Each month, property data expert Greg Dickason answers your questions regarding available property data to help you fast-track your research.
Question: Why is there a discrepancy among data providers with regard to median prices? If they are getting their information from the same source, why are they different? Do I even bother with medians?
Answer: Data providers have different coverage of the market as they invest different amounts in gathering data. Some only have access to the government data, which often has a lag of up to three months from the settlement dates of properties transacting. Other providers get recent sales data from property professionals such as real estate agents, which makes their data more recent and accurate.
But that is only half the story. Even when the provider has full coverage of the market, recent prices are not a good way of tracking value.
Prices are important, but they need to be put into the context of value. A simple example explains.
Say this month we see five sales in a suburb. All the sales are for lovely properties on a hill, all with five bedrooms and great views. The median price achieved is $700,000 for these five properties.
Last month in the same suburb there were six sales, but most of these were for properties in the older section of the suburb. These are mainly three-bedroom houses with one bathroom, and most need some renovation. The median price achieved for these very different properties was $520,000.
If you were looking at the median price for the suburb, you would see a big jump from one month to the next: from $520,000 to $700,000 – a jump of 34%!
Data providers try to correct this ‘volatility’ by calculating a rolling median price over
the previous 12 months, rather than month to month. This is something RP Data provides in its suburb scorecard product and is introducing into its RP Professional product.
But even a rolling median price only partially masks the effect of different types of properties selling in different months.
To get the real trend, the value changes need to be reported. I will extend the example to explain.
If we look at those five properties that sold for a median price of $700,000 and go back to the previous time each sold, we can calculate how they have changed in price over time, at the individual property level. We are no longer comparing different properties; we are comparing the same property and how its price has changed.
Taking into account the different time periods and any renovations or other improvements, we can calculate the change in value for those properties as we are looking at the actual difference in their sales price over time. If the median previous price achieved for those five properties was $650,000, with a median hold period of two years, the value of the properties has increased by 3.8% a year.
Extending this value calculation to all properties, but using this ‘repeat sales’ model and other tested models, establishes the value of each property in a suburb. This takes out the effect of different property types sold in different months changing price – the type of property and what it is worth is also taken into account.
This is what we mean by hedonic valuation. ‘Hedonic’ means we take into account attributes of the property and calculate the value of the property using these attributes. We can apply this at the individual property level or higher levels, such as for suburbs, postcodes and capital cities. This is then a valuable metric often used by banks in both valuation and portfolio risk management.
Question: When I want to buy an individual property, how do I go about valuing it? How reliable are comparable sales with regard to determining the value of a property? Are there better indicators I should be looking at? I want to be able to quickly assess the value of a property before I order a valuation.
Answer: The answer lies in your question. ‘Comparable sales’ are reliable, just as long as they are similar properties with similar attributes. This takes into account the ‘hedonic’ characteristics of the property you are looking to value.
Of course, getting enough comparable sales that are both close enough in distance to the property you want to value and have sold sufficiently recently is the hard bit. RP Data has an ‘AVM’ (‘Automated Valuation Model’) that does this hedonic comparison for you and comes up with a value and a ‘confidence range’ for a property. An alternative is to use a data provider with ‘deep’ data that enables good searching and filtering of results to find good comparable properties.
Of course, once you are very serious about a property, paying a professional valuer to give you a valuation is a wise move. They have spent the time evaluating many properties and have the training and experience to give you the confidence you need for a major investment.
Question: I think I get it now. So, when RP Data reports that values drop, say 20%, does that mean my property has lost value?
Answer: Unfortunately, yes! Luckily, values dropping by 20% is not that common. At RP Data we strive to make the market transparent so that decision-makers can make decisions with as much information as possible. This means values will go up as well as down, and if you follow our RP Data–Rismark Daily Home Value Index you will see that values do indeed go up and down on a daily basis in the property market.
