UNIVERSITY OF IBADAN
3.2. Theoretical review
3.2.3. Lancaster’s demand theory
An alternative theory to the popular traditional consumer theory can be traced to Strotz (1957) who used utility tree to show that a particular commodity is associated with a particular type of utility. Gorman (1959) considers separation of utility function and its aggregation. Lancaster (1966) extends and harmonises this alternative reasoning into a formal consumer theory based on assumption that consumers derive utility from the properties or
UNIVERSITY OF IBADAN
characteristics of commodities rather than directly from the commodities. The resulting alternative theory, used in this thesis, is what is referred to as Lancaster‟s consumer theory.
This theory assumes that commodities are inputs to the consumers‟ consumption activity which generate collection of characteristics as output. A particular commodity is assumed to have more than one characteristic while many characteristics are shared by more than one commodity. It also assumes that combination of goods may possess different characteristics from individual commodities. Some distinguishing features of the Lancaster‟s consumer theory in comparison to the traditional consumer theory are summarised in Table 10.
Lancaster‟s theory of consumer demand is the dominant theory of the economics of quality. It is probably the most popular theory of consumer choice of quality. The theory was presented in two papers and two books (Lancaster 1966, 1971, 1975, 1979). According to Bowbrick (1994), Lancaster‟s popularity appears to derive from the points that:
- The analysis is presented rigorously, based on assumptions which are usually made explicit.
- The theory is clearly presented.
- The theory uses objective characteristics rather than perceptions and beliefs. This gives the promise of “hard” results. It is also cheaper to work with easily measured characteristics than it is to identify and measure consumer perceptions.
Lancaster‟s theory originated from the observation that traditional demand theory was ignoring highly pertinent and obvious information, the properties of goods and he proposed to concentrate on this aspect. After all, one would expect information on the properties of goods to be more easily obtainable and to be more universal in character, than properties of
UNIVERSITY OF IBADAN
Table 10. Lancaster’s consumer theory versus traditional consumer theory Lancaster’s consumer theory Traditional consumer theory Characteristic determines the relationship
between good. For example, wood will not be a close substitute for bread, since characteristics are dissimilar. And a red Buick will be a close substitute for a gray Buick.
No reason except „tastes‟ why wood and bread should not be close substitutes and why a red Buick should be any closer substitute to gray Buick than wood and bread.
Relationship between goods is frequently intrinsic and objective, and will be observed in many societies under many market conditions.
For example, substitution between butter and margarine.
No reason why close substitutes in one context should be close substitutes in another.
A good may be displaced from the market by new goods or by price changes. A monetary asset may cease to be on the efficiency frontier, and will disappear from the economy.
No presumption that goods will be completely displaced or disappeared from the economy.
The labour-leisure choice may have a marked occupational pattern.
Labour-leisure choice determined solely by individual preferences; no pattern, other than between individuals would be predicted.
An individual is completely unaffected by price changes that leave unchanged the portion of the efficiency frontier on which his choice rests.
An individual is affected by changes in all prices.
Source: Lancaster (1966)
UNIVERSITY OF IBADAN
It is assumed that all characteristics are quantitative and objectively measurable. This is an assumption on how individuals perceive the characteristics as well as on its being objectively measurable. It is assumed that if one has x times as much of a good, one has x times as much of each characteristic. Given quantities of the two goods xj and xk the total amount of the ith characteristic possessed by the goods collection (xj, xk) is the sum of the amounts of the characteristics possessed by xj xk separately (Lancaster, 1971). Lancaster‟s assumptions on preferences simply carry over traditional preference theory applying it to collections of characteristics instead of collections of goods. These assumptions are that of transitivity and completeness, continuity, strict convexity, non-satiation and positively desired characteristics.
Lancaster (op. cit.) argues that it is not possible to base any aggregation on the demand of a single representative consumer, as in traditional theory, for the situation is not approximated by a single customer buying different quantities of a single good at different incomes and prices, rather, it is one of different consumers each buying different goods and possibly changing the goods bought as price and income changes. Accordingly, he argues there should be at least one representative customer for each facet, vertex and edge of the characteristics rays.
This thesis uses the Lancaster‟s theory as formalised for tourism demand by Giacomelli (2006a). To consider the effects of destinations‟ heterogeneity, Giacomelli analyses tourism choice process in the light of Lancaster‟s consumer theory. To consider the effects of risk arising from the experience tourism good quality, the same process was analysed in the light of the expected utility theory of Gravelle and Rees (2004). Most of the authors criticising the neoclassical homogeneity assumption, see in Lancaster‟s demand theory an effective tool to gain useful insights into the tourism choice process.
UNIVERSITY OF IBADAN
characteristics are observed and a regression is run to show the relationship between prices and characteristics. On the basis of this, predictions are made from the price of a product with a certain characteristic. A major advantage of the hedonic approach is that it works directly with easily observable market prices and characteristics. It neither requires observations of individual buyers nor a process of aggregation from the individual buyer to market demand. Hedonic methods have been used in many studies related to price of houses or agricultural land in which only a small proportion of the stock is sold in one year. There is an enormous hidden reservation demand in these markets as most farmers want to keep their land and most house owners want to keep their houses. Hedonic models are generally confined to goods within a single group. They usually refer to units of a good, to a package or to a fixed amount without linearity and additive assumptions.
ii. Rosen’s model
Rosen‟s (1974) model rests on the assumptions that: Goods are valued for their utility-bearing characteristics; characteristics are objective and are perceived in the same way by all consumers though they may be valued differently; goods in a group are defined uniquely by their characteristic mix; goods and characteristics are valued positively; characteristics are positively priced; linearity is assumed but not additivity. Broadly speaking, Lancaster is primarily concerned with comparison of individuals‟ demand, market demand being mentioned but not analysed, while Rosen‟s (op. cit.) main objective is to work from assumptions about individual producers and consumers to market prices. The argument is that the consumer attempts to maximise his/her total utility subject to a budget constraint and does this when the ratios between the marginal utilities from any pair of characteristics are equal to the ratios between their marginal prices. Rosen (op. cit.) develops a bid function, resembling an indifference curve, showing the price the consumer is willing to pay for alternative bundles of characteristics. A similar supply model is constructed, with an offer curve joining combinations of characteristics that a given producer will supply for a given sum of money.
Where the offer curve is tangential to the bid curve a bargain is struck. From the offer prices of different producers and the bid price of different consumers an equilibrium price is reached for each characteristic mix. Rosen (op. cit.) presents his paper as a structured interpretation of the hedonic method. It is sometimes stated that He provides the theoretical basis for hedonic analysis (Steenkamp 1989, Earl 1986, Ratchford 1979).
UNIVERSITY OF IBADAN
iii. Houthakker-Thiel approach
Thiel (1952) and Houthakker (1952) present an approach which analyses a situation where someone decides first on what group of goods someone wants to buy then what good is within that group. Thiel admits that his analysis is a highly simplified one but sometimes it has a certain degree of reality. The theory uses a homogeneous good within a group of goods.
The buyer decides, first, whether to buy in the group and then which good to buy in this group. Clearly, the result of the analysis depends on whether, for example, butter and margarine are seen as goods in the same group, so Houthakker and Thiel give some attention to what a group is. Thiel explains that characteristics are positively valued and their price is related to the cost of production. By implication, a characteristic in one group of goods is not equivalent to the same characteristics in another, sugar in sweets or in puddings for instance.
iv. Ladd-Zober model
Ladd and Zober (1977), like Lancaster, assume everybody sees the same objective characteristics, that a consumer has a fixed income that can be spent on products and that it is spent to maximize utility. They assume infinitely divisible products. They are concerned with consumers‟ reactions to objective measurable characteristics, not to perceptions or beliefs which may or may not affect sales. In contrast to the Lancaster theory, it is assumed that the consumer consumes products by consuming services (rather than characteristics) provided by the products.
v. Becker’s household production approach
UNIVERSITY OF IBADAN
vi. Dixit-Stiglitz model
Dixit and Stiglitz (1977) tackle a welfare economic problem by stating that the basic issue concerning production in welfare economies is whether a market solution will yield the socially optimum kinds and qualities of commodities. They are mainly concerned with scale economies. They use indifference curves similar to those used by Lancaster, positive costs of characteristics and economies of scale in producing a variant. Unlike Lancaster they use a representative consumer. Lancaster (1990) cites Spence (1976) and Dixit and Stiglitz (1977) as having a single representative consumer who buys some of each product, unlike his own approach. Market demand is then a matter of aggregating this consumer‟s purchases.
vii. Chamberlin’s monopolistic model
Chamberlin‟s original work is readily applicable to a large number of real markets, and Chamberlin‟s (1953) model is a rich analysis of quality problems in the real world, an analysis which has been largely ignored, possibly because it raises points not normally incorporated into generalised mathematical models. The new models, sometimes referred to as neo-Chamberlinean, can only have an application if their assumptions on individual consumer demand and individual producer supply have some realism. In that sense the models must be considered as users of basic consumption theory, rather than theory on its own right. Lancaster (1990) uses product variety in a way quite different from that of Chamberlin, rather like Chamberlin‟s product differentiation.