Housing markets, signals and search

Abstract

Research into the spatial structure and functioning of local housing markets typically focuses on market outcomes, particularly house price changes and household movement patterns. Explanatory models are usually based upon a standard neoclassical analysis of the housing market. That approach de-emphasises the importance of imperfect information, real market processes and the signals they generate. The inherent nature of housing means that partly informed households typically engage in search activity prior to purchasing a property. Search is inevitably a spatial process. Housing market search modelling remains relatively undeveloped. However, analysis of this process can provide important additional insights to both better explain consumer behaviour and support more informed decision-making by housing planners and market providers. We illustrate these arguments using housing search data for Scotland.

Keywords:

• market signals
• housing market areas
• models of search
• spatial search patterns
• planning

Acknowledgement

Thanks are due to David Watkins for providing figure 1.

Notes

1. Earlier work by Maclennan and Wood (1982) points out that some households may miss some stages and that failed search leads to a recursive adjustment process.

2. This stands in marked contrast to the Austrian perspective that saw the knowledge properties of market signals, and not just prices, as being the raison d’etre for both analysing and preferring markets (Maclennan, 2012)

3. Survey generated price information was compared with price data from the Register of Sasines and the map-based Land Register of Scotland that is slowly replacing it – these being the most authoritative sources of residential property transaction information for Scotland. The aggregate sample price data matched well with the Sasines/Land Register (S/LR) price profile of new property purchasers, although in a small number of Districts sample prices differed significantly from mean S/LR prices at the District level.