Modeling Spatial Patchiness and Hot Spots in Stated Preference Willingness to Pay

Stated preference analyses often impose strong assumptions regarding spatial welfare distributions that can influence the validity of welfare analysis and aggregation. These include spatial homogeneity and continuous distance decay. Global assumptions such as these are increasingly questioned by non-economics disciplines in favor of approaches that allow for local patchiness. Drawing from this literature, this article proposes methods to identify and evaluate hot spots in stated preference welfare estimates using local indicators of spatial association. Methods are illustrated using geocoded choice experiment data addressing river restoration. Results suggest the presence of statistically significant, non-continuous patterns overlooked by current approaches.