earticle

영어

Due to the natural process of beach erosion in the Cape Hatteras National Seashore in North Carolina, the Cape Hatteras lighthouse, a National Historic Monument, was moved inland in 1999 in order to be protected from being damaged or destroyed with persistent beach erosion. Beach renourishment was among the project alternatives considered to protect the lighthouse. However, neither the decision to move nor the possibility of using beach renourishment was considered as part of an economic evaluation of the National Park Service's decision to protect the lighthouse. This paper develops an ex post analysis of the benefits of protecting the lighthouse. The analysis takes advantage of a unique data set collected in 1993, prior to the decision to move the lighthouse. This survey assembles a panel with two surveys of the same respondents. The first is a baseline telephone survey to a random sample of 1000 NC households.It was intended to recruit respondents and collect information about their socioeconomic characteristics. The second interview contacted the same person, each individual agreeing to participate who had received a booklet describing the proposed beach renourishment plan. Benefit estimates are developed using a discrete choice contingent valuation framework. The panel structure allows the analysis to take account of a variety of selection effects that arise due to the structure of the two surveys and the specific valuation questions. An important aspect of the model derives the economic restrictions that would give rise to the selection effect associated with respondents' choices among beach protection plans offered as part of the survey. As a result, this analysis can develop benefit estimates that adjust for both the effects of interest in the topic on the response rate to a telephone based contingent valuation survey and the selection effect arising from different beach policy choices. Improved coastal planning requires better information about the costs of policy alternatives and the preferences of citizens affected by those policies. A non-market benefit analysis puts these preferences in monetary terms, comparable to the costs. This research has a unique opportunity to evaluate whether the consideration of policy alternatives for a decision that has been already made would have recommended a change in the outcome. Moreover, the analysis demonstrates how economic analysis and statistical modeling of selection effects can be integrated in the evaluation of discrete response contingent valuation data.