Before considering the housing search process in detail it is necessary to recognise that housing is an unusually complex economic commodity. Dwellings possess attributes that rarely apply to other consumption decisions (Quigley, 1979), therefore the standard conditions for economic theories of simple goods may not fit housing, when viewed as a commodity.
“Revealed preference theory implicitly assumes that chosen commodities are well defined and particular commodities are uniform, with few choice relevant characteristics. We shall call such commodities simple commodities. Housing has all the characteristics of a complex commodity, the housing stock is highly varied and this variation occurs with respect to a large number of locational, neighbourhood and internal characteristics.” (Maclennan, 1977, P.112)
The complexity of the search process is partly a result of the complexity of housing as a composite good. A dwelling is a bundle of goods that cannot be easily separated from each other; the housing search process is therefore simultaneously a search combining location, size, style and quality (Van Ham, 2012). Whilst there may be ‘irrational’ behaviours in the market of other consumables (such as second hand cars, see Akerlof, 1970 for one of the key original papers on information asymmetries and economic rationality), these issues combine even more complexity in the housing market. Four key complexities are explored: spatial fixity; product heterogeneity; expense; and joint consumption and investment purchases.
The purchase of a dwelling involves not only ownership of a building but also a specific space of fixed position, size and shape (Meen, 2001). Unlike goods, such as second hand cars, art, or chocolate bars, which can be moved from location to location at the will of the owner, space is a key component of a dwelling (whilst it is possible to relocate the physical structure of a property it is prohibitively expensive in the vast majority of cases, and therefore the product can be considered as spatially fixed). The housing market takes into account not only the physical characteristics of the property, but also the exact location and the space surrounding it. This is important for purchasers motivated by consumption of access to other amenities that the dwelling affords (Galster, 1996). The search for a dwelling amongst all opportunities is therefore also a spatial search, in which locations as well as opportunities are sifted (McPeake, 1998).
Smith, Rosen and Fallis (1988) include distance from employment or cultural locations, the land use types in the neighbourhood and the local government jurisdiction as key spatially fixed attributes of housing. Socioeconomic characteristics of the neighbourhood as well as its physical nature are included in Megbolugbe et al, 1991). Spatial fixity is also important for investment motivated purchases because the space cannot be relocated to take advantage of prices in another location (unlike arbitrage processes in many other markets), or in changes to the neighbourhood (Maclennan, 2012). Attempts have been made to account for spatial fixity in NCE models, although specifying the location variables to be included in a model remains complex (Megbolugbe et al, 1991).
Dwellings are heterogeneous in a strict sense spatially and are frequently heterogeneous in their non-spatial characteristics (Meen, 2001; Tu, 2003). A row of terraced houses may occupy very similar locations (but not exactly the same) and may have similar physical characteristics. However, properties are regularly changed by occupants and micro geography plays a key role in the desirability of a dwelling15. NCE often assumes that housing services are a homogeneous good, or if heterogeneous, the housing characteristics are objectively measurable and any dwelling can be operationalized in an economic model based on their characteristics as a vector (Fallis, 1985). However, in reality all properties are different and heterogeneity prevents dwellings being bought and sold as a standardised good as parties need to gather information about the specific characteristics of the properties available (Smith et al., 1988). The characteristics of the property (size, style etc) allow a household to satisfy a range of household activities in a limited way. The ability for a property to satisfy the desire for household activities introduces the concept of derived demand, whereby the property provides the ability for a subset of desires to be achieved rather than the property in itself satisfying household demands.
“Adopting Lancaster's perspective, housing should not be viewed as being desired as a good per se (a point which immediately casts doubt upon the validity of studies, which estimate elasticities of "housing" demand). Rather, housing can be viewed as a collection of attributes (or characteristics) which in conjunction with the household's consumption technology, are used to satisfy more basic consumption objectives such as shelter, comfort, aesthetics, accessibility, etc. That is, goods are only intermediaries in the consumption process.” (Maclennan et al., 1987, P. 33)
Dwelling heterogeneity is key therefore in considering the satisfaction of desires.
“Supply does not, however, take the form of modules of a homogeneous commodity. Indeed, housing units are enormously heterogeneous. They differ in numerous structural characteristics, lot features, neighbourhood characteristics, local public services and access to desired destinations. Occupants demonstrate the importance of these features through their behaviour. Therefore, housing is a package of many salient attributes, only some of which are under the control of the owner. And this heterogeneity ensures a wide spectrum of degrees of substitutability among dwellings. Housing is not, then, a single commodity but rather a complex of variously related commodities.” (Galster, 1996, P.1798)
Heterogeneity is more significant if the dwelling purchase acts as a communicative aspect of consumption, both reflecting and potentially defining the household’s values (Bourdieu, 2005; Kauko, 2006). The similarity of properties in location and dwelling attributes may be so close that they can be considered functional substitutes for each other and therefore part of the same sub-market (McMaster and Watkins, 1999). However, defining a sub-market in an academic study is not necessarily the same as the consumer’s process of searching for information on the type of property and the areas available. The lack of a uniform product places the emphasis on consumer search behaviour as consumers seek to bring together their household aspirations and the varied characteristics and geographically distributed products available (Maclennan, 1982, also see below).
The purchase of a dwelling is often the largest single expenditure a household has made, particularly for first time buyers. The scale of the purchase is often a magnitude order above everyday purchases, and the household’s annual income. In addition to the price paid for the dwelling, other transaction costs make the venture more expensive including: search, legal and administration, adjustment, finance costs, and the cost of uncertainty (Quigley, 2003).16
“The considerable search warranted by the extreme heterogeneity and immobility of housing, the complex legal and other transactional services, and the household move itself require a heavy outlay of time, effort and money. Both current and capital cost aspects are mingled, and liquidity as well as income constraints are involved. An important consequence is that most households change occupancy infrequently.” (Galster, 1996, P.1798)
Joint consumption and investment
For many households the purchase of a dwelling acts as both a consumption choice, and an investment decision (Rothenberg et al., 1991). The high cost of dwellings emphasises the cost of the purchase and the potential to both make money and lose money on the purchase. Increases in the prices of dwellings also raises the aspirations of households to not only recoup the cost of purchase, but also to sell the property for a higher price than was paid for it. This investment potential suggests that the owner-occupier housing market cannot be considered as simply the market for a consumption good.17
The complexity of housing choice therefore suggests that housing demand is particularly unsuitable for economic assumptions that are designed for small world problems (Ferrari et al, 2011a). A more appropriate economic approach to the decision making process is to consider the choice of a dwelling from a BE perspective that considers the complexity of decision making in light of these characteristics (Marsh and Gibb, 2009).
“Households engage in a search process that involves formulating aspirations, gathering information, learning and revising plans before making a decision to adjust their housing consumption. The process would not be necessary if the assumptions of the neoclassical person were all satisfied.” (Megbolugbe et al., 1991, P. 390)
3.3 The stages of housing search
As discussed in chapter two, economic schools have very different approaches to housing markets and the appropriate subjects of study. Using the above definition of housing search, this section provides an overview of the viewpoint of the schools of economic thought outlined earlier. A more detailed exploration of the variations within BE is provided in the following chapter.
Most theoretical approaches to housing search, which are sensitive to some of the complexity of the decision, break the process down into multiple stages. Rossi’s (1980) seminal book Why Families Move sparked a considerable research agenda in geography of housing and migration patterns. He distinguishes three stages in the housing search process: the decision to leave the old dwelling; the search for a new place; and the choice among alternatives (Rossi, 1980, P.222). McCarthy (1982) likewise uses three stages, although refers to the choice among alternatives as the housing outcome. There are, however, many different views about the number of stages in household’s economic decision making.
“It is well documented that the decision-making process follows a number of stages, although there has been little consensus on the number of stages an individual goes through before making a final choice (Lee and Marshall, 1998). Studies using self-reports have included three stages (Davis and Rigaux, 1974), four stages (Moschis and Mitchell, 1986) and nine stages (Woodside and Motes, 1979).” Levy and Lee, 2004, P.323
Levy and Lee’s (2004) research with estate agents suggested that for the housing market five stages in the decision-making process are discernible: problem recognition, product specification, information search, alternative evaluation, and final choice.
Fig. 3.1: Alternative views of housing choice
(left side represents a linear process through the stages of purchasing a dwelling, the right hand side represents an iterative process which may skip stages and occur non-sequentially)
Marsh and Gibb (2009), expanding on Maclennan and Wood’s (1982) model, suggest that there are six major steps: selection of a search strategy, orientation to an area, establish the possible properties (vacancies), personally visit those properties, evaluate in detail the characteristics of the properties and then form and place a bid. There is a linear progression in this model through the stages, yet this may be an over-simplification of the progress involved in many cases. Figure 3.1 shows two alternative ways of conceptualising the process of housing choice, based upon Marsh and Gibb (2009). The left side suggests that the household progresses through the decisions systematically, narrowing the housing selection until a decision to move or stay is reached, a view typically associated with NCE (e.g. Boehm, 1982). An alternative conceptualisation of housing choice is that it is messy, with previous decisions being reformulated throughout until a decision to move or stay is made.
Whilst NCE has viewed decisions about housing choice as a straightforward progression from motivation to outcome, there is growing evidence to suggest that the decision is rarely experienced as a logically linear process and may look and feel more like a ‘muddling process’ (Park, 1982). Levy et al. (2008) summarise the unstructured approach to housing search:
“It is clear from the interviews that although certain distinct stages can be identified within the decision-making process, these stages may not take place in a structured linear fashion.” (Levy et al., 2008, P.285).
The alternative, non-linear progression through stages of the housing search process, signifies that the purchase of a dwelling is not a “small world” problem. In order to analyse and theorise the housing search process it is necessary to create a linear description and analytical structure. Although this structure necessarily simplifies the process it is possible to understand the stages as iterative and recursive, whilst potentially moving towards the goal of purchasing a home to love in. Figure 3.2 provides a conceptualisation of the different stages of the search process, which is linear but includes feedback between each stage. Thus it is possible to conceptualise a movement in the search process from left to right whilst recognising that the search may stop at any stage, and that stages may in essence run concurrently (e.g. search and visiting properties).
NB: Different housing search stages are in red text and the recursive feedback links are black dotted lines
In the text that follows the stages are considered in a sequential manner, however this is for analytical purposes and it should be remembered throughout that in practice search processes may be non-linear. Each stage is described and a review of key research is provided. A very brief illustration of the types of view from the five different schools (OBE and NBE are separated) allows some comparison of the approaches.
3.3.1 Household Formation
The decision to form a new household is not always considered as a key stage in the housing search process as many household moves are entire households relocating. However there are concealed households moving out (e.g. adult children moving out of the parental dwelling) and divisions of existing households (e.g. separation). The relationship between household growth determined housing demand and supply is a two-way relationship as new supply opportunities may facilitate new household formation (Mulder, 2006).