2008 Volume 1 Issue 2 Pages 64-75
The purpose of the study is to perform a comparison of marginal willingness to pay for drinking water quality using the averting expenditures method and choice experiment. The averting expenditures method can reveal non-market values based on the actual averting actions of citizens. Choice experiment is based on choices made by the respondents from a hypothetical choice set. Data was collected from a drop-off survey of residents of Tsukuba, and 220 samples were collected in total. Estimated models of averting expenditures and choice experiments were formulated as the random utility theory, and both methods were analyzed using a conditional logit model and a mixed logit model. Models combined SP-RP data were analyzed using a nested logit model. The estimated coefficient of the residual chlorine using averting expenditures was negative, although that of choice experiments was positive. SP-RP combination models estimated with a nested logit model obtained the highest goodness-of-fit. Estimated marginai willingness to pay to reduce chlorine was 23.5 yen for the averting expenditures method, and 11.5 yen for choice experiments. Marginal WTP for a 1% reduction in THMs was 28.4 yen for the averting expenditures method, and 12.5 yen for choice experiments. The Iindings of this study suggest that the averting expenditures may generate relatively higher benefit estimates than the choice experiments due to the complexity in estimating actual averting expenditures.
