Skip to section navigationSkip to content Commonwealth of Australia Coat of Arms Parliament of Australia - Parliamentary Library
HomeSenateHouse of RepresentativesLive BroadcastingThis Week in Parliament FindFrequently asked questionsContact

Research Note Index

Research Note no. 7 2006–07

House prices

Tony Kryger
Statistics Section
9 October 2006

House prices in Australia have risen substantially over the past 20 years, far outpacing the growth in inflation, average earnings and household income.

Nationally, there are four main measures of established house prices—those produced by the Australian Bureau of Statistics (ABS), the Real Estate Institute of Australia (REIA), the Housing Industry Association / Commonwealth Bank (HIA/CBA) and Australian Property Monitors (APM).(1) The purpose of this Research Note is to examine some of the issues surrounding each of these house price measures and to look at the trend in house prices over the past 20 years.

Existing house price measures

The REIA is the longest running house price series, providing quarterly data on median prices for established houses sold in each of the eight capital cities. Data are collected from a variety of sources—APM in NSW, Residex in Queensland, a survey of real estate agents in Victoria and Tasmania, and state land titles offices in the remaining states.  Prior to September quarter 2004, prices were based on the settlement date of sales. That has since changed to the date of contract exchange, ensuring that the pricing point of the series more closely approximates developments in the market.

The ABS produces a quarterly index of established house prices in each capital city. Sales prices are determined using Valuers’ General data combined with data from home mortgage lenders. The ABS improved the timeliness of its data when, from March quarter 2002, the pricing point of its series was changed from the date of settlement to the date of contract exchange.(2)

The HIA/CBA series provides median ‘first home buyer’ prices(3) for established dwellings (houses and units) for each capital city and rest of state region. Price data are collected by means of a census of CBA home loan approvals during the quarter. A major advantage of HIA/CBA price data is that they are dated to when the CBA agreed to provide funds for purchase. A disadvantage is that although the figures come from a large national lender, the sample may not be representative of all transactions in a period.(4)

APM, a property research firm, produces quarterly median house price data for the eight capital cities, based on Valuers’ General data as well as agent reported sales data. The series is based on property prices exchanged in the quarter.(5) This Research Note does not incorporate any data from APM as data from this source are only provided on a fee for service basis.

All four house price measures have the disadvantage that they fail to adjust for improvement in the average quality of houses over time. The quality of the national housing stock has improved substantially as houses have increased in size and as the existing housing stock has undergone renovations and extensions. These improvements result in higher prices that are the result of quality changes rather than pure price increases.

Another problem faced by the various house price measures is that the mix of houses (size, quality and geographic location) traded in any period may be very different from the mix of houses traded in other periods and from the much larger stock of dwellings not traded and for which no price data are available.(6) Left unadjusted, price movement from one period to the next will partly reflect changes in the mix of properties that are being traded, as well as pure price changes. To minimise this influence, the ABS uses location (suburb) to group together houses that are similar in terms of certain price determining characteristics. Those characteristics are the social and economic condition, percentage of three bedroom houses and geographic location of the suburb.(7)  A broad based measure of price change is then calculated by weighting the change in the median prices for each group.

The approach adopted by APM to the problem of compositional change is to allocate a city’s suburbs into various groups ordered from the most to least expensive suburb. Typically, there are ten such groups of suburbs for each capital city. A median price series is then compiled for each group. By weighting the median price of each group it is then possible to come up with a composition adjusted median price for the capital city as a whole.(8)

The trend in house prices

Bearing in mind the differences and problems that exist with each of the house price measures, established house prices have more than quadrupled in the past 20 years (Table 1)(9). During this time there have been two periods when prices rose particularly fast. The first was the two–year period from 1987 to 1989, when house prices (based on ABS data) rose by 56 per cent. The second was the seven–year period from 1997 to 2004, when prices rose by 108 per cent. The second period differs from the first in that it was mainly characterised by strong, but less extreme rates of increase, sustained over a long period.

The increase in house prices in Australia has far outpaced the general rate of inflation (Chart 1). On an annual average basis, house prices over the past 20 years have increased by 7.5 per cent (ABS measure) compared with an increase of 3.6 per cent in consumer prices. Over the same period, REIA house prices have increased on an annual average basis by 8.3 per cent, or an increase in ‘real’ terms (i.e. after adjustment for inflation) of 4.5 per cent.

Another way of highlighting the increase in ‘real’ house prices is to express prices relative to adult earnings. In 1986, 3.6 years of adult full-time earnings were needed to purchase a house (using REIA house prices), increasing to 7.0 years in 2006. However, because households often consist of two income earners, and because house purchases are made from after tax income, a better indicator is the ratio of house prices to household disposable income. On this basis, the number of years of household disposable income needed to purchase a house has increased from 2.5 to 5.4  between 1986 and 2006. The current figure is somewhat down on the figure for 2004 when a peak of 5.7 years of household disposable income were needed to purchase a house. (Table 1 and Chart 2.)

Established house prices (which include land) have increased significantly over the last couple of decades. However, project house prices (which exclude land) and the cost of materials used in house building, have experienced increases more in line with the general rate of inflation (Table 2). The implication is that an important component of the increase in established house prices is the higher cost of land.(10) The change in the quality of housing is another factor.

Conclusion

House prices in Australia have increased substantially over the past couple of decades, both in actual and ‘real’ terms. However, it needs to be remembered that there are a number of different house price measures and each is not without its limitations. Importantly, none of the series measures only pure price change, the movement in house prices from one period to the next being a combination of pure price increase, quality improvement and compositional change. Consequently, all house price series need to be treated with caution.

  1. Another provider of house price information is Residex, a property research firm. Residex compiles ‘repeat sales’ indices for Sydney, Melbourne and Brisbane only, based on data received from state land titles offices. The series goes back to 1978 and attempts to control for compositional shifts in the stock of houses sold by measuring only price changes between successive sales of the same property.
  2. Australian Bureau of Statistics, House Price Indexes: Eight Capital Cities, September Quarter 2005 (Cat. No. 6416.0), pp. 2, 12.
  3. The CBA/HIA makes an adjustment to the median price of established dwellings to approximate ‘first home buyer’ prices. This adjustment was made necessary by a break in the series in the December quarter 1987 after which separate information on loans approved to first home buyers was no longer available.
  4. Reserve Bank of Australia, ‘Measuring Housing Prices’, Reserve Bank of Australia Bulletin, July 2004, p. 3.
  5. Australian Property Monitors, New housing price series reveals a flat market, media release, 8 August 2005.
  6. Reserve bank of Australia, ‘Measuring House Prices: An Update’, Reserve Bank of Australia Bulletin, June 2006, p. 2.
  7. Australian Bureau of Statistics, Renovating the Established House Price Index, Australia, November 2005 (Cat. No. 6417.0), pp. 7–8.
  8. Australian Property Monitors, loc. cit.
  9. A 20 year reference period has been chosen as it is the longest period for which data is available for all  three house price measures. REIA data has been available from the late 1970s, HIA/CBA data from 1984, and ABS data from 1986.
  10. A. Moran, ‘Prices and Planning: The State of the Housing Industry’ accessed on 13 September 2006.

For copyright reasons some linked items are only available to members of Parliament.

top