FOI Seminar 2 December 2011

Danny Campbell
(Queen's University Belfast):

 

The geography of latent classes: the role of distance decay

(Contact: ) 

 

Friday 2 December 2011, 12.30-13.30, Building A, first floor (the B.S. Jørgensen seminar room)

Abstract:

While stated choice experiments provide useful insight for many stakeholders, the models, on which the results are derived, typically do not adequately account for the inherent spatial dimension that is likely to exist. This is in spite of the fact that it is well recognised within the non-market literature that there is generally a negative relationship between the value people attach to an environmental resource and the distance they live from it. This is widely referred to as the 'distance decay' effect. Using responses to a stated choice experiment that aimed at establishing willingness to pay for a number of rare and endangered fish species in the Lough Melvin Catchment (Ireland), this paper sets out to explore whether the distance that respondents lived from the Catchment had any bearing on the willingness to pay estimates. We begin our analysis using a novel 'distance decay willing to pay space' specification, whereby we estimate a distance decay function for each of the fish species. We then build upon this model to accommodate the 'aspatial' heterogeneity that exists in the willingness to pay estimates. Notwithstanding the advantages that these models are shown to possess over the naive specifications, that do not account for the distance decay, they are based on a stringent deterministic relationship between willingness to pay and distance. For this reason, we develop a more flexible latent class logit model that is capable of 'probabilistically' describing the spatial patterns and variations in the willingness to pay estimates. Specifically, we define the latent class membership to be a function of distance. Subsequently, we are able to derive weighted willingness to pay estimates across the complete spatial extent. In this paper we discuss the importance of accounting for the spatial heterogeneity (attributable to the effect of distance decay) in willingness to pay and the additional insight that these more flexible models offer. We also address the implications for policy appraisal.