A random utility model of beach use on the East Coastof the United States: per-trip values and hypothetical beach closures

Abstract

Beach recreation is a popular activity on the East Cost of the United States. Individuals travel to the shore to enjoy ocean views, participate in various activities in the water and sand, have a good time on the boardwalk, engage in four-wheel driving on the beach, go fishing, and so on. Given the high volume of beach visitations and their use for recreational purposes, many policy concerns arise when it relates to managing these beaches. Specifically, the issues of compensatory valuation in the event of beach closures due to an oil spill, water contamination or other damage to the site, becomes prominent. In order to address these issues, one needs to asses economic value of beaches. Using the 2015 beach visitation data from the Offshore Windfarm survey, I estimate random utility maximization models of beach recreation. The survey area includes 275 ocean beaches stretching along the shoreline from Massachusetts down to South Carolina. This data set is the largest beach visitation data set for the east coast beaches thus far. The four empirical models of site choice are considered in my dissertation such as standard multinomial logit, multinomial logit with interactive terms, mixed logit and alternative-specific constants logit. All of these models are estimated for the three different trip types: day trips (30 minutes to several hours long) short overnight trips (2 to 4 nights long) and long overnight trips (4 to 30 nights long). The results highlight the importance of considering longer trip types in beach valuation which has not been given enough attention based on the existing literature of beach recreation. The main difference in modelling longer trips lies in the calculation of travel cost which includes lodging and meals and incidental expenses for the short and long overnight models. I pay special attention to the calculation of travel cost as the most important component of the models. I find that short and long overnight trip models produce significantly lower travel cost coefficients which indicates the higher marginal utility of longer trips. Finally, I provide welfare estimates for beach closures based on the results of the econometric RUM modelling. The loss-to-trip ratios for individual beach closures range from $17.7 to $32.5, $88.8 to $149.1 and $324 to $1865.9 for day, short and long overnight trips, respectively. The aggregate welfare losses for beach closures range from $4.9 million to $70.5 million, $9.8 million to $159.7 million and $11 million to $2.3 billion for day, short and long overnight trips, respectively.