• No results found

The term hedonic is used to describe ―the weighting of the relative importance of various components among others in constructing an index of usefulness and desirability‖ (Goodman, 1998: 292). Rosen (1974: 34) defines hedonic prices as ―the implicit prices of attributes and are revealed to economic agents from observed prices of differentiated products and the specific amounts of characteristics associated with them‖ (Ustaoğlu, 2003). Rosen (1974), comprehensively laid down a theoretical foundation for determining the bid price or implicit value of the attributes of a commodity for different consumers. The bid price (φ) is defined as the maximum amount of money which a consumer is willing to pay for a good under the condition that he or she retains a specific level of happiness or utility. He proposed to utilise the information from the tangent of the market price curve with which the consumers or producers share the same value of the equilibrium conditions. The methods used to identify the consumer‘s bid price function and the producer‘s offer function (o) was fully discussed by him. The offer function is defined as a function to determine the minimum value of price which a producer should accept to sell a good for a certain profit. The relationship among market price, bid price and offer functions are shown in Figure 4.2 (Hidano, 2002: 10).

Diagram.

92

Hedonic Pricing Function Fig.4.2:

Hedonic methodology is mainly used for market valuation of goods for their utility bearing characteristics. The Hedonic Housing Price Model is a powerful econometric tool for capturing important determinants of housing values. The lack of theoretical foundations in hedonic price theory was overcome by Lancaster (1966) who states that a commodity can be decomposed into a bundle of attributes. The correct interpretation of these hedonic functions was widely misunderstood until the work of Rosen (1974) . The goods under consideration embody varying amounts of attributes and are differentiated by the particular attribute composition they possess. In most cases, the attributes themselves are not explicitly traded, so that one cannot observe the prices of these attributes directly. In such a case, hedonic pricing models are very essential in order to determine how the price of a unit of commodity varies with the set of attributes it possesses. If the prices of these attributes are known, or can be estimated and the attribute composition of a particular differentiated good is also known, hedonic methodology will provide a framework for value estimation (Ustaoğlu, 2003).

As stated above, the theory of hedonic price functions provides a framework for the analysis of differentiated goods like housing units, whose individual features do not have

 

, ,

P Z  

 

P Z

0

93

observable market prices. The traditional use of hedonic estimation in housing studies has been for the purpose of making inferences about non-observable values of different attributes like air quality, airport noise, commuter access (railway, subway or highway) and neighbourhood amenities (Janssen,VanVliet,Aart,Harssema and Brunekreef,2001). Over the past three decades, the hedonic-based regression approach has been utilised extensively in the housing market literature to investigate the relationship between house prices and housing characteristics. The primary reasons for such extensive application are for analysing household demand for these characteristics as well as constructing housing price indices (see, for example, Can, 1992; Sheppard, 1999).

Residential housing is an important aspect of quality of life in any community.

Therefore, appropriate valuation of specific characteristics of a residential house is in order.

To achieve this objective, empirical researchers often specify hedonic price functions or hedonic models (Ogwang and Wang, 2003).

4.5.1 Empirical Studies on Hedonic Pricing Model6

Hedonic price theory has found useful applications in the housing market right from time of Ridker and Henning (1976) who analysed the effect of air pollution on housing prices.

Following this study, a number of empirical studies appeared in the hedonic price literature regarding the housing market. A brief list of some selected findings is presented in the appendix after giving consideration to the review of certain empirical studies.

Follain and Jimenez (1985) used data from a household survey similar to PNAD for five cities of Colombia, Korea and Philippines. They have used rent as a proxy for property value. From the estimates of the household willingness to pay for the property attributes, the authors estimated the optimum size and characteristics of the properties addressed to low income population that would maximise the producers‘ profit and the consumers‘ utility. In this way, it could be possible to estimate which housing programme would be most suitable for the low income population, at the minimum cost to the government and still respecting the consumers‘ preferences for the various attributes of the property. Hence, such methodology could be permitted researchers to answer the following question: given a certain cost and a target group previously defined for an urban policy, what would be the best project, in the sense of maximising the social welfare derived from that policy?

6 Summaries of some earlier studies were provided in the Appendix

94

Figueroa (1993) has estimated a hedonic prices function for Paraguay, by Iterative Least Squares. The advantage of this method is that it avoids the use of ML to estimate an optimum functional form, so that the Box-Cox transformation could be derived by OLS. To find the optimum λ, several values were imputed and equations were estimated by OLS for each specification, choosing those with the smallest Sum of Squared Errors (SSE). Such methodology permits obtaining the optimum functional form, without the need of using non-linear methods like ML. The data comes from a survey carried out especially for this study in housing programmes implemented by the Paraguayan government. After estimating the equations, the author analysed the social impact of housing programmes for low-income populations. The property price was obtained by questioning the owner directly, in order to evaluate his house. According to the author, this would allow the capturing the number of people who are willing to pay for their properties. Figueroa has demonstrated how urban infrastructure policies affect the property selling price, and consequently, the families‘

patrimony. He has also shown how hedonic models can be used to estimate some of the positive externalities of urban infrastructure policies, such as the increase in families‘ wealth and living conditions.

Santos et. al. (1999), have applied the hedonic prices model to the RMs of Recife, Curitiba and Brasília, using data from PNAD/97. They have used a log-linear model and the OLS technique for the estimation of the regressions for each RM, separating families per income levels. The great contribution of this study was trying to explain the families willingness to pay for housing services, taking into account their income level, with an emphasis on governmental housing programmes (families with monthly income below Brazilian minimum wages). However, their results can be biased, once the data was censured a priori, because of the partition of the sample by income strata.

Aguirre and Macedo (1996), have estimated a hedonic function for Belo Horizonte (Minas Gerais), using Box-Cox transformation and data from the Institute of Economic, Administrative and Accounting Researches of Minas Gerais (IPEAd). The results were obtained by Ordinary Least Square (OLS), Maximum Likelihood (ML) and non transformed data. The sample is limited to information on flats, with an average size of 120 m2. However, some of their findings indicate a possible bias in the sample because the presence of a garage was not significant to increase the property price. This is probably due to the fact that flats with 120 m2 are targeted at higher income groups, who require a priori the existence of a garage in the property. Perhaps the inclusion of an extra parking space would be more important to explain the variation in flat prices than the existence of a garage in the building.

95

Selim (2008) used Hedonic model to examine the effect of characteristics of goods on their prices. Factors that determine the house prices in Turkey are analysed in this paper using 2004 Household Budget Survey Data. The most important variables that affect house rents are type of house, type of building, number of rooms, size and other structural characteristics such as water system, pool, natural gas.

Some studies measure the benefits of air-based amenities (Harrison and Rubinfeld, 1978; Nelson, 1978; Graves et. al., 1988); others measure the benefits of water-based amenities (Brown and Pollakowski, 1977; Lansford and Jones, 1995; Epp and Al-Ani, 1979;

Young, 1984; Milon, Gressel, and Mulkey, 1984; Wilman, 1981). All these studies apply the hedonic price model, which assumes that a continuous function relates the price of a house to its attributes — the hedonic price function — and that people select a house by equating the marginal utility of each house attribute to its marginal price (Rosen, 1974).

96

CHAPTER FIVE

Theoretical Framework and Methodology 5.1 Introduction:

This chapter presents the study‘s theoretical framework, model specification and methodology under which the estimation technique and the source of data used were also discussed.