Conference Agenda

Discrete choice experiments (DCEs) are widely used to estimate willingness-to-pay (WTP) for environmental attributes, but derived welfare measures may be sensitive to experimental design when preferences are uncertain and therefore dependent on contextual cues. A prominent concern is the cost vector effect, whereby WTP increases with the range of the price attribute presented in choice tasks. We hypothesize that familiarity with the valuation context mitigates this bias by stabilizing preferences. To test this, we designed a 2 × 2 split-sample DCE valuing recreational access to temperate beaches with varying ecosystem condition, reef type, congestion, and travel time. An online sample of respondents was stratified into experienced users (divers/snorkelers) and non-experienced users (occasional beach visitors) and randomly assigned to one of two price vectors. After excluding protesters and speeders, data were analysed using an error-component mixed logit model in WTP space, estimated via weighted maximum likelihood to reflect individual exposure to coastal recreation. Results show strong evidence of a cost vector effect among non-experienced users, whose WTP estimates are systematically and significantly higher under the high-price treatment for most attributes, especially for high ecosystem quality, natural reefs, and low congestion. In contrast, experienced users’ WTP is statistically stable across price vectors for all attributes, indicating robust preferences. Alternative sorting of sub-groups and complementary models accounting for attribute non-attendance and income effects helped us to interpret results and test their robustness. The study concludes that welfare estimates from experienced respondents may be more reliable for complex environmental amenities and provides useful recommendations for experimental design.