A site-portfolio model for multiple-destination recreation trips: valuing trips to national parks in the southwestern USA

Date
2016
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
The multiple-site visitation problem has afflicted travel cost models since their inception. Over the years, several approaches have been proposed to address the issue of allocating travel costs when multiple-site visitations are involved; however, these approaches have proven to be problematic for generalized application. I propose a new method for analyzing multiple-destination recreation trips and apply it to visitation to national parks in the southwestern United States, including well-known parks such as the Grand Canyon and Zion National Parks. I use conventional random utility theory and treat groups of parks (portfolios) as choice alternatives. I consider one choice occasion per respondent and condition that choice on the person visiting at least one park in the choice set, so the participation decision (go/no-go) is not modeled. Trip cost includes time, travel, lodging, and food cost for visiting all sites in the portfolio. Variation in trip cost is generated by where individuals enter and exit the region and by variation in the specific set of parks in each portfolio. Specialized sampling weights are used in the model to correct for on-site sampling. I estimated three empirical versions of this choice model: Standard Logit with Additive Site Utilities (SL), Mixed Logit with Additive Site Utilities (MXL), and Portfolio Specific Constants as Utilities (PSC). I found that the PSC model performs relatively better than the SL model in terms of accounting for the complementary effects among parks. MXL model with a constrained distribution of the random parameters provides more behaviorally reasonable estimates compared to other traditionally assumed distributions. Finally, I provide estimates of values for closing individual parks or groups of parks. The loss-to-trip ratios (per trip value) for individual park closures range from $143 to $255 for Additive Site Utility Models. The aggregated welfare losses for individual park closures over the season (June 2002) range from $2.4 million for Canyonlands to $40.9 million for Grand Canyon.
Description
Keywords
Citation