price Index. It was one of number of initiatives dreamt up by the company's public relations advisers to boost the bank's profile. It was an inspired idea, capturing the growing obsession with house prices that had taken over much of Irish society.

To ensure that the new index had credibility the Irish Permanent engaged the respected Economic and Social Research Institute to draw up the index and carry out the calculations. The Irish Permanent index is based on mortgages taken out with the bank, which has between 20 and 25 per cent of the market.

The ESRI adopted a radically different approach from the Department and this explains the large gaps between the Irish Permanent Index and the official figures, according to Mr David Duffy of the ESRI.

The simple average approach adopted by the Department means that the figures for any one period can be very easily skewed. If the same number of the same type of houses was sold every quarter, a simple average would be an excellent measure of house prices but that is obviously not the case.

If a significantly larger number of very expensive houses is sold in one three-month period compared to the next, then the average price figure is pushed up. The opposite happens if a greater than normal number of large developments aimed at first-time buyers comes on-stream. "A few Sorrento Road-like properties can make a big difference," explains Mr Duffy.

The ESRI uses a complicated statistical method to try to compensate for variations in both the type and character of houses sold. They take the mix of houses sold in 1996 - the year on which the index is based - as a reference. Adjustments are then made for differences between the 1996 mix and the mix of houses sold in the period under review.

"We try to strip out the effect of the number and character of the houses sold," says Mr Duffy. The technique is based on academic research in the US and in Britain.

The mix-adjusted approach adopted by the ESRI is the main cause of the variation between its index and the Government's figures, according to Mr Duffy. Another factor is that the Irish Permanent index is calculated on the basis of loans paid out rather than loans approved. On average there is a six-week lag between a loan being approved and a cheque being issued.

What the mix-adjusted approach does not explain is the huge difference between the Irish Permanent index and the First Active index, which also uses the mix-adjusted approach. The First Active index, which started last year, was designed by two University College Dublin academics, Mr Joe Durkan and Mr Colm Harmon.

Neither academic was available for comment this week but their colleague at the ESRI put the difference down to the fact that they use another variant of the mix-adjusted technique.

Mr Richard Hoare, the operations director of First Active, stoutly defended the mortgage company's index, which is based on its business data. He acknowledged that the index had been extremely volatile in recent months but said that it could be affected by a number of factors.

"If you have a couple of bank holiday weekends in a month, that will skew the figures because fewer house sales close in those weeks," he explained. He also said that First Active, which has around 16 per cent of the market, gets more of its business through intermediaries such as brokers than the Irish Permanent. "We probably have more first-time buyer business as a result," he said.

There is considerable variation between the two indices regarding house prices in Dublin, which is a reflection of the fact that First Active includes the surrounding counties in its definition of Dublin, says Mr Hoare.

The two other published indices, the IPAV's and Sherry FitzGerald's, adopt a less rigorous approach and are not held out as comprehensive surveys. Both are based on repeat sales figures compiled by estate agents. In Sherry FitzGerald's case, it concentrates on Dublin, while the IPAV uses data from 200 estate agencies across the State.

Last month the Minister for Housing, Mr Bobby Molloy, announced that the Department of the Environment had hired Davy Kelleher McCarthy (DKM), a firm of consultants, to overhaul its house price index. The Minister expressed concern that house buyers were being misled and confused by the various conflicting indices.

DKM is currently designing a mix-adjusted index based on the Department's data, which should be published from the start of next year. It will be based on returns from all the lenders, including both First Active and Irish Permanent, and should be more representative.

Although both Irish Permanent and First Active adjust for the fact that the mix of houses sold will vary from month to month, their indices do not take into account the fundamental difference in the type of business they do, according to Ms Annette Hughes of DKM. It is generally accepted that first-time buyers comprise a bigger proportion of First Active's business, while Irish Permanent deals with more proportionally more second-hand properties than other mortgage companies.

The Government's new index will capture nearly all the house sales in the Republic, but will not include sales that do not involve a mortgage.

Will it agree with any of the existing indices? "I can tell you now that it won't," says Ms Hughes.