Introduction from the Editor
Article Type: Editorial From: International Journal of Housing Markets and Analysis, Volume 6, Issue 1
This is a very timely special issue focussed on modelling housing markets. In many regions and countries there has been an improvement in housing market sentiment and investment in recent times. A general observation is that this follows the downswing in the same markets which was partly associated with the global financial crisis. To those who are familiar with the fundamental operation of housing markets (and many other land uses), this upswing both in timing and magnitude is of no surprise. Whilst the common reference to the broader property market in wider society refers to “location, location,location”, property cycles confirm the appropriate catchcry for those who understand and model housing market behaviour is actually “timing, timing,timing”.
Although housing differs from other land uses due to the basic need for shelter and the associated underlying demand, housing also has the same fundamental economic characteristics as other land uses, e.g. office, retail and industrial. This is linked to the relatively long lag between the decision to increase supply of housing and the final release onto the market (between which time the supply/demand relationship may have adversely changed), lack of a central market where buyers and sellers meet and a final product which is generally indivisible. However, the largest hurdle is arguably the lack of research in this area. Whilst substantial resources are provided to conduct research into other land uses, the relatively low institutional investment in housing reduces the amount of research in this area. Despite housing representing the largest single land use with the greatest aggregate capital value, the marketplace predominantly consists of individual families who are trading in the market place and lack the collective need for a high level of research.
This special issue is “timely” as it publishes high level quality research about modelling housing markets from an international perspective. It is clear an improved understanding of how housing markets work will not only assist stakeholders such as government bodies and lenders to plan for the future, it will also reduce a household’s exposure to risk when this asset is typically their single largest investment. Modelling housing markets allows such stakeholders to understand how this asset class operates and facilitates forward-looking projections.
Each of these six papers differ with reference to location and topic. Nevertheless, important lessons can be transferred between regions as each is also operating in a different part of the cycle. The first paper examined residential price indices and examined if they were relevant in their current form. The focus was placed on residential property price indices (RPPI) in the UK with reference to average prices, price trends and market turning points. This robust discussion raised different positive and negative aspects of using RRPI and contributed to the debate in the literature. It was concluded it is unclear which of the current RRPI indices should be used and this contributes to the ongoing discussion in this area. The second paper examined the housing market in Brazil with reference to house price indexes and cyclical behaviour. The timeframe was between 2001-2008 with a detailed examination of 1,487 residential properties and their associated attributes. The authors proposed an alternative model which decomposed the sources of cyclical behaviour and argued it should be possible to filter out real shocks that potentially affect the real estate sector. It was concluded that real shocks and US shocks have an upward effect on the house price index.
The third paper analyses the future demand for retirement housing in England based on modelling behaviour linked to health and housing issues, given there are approximately 560,000 independent-living dwellings for the elderly in England. This topic has become increasingly important when considering the ageing of the “baby boom” generation and the ability to accurately predict demographic trends. The modelling process used data from the 2008-based forecast of household numbers up to 2033, individual age profiles, elderly activity constraints, owner occupation levels, affordability and household types. The study concluded there was a substantial need for an increase for retirement housing in England as the population ages. The fourth paper is also ideally suited to this special issue and focuses on the “unobserved attributes in hedonic house price models” based on the original discussion by Rosen (1974). A review of previous studies in this area is undertaken followed by an examination of housing data in Norway over a ten year period. The findings provided an insight into the use of panel data as well as overcoming problems associated with hedonic modelling in reference to omitted,time-invariant explanatory variables.
The fifth paper examines housing data at the neighbourhood level and applies fuzzy equality. It concluded that Thiessen pologons have a slight superiority with based on OLS regression and this finding has relevance for the use of different clustering techniques for market analysis at the neighbourhood level. The final paper examines housing data in Malta to construct three hedonic indices from 1980 to 2010. The analysis of over 28,000 observations enabled the development of two variants of the constrained hedonic index, as well as an unconstrained hedonic price index.
The six papers in this issue collectively form an important special issue on modelling housing markets and your on-going support and discourse in this area is sought. If you have any comments or are interested in submitting a research paper, please contact the editor at: richard.reed@deakin.edu.au
Richard G. Reed
