Abstract
Port expansion has been seen as the origin of negative externalities, affecting local residents’ well-being and contributing to the poor public image of ports. In this study, the contingent valuation method is used to estimate the costs borne by local residents as a consequence of the negative externalities derived from the growth of the Port of Valencia (Spain) in the last 30 years. As transport project appraisal has become more complex, this technique complements existing methodologies in this field, such as the social cost benefit analysis and the multicriteria analysis. Given the perceived property rights of families that have been living close to the port for a long time, a willingness to accept (WTA) framework was used. The econometric analysis undertaken reveals that among the different externalities derived from this growth, the only concern that truly affects individuals’ WTA is the reclamation of land from the sea during the port’s expansion. Finally, the results obtained show that the present value of the costs potentially borne by local residents ranges from a minimum value of €64.4 million to a maximum value of €107.4 million, depending of the aggregation criterion chosen.
Acknowledgements
Special thanks are due to professor Peter Groothuis for the technical advice in estimating the models and to two anonymous referees for their helpful comments.
Notes
1. Phillips (Citation2003) distinguishes between normative and derivative stakeholders. The first group are those stakeholders to whom the organization has a moral obligation, while the derivative stakeholders are those groups whose actions and claims must be accounted for by managers due to their potential effects upon the organization and its normative stakeholders.
2. Two pretest surveys were carried out, and 100 respondents were interviewed in each. The first one was carried out in June 2010 and the last one in July 2010.
3. Due to funding constraints, the sample size was 400 interviews. Nevertheless, this sample size implies a maximum error of 4.5% with respect to the population included in this study. As it is obvious, larger sample sizes improve the statistical accuracy.
4. As it is customary in CVM studies, when explaining the determinants of WTA, a dummy variable was created for each one of the interviewers in order to ascertain if they influenced the responses of the respondents (interviewer’s bias; see, for example, Loureiro and Lotade, Citation2005). Neither of them was statistically significant, so differences in WTA responses were not motivated by socio-demographic differences in the interviewers (age, gender, race, nationality, etc.).
5. If the 66 protest responses obtained are excluded, this figure drops to a reasonable 37.5%.
6. In order to address preference uncertainty, a ‘don’t know’ option should be included in the discrete choice question. Therefore, a practical issue arises of what to do with these ‘middle’ responses since they can be (i) excluded from the sample, (ii) treated as ‘no’ responses or (iii) treated as ‘yes’ responses. In any case, the treatment applied has an effect on the estimated welfare evaluations (Saengsupavanicha et al. Citation2008).