An Innovative Stated Preference Computer Survey Model for Valuing Noise Impacts from Road Traffic



An Innovative Stated Preference Computer Survey Model for Valuing Noise Impacts from Road Traffic

Authors

ARSENIO E and BRISTOW A L, ITS, University of Leeds, UK

Description

The objectives of the research reported in this paper are to develop an improved methodology to obtain monetary valuations of the noise externalities of road traffic as experienced in the home.

Abstract

The objectives of the research reported in this paper are to develop an improved methodology to obtain monetary valuations of the noise externalities of road traffic as experienced in the home.

The literature on the valuation of the environmental impacts of transport is vast, as a number of studies testify (for example, Maddison et al., 1996; ECMT, 1998; IWW/INFRAS, 2000). Here we wish to explore the methods used to value noise from transport and relevant issues arising from those methods.

Hedonic Pricing is the most commonly used method for valuing noise and has been applied to the valuation of noise from aircraft (Pennington et al., 1990, Levesque 1994, Tomkins et al., 1998) and from road transport (Bateman et al., 2000). This method has been questioned on several accounts, including imperfect knowledge of the attributes of each location and other market imperfections, correlation of explanatory variables, and the difficulty of measuring intangible influences. As eadier noted by Harris (1981), the lack of understanding as to how individuals perceive environmental variables such as "quiet" means that the specification of a hedonic demand function is somewhat arbitrary.

Alternative cost approaches have also been applied to the challenge of noise valuation. One method is to examine the costs of averting behaviour, whether incurred by the individual (e.g., double glazing, behavioural changes) or other bodies (e.g., noise barriers, noise regulations). This approach does not attempt to value the impact on the individuals affected but the costs of mitigating the noise imposed. A related approach, which has been used recently, is damage or consequential cost (Delucchi, 1996; Mayeres et al., 1996; Haling and Cohen, 1996; IWW/INFRAS, 2000), which examines the health, cost implications.

The limitations of these methods led researchers in the field of environmental economics to investigate the potential of a more disaggregate econometric modelling approach centred on individual willingness to pay.

The contingent valuation method (CVM) was the first such approach to be adopted and it has been extensively used for valuing environmental resources (Bateman and Landford 1997; Carson 1999) with limited application to noise valuation (Soguel, 1994; Feitelson et a/., 1996; Maddison and Mourato 1999).

Within this technique an hypothetical scenario is constructed for the purpose of valuation, and respondents are directly asked their willingness to pay for the good (or changes in its levels). CVM studies are usually associated with higher estimates than those derived from hedonic pricing techniques, with Verhoef (1994) noting that CVM estimates can be up to 15 times greater than those derived using hedonic pricing techniques.

The Stated Preference (SP) approach is a generalisation of CVM, where individuals express choices between alternative scenarios characterised by a number of relevant attributes. Concerns about the CVM approach, in particular that it is more susceptible to strategic bias, and that the SP approach more closely resembles everyday decision making, provided a stimulus to use of SP in the valuation of noise (Wardman et al., 1998; Saelensminde, 1999; Garred eta/., 20OO).

The emphasis of this paper is on the use of the SP approach to value noise externalities. The methodological issues that we discuss are:

* innovations in the means of representing changes in noise levels to respondents within the SP experiment

* analysis of the relationship between perceived noise levels and measured noise levels, considering the modifying influence of contextual and personal variables

* a comprehensive comparison of different methods of valuing noise

Each of these issues is considered in turn, in sections 2, 3 and 4 respectively, and the precise methodology used in this study is discussed in section 5. Preliminary results are presented in section 6 with conclusions in section 7.

Publisher

Association for European Transport