Accounting for substitution and spatial heterogeneity in a labeled choice experiment

Abstract

Many environmental valuation studies using stated preferences techniques are single-site studies that ignore essential spatial aspects, including possible substitution effects. In this paper substitution effects are captured explicitly in the design of a labeled choice experiment and the inclusion of different distance variables in the choice model specification. We test the effect of spatial heterogeneity on welfare estimates and transfer errors for minor and major river restoration works, and the transferability of river specific utility functions, accounting for key variables such as site visitation, spatial clustering and income. River specific utility functions appear to be transferable, resulting in low transfer errors. However, ignoring spatial heterogeneity increases transfer errors.