Automated property valuations

Each website offers a range of reports, including individual property, suburb and postcode profiles. Free sample reports are available on each of the websites, so you can see what you're likely to get.

Single property reports include varying amounts of information, including:

• Price estimate or valuation, usually with a big disclaimer. The estimates are based on past sale prices, information from estate agents and statistics.
• Price ranges.
• Sales history.
• Recent sales in the area.
• Property attributes, such as the number of bedrooms, area size, the presence of a car space or pool, nearby amenities.
• Photographs.

New initiatives on the way

On 1 March 2008 Residex plans to launch a website that will allow the public, vendors and agents to add their opinions on its estimated property prices.

RP Data plans to offer more accurate electronic valuation reports, a hybrid between an automated valuation model and a physical valuation, using both the automated data and a licensed valuer. Reports are expected to be available in about five hours for less than $200.

$140,000 difference

As part of the research for this article, Rachel and Paul, who were preparing to bid for a Sydney house at auction, bought property reports from four websites. They also arranged an independent valuation.

While not an estimate of which websites would be most accurate in a large sample, the exercise did prove that there are risks in relying on a computer-generated valuation.

• The online estimates obtained for the same three-bedroom house differed by over $110,000 — ranging from $548,000 to $662,661.
• The least accurate website was over $140,000 off the eventual sale price.
• Some websites gave ranges so wide that they were of little use: for example, $543,000 to $782,000, or $484,800 to $706,000.
• The licensed valuer was somewhere in between, putting a price of $630,000 on the property. (See the table below for more details.)

With such wide differences, the couple felt they couldn’t rely on the reports, and they weren’t sure which to believe. They eventually used the reports plus information from the selling agent to derive their own valuation.

"Our own estimate was based on previous sales in the area and the price estimation in the Residex report, which seemed the most comprehensive," Rachel says. "We added $25,000 for improvements that had been made since the property’s last sale, so we reckoned its market value was about $621,000. That was pretty close to the licensed valuation report of $630,000."

However, the price the property fetched at auction — $690,000 — exceeded everyone’s expectations. "The unknown factor is what someone else is prepared to pay," says Rachel. "The highest bidder might have been looking for two years and be willing to pay above market value, or might just have loads of money."

Although all the automated reports undervalued the property relative to its sale price, Rachel still thinks some are worthwhile for buyers.

"$50 is a small price to pay to be informed about the market," she says. "But some reports are better than others. The least useful ones were those that gave little information about comparable sales in the area or the property’s historical sale prices. The reports helped up to a point — but they don't take into account things like recent renovations, and the prices are really only a guide."

Table: e-Valuations compared

Internet estimations and valution for Sydney property, sold for $690,000 at auction two weeks later.
Company name	Australian Property Monitors	Propertyvalue.com.au	Residex	RP Data	Licensed valuer
Report name	Property Report	Individual Property Report	Property Explorer Report	RP Estimate	Certificate of valuation
Report price ($)	59.95	39.95	65.00	79.95	550.00
How is the estimate described?	Market price estimate	Best estimate or estimated value	Initial estimate	Unassisted valuation result / estimated value	Current market valuation
Range quoted ($)	Not given	496,415 to 606,729	484,800 to 706,000	543,158 to 782,163	Not given
Valuation or estimate ($)	548,000	551,572	596,100	662,661	630,000
Difference from licensed valuation ($)	–82,000	–78,428	–33,900	32,661
Difference from licensed valuation (%)	–13.0	–12.4	–5.4	5.2
Difference from sale price ($)	–142,000	–138,428	–93,900	–27,339	–60,000
Difference from sale price (%)	–20.6	–20.1	–13.6	–4.0	–8.7
Number of pages in report	4	4	17	2	8
Suburb median ($)	540,000	594,000	592,500	Not given	Not given

"While Automated Valuation Models (AVMs) may be useful for lending institutions in a portfolio and mortgage situation, I have grave reservations about their general use by consumers."
So says Peter Rossini, Program Director of the University of South Australia’s School of Commerce. Peter has published several research papers and studies of automated property valuation models.

The competing websites get most of their data from the same sources — state government property sales registers. They supplement this with information about recent sales from their estate agent contacts, and details of properties’ attributes. Complex statistical modelling of various types (including hedonics, which takes into account individual property attributes, such as the number of bedrooms and presence of a parking space) is then used to generate estimations and values. Each model uses its data differently, putting more or less weight on particular components.

However, each company knows that only a portion of the automated estimates it sells is within the right range.

Residex, for example, says estimates will generally “only be in an acceptably tight range in around 70% plus of the cases regardless of who undertakes the calculation”. If correct, that means around 30% of estimations from all companies are off the mark.

Residex concedes that “where there is the potential for such significant errors it is possible to unintentionally mislead the public.” For that reason, the company’s reports provide “rules and test values for the users to confirm and identify a value from and to help the user become more familiar with the valuation process”.

RP Data points out the limitations too. “We don’t want people thinking they can rely on an automated valuation,” it says.

Research in the UK, where automated valuations are also used, acknowledged the overall good performance of the tested models, but found that low-value properties are more likely to be overvalued, and high-value properties are more likely to be undervalued. It also found more variation (errors) in less homogenous segments of the property market, such as regional or rural areas, or suburbs where properties aren’t similarly designed in a uniform pattern — which is a much more common situation here than there.

In the UK, automated systems provide an indication of confidence in the accuracy of the valuation (sometimes referred to as confidence levels or uncertainty levels). In Australia, however, only RP Data reports give an indication of the accuracy you can expect, which varies from suburb to suburb. Its ‘Forecast Standard Deviation’ (FSD) is a 68% probability that the report’s estimated value and the ‘true valuation’ of the property are within a certain range. An FSD of 15, for example, should mean there’s a 68% chance of the estimated value being within plus or minus 15% of the true value.

Australian Property Monitors claims that on average, half its reports are within 10% of the property’s sale price, and that 80% are within 20% of the actual value. Residex doesn’t display its confidence level for individual property estimations to the public, only to banks.

So property websites recognise their own limitations, and legal disclaimers point out that consumers shouldn’t rely solely on their figures. “Our reports are a guide or tool, but you need to get a valuation later,” says Residex. “A valuation is much more detailed and accurate than our Property Explorer reports could ever be 100% of the time.”

Australian Property Monitors echoes those comments. "Our reports should be used as part of a suite of research undertaken when buying a property. Customers shouldn't rely on our reports in isolation for financial decisions. Buyers also need building inspections and other due diligence. We have specific disclaimers to protect us because we want customers to use the report as an estimate not a valuation. Only a licensed valuer can give a valuation."

Limitations of e-Valuations

These websites use sophisticated statistical analysis, huge databases and contacts in the industry. But there are many things automated valuations and estimations, based purely on statistics, can’t do.

• There’s no substitute for physically examining a property. The stats don’t take into account renovations and improvements that’ve been made, or if the property’s fallen into disrepair. There’s no inherent reason why this information couldn’t be collected – it’s just not happening at present.
• Automated systems remove human subjectivity and bias – a possible advantage of automated valuations over physical valuations. However, as anyone in the property market knows, emotion and personal bias are often key determinants of a selling price.
• Computer-generated figures are slower to adjust for an under- or oversupply of properties in a particular suburb. These supply and demand factors also affect prices.

Industry criticism

As you’d expect, licensed valuers are often critical. The Australian Valuers’ Institute (AVI) says errors in value from automated reports are common and opposes their use of the term 'valuation'.

"In general, the people behind these websites and statistics aren’t qualified to give a valuation report based on statistics,” says the AVI. "A property valuer needs to look at much more than statistics to form a rationale and opinion to provide a valuation assessment. Unless you inspect a property internally and externally, analyse the market, the location and other prudent things that a qualified valuer would do, you can’t get a reliable valuation or figure."

Herron Todd White, a national valuation company, said automated reports can be useful for banks valuing their portfolios, as the overvaluing and undervaluing of individual properties should even out over a large sample of properties.

"But for an individual purchase or seller of an individual property, these automated calculations are next to useless, and in fact could do more harm than good in determining the true market value," the company says.

"If you’re spending hundreds of thousands of dollars on a property, spending an extra $400 for a trained and qualified valuer to value the property properly could save you thousands."