This study analyzes ways to identify protest zero-bids (PR zero-bids) in the contingent valuation method (CVM) and to effectively deal with PR zero-bids. In order to distinguish PR zero-bids from true zero bids (non-PR zero-bids), this study adopts the concept of the implicit willingness to pay employing the Hicksian compensating surplus and the Taylor's 1st order approximation. When a respondent proposes zero WTP and chooses a PR filtering item to indicate that his/her implicit WTP is not necessary zero, his/her response is identified as a PR zero-bid. PR filtering items falling into the PR zero-bids category include the uncertainty of information, distrust in the government and project achievement, disagreement to project plans, discontent with the fairness of public works and their payment method and animosity against the CVM itself.
The empirical analysis shows that PR zero-bids take place systematically in particular respondent groups; respondents who have never used similar facilities before nor plans to use the facility provided by the public project, the employed, and low income groups.
According to estimation outcomes using CVM data at KDI, the correlation coefficient of the estimated errors associated with PR zero-bids and WTP estimates is 0.9696, which means that the presence of sample selection bias is statistically verified. Also, the analysis finds that there is a gap of about 940 won per household between the estimated WTP excluding PR zero-bids without any adjustment and the estimated WTP with consideration of sample selection bias. Such gap can be significant in altering the decision in regard to economic feasibility of concerned project.
In conclusion, the study suggests that a CVM questionnaire needs to be designed carefully to minimize problems associated with PR zero- bids and the potential risks of having sample selection bias should be concerned. In empirical estimation, it is possible to exclude PR zero- bids when characteristics of a sub-sample excluding PR zero-bidders are identical to those of entire sample including PR zero-bidders. Otherwise, it is necessary to take additional steps to deal with sample selection bias in estimating WTP. In this study, the single-bounded dichotomous choice model is suggested as a possibility.