Contingent Valuation Versus Choice Experiments: Estimating the Benefits of Environmentally Sensitive Areas in Scotland: Comment

Following Boxall et al. 1996, Hanley et al. (1998) compare welfare measures derived from contingent valuation (CV) to those derived from choice experiment (CE). Using the same Gumbel distribution assumption of the unobserved component of indirect utility, they estimate welfare measures that they expect to be the same. However, they fail to acknowledge that the indirect utility specifications underlying the two sets of estimates differ. Hence, they do not compare like with like and the difference in welfare estimates cannot be definitely attributed to the difference in the two stated preference methods employed. This comment argues that their econometric analysis is flawed and does not support many of their concluding remarks. Further, disagreement is expressed with their alleged findings on the potential performance of CV in benefit transfer studies.