4.2 A Neoclassical approach to housing search: Tu and Goldfinch, 1996
Neoclassical models of housing search come in a range of forms and with different emphases. However, one of the key denominators is the conceptualisation of human behaviour in the model. The model for consideration here, by Tu and Goldfinch (1996) has been selected as an example of a NCE approach to human behaviour in the housing search process, it uses a two-stage model. Its theory and model are closely linked to NCE, and the antecedents of the model are lined to the assumptions of human behaviour in the housing market that conform to some of the major tenets of homo economicus.
The amount of information possible and appropriate to acquire is contested. Some models assume that the household has incomplete information at the beginning, but through the search process they obtain perfect information and then are able to maximize their utility (or at least satisfice with complete information). Tu and Goldfinch (1996) adopt this approach in their model in Fig. 4.2. They argue that:
“A buyer’s housing market information level will influence his/her final choice only if the choice is made under imperfect market information. As a result, this choice is probably not his/her maximum utility choice. However, this will not happen very often. The reason is that buying a dwelling is the largest expenditure and probably one of the biggest decisions that a household makes. A household will therefore be very careful in choosing a dwelling. If they cannot find a suitable dwelling they may quit the market. So it is reasonable to assume that each buyer will buy a dwelling only after obtaining full market information. Under this assumption, a buyer’s final choice will not be influenced by his/her market information level.” (Tu and Goldfinch, 1996, PP.519-520)
Tu and Goldfinch (1996) maintain the language of choice in the decision making process, despite the challenge to NCE understandings of the process where utility maximisation essentially removes the deliberation of preference and simplifies it to an information sourcing problem (Latsis, 1972; Ferrari et al, 2011). Tu and Goldfinch translate this approach into a schematic diagram, see Fig X. This approach assumes that the search process does not influence preferences, nor that individual’s have limitations in their abilities to gather information or to process the optimum stopping point based on a rational approach.
Fig 4.2: Decision-making, search and choice: Tu and Goldfinch, 1996
Acquisition of information in the housing search stage is also contested.
Source: Tu and Goldfinch (1996, P.521)
In Tu and Goldfinch’s model the order of properties viewed, indeed whether the area or the type of property chosen first is insignificant as the household will gather complete information about their housing opportunities, and maximise the utility of their purchase in a rational manner. The ‘housing search’ box therefore is a simple black box, the inner mechanisms of which are not of interest. The choice is separated into spatial and non-spatial characteristics (utility functions) in Tu and Goldfinch’s model, and is divided into sub-groups of housing searcher based on life-cycle stage. Preferences are explained in terms of life-cycle stage, and outcomes are modelled accordingly. There is no variation in search behaviour between sub-groups as they are all assumed to obtain perfect knowledge and act rationally with regard to that knowledge. This approach is symptomatic of a mainstream approach. The research agenda therefore is concerned less with the mode of information acquisition or the decision making process, instead it focuses on defining the choice set and preferences. This approach is unlikely to be described as behavioural, and would be on the extreme left of our simple taxonomical spectrum. The remainder of the models discussed offer variations on this model, incrementally developing richer behavioural models.