Multiscale Spatial Pattern in Nonuse Willingness to Pay: Applications to Threatened and Endangered Marine Species Robert J. Johnston, Daniel Jarvis, Kristy Wallmo, and Daniel K. Lew ABSTRACT. This paper demonstrates methods that may be combined to characterize otherwise undetect- able spatial heterogeneity in stated preference will- ingness to pay (WTP) estimates that may occur at multiple geospatial scales. These include methods ap- plicable to large-scale analysis with diffuse policy im- pacts and uncertainty regarding the appropriate scales over which spatial patterns should be evalu- ated. Illustrated methods include spatial interpolation and multiscale analysis of hot/cold spots using local indicators of spatial association. An application to threatened and endangered marine species illustrates the empirical findings that emerge. Findings include large-scale clustering of nonuse WTP estimates at multiple scales of analysis. (JEL D61, Q51) I. INTRODUCTION An expanding literature addresses spatial heterogeneity in stated preference willingness to pay (WTP). 1 Most research in this area ap- plies some variant of the distance decay par- adigm, in which WTP is assumed to diminish as a continuous function of distance from an affected resource (Bateman et al. 2006). Yet while distance decay has been verified in em- 1 For example, see Bateman et al. (2000, 2006, 2011a, 2011b), Bateman and Langford (1997), Brouwer, Martı ´n- Ortega, and Berbel (2010), Campbell, Hutchinson, and Scarpa (2009), Campbell, Scarpa, and Hutchinson (2008), Georgiou et al. (2000); Hanley, Schla ¨pfer, and Spurgeon (2003), Imber, Stevenson, and Wilks (1991); Johnston and Duke (2009), Johnston and Ramachandran (2014), Jørgen- sen et al. (2013), Loomis (2000); Moore, Provencher, and Bishop (2011); Pate and Loomis (1997), Rolfe and Windle (2012), Schaafsma, Brouwer, and Rose (2012), and Suth- erland and Walsh (1985). Land Economics • November 2015 • 91 (4): 739–761 ISSN 0023-7639; E-ISSN 1543-8325  2015 by the Board of Regents of the University of Wisconsin System pirical settings involving both use and nonuse values, the theoretical justification for this ef- fect is limited to use values (Bateman et al. 2006). There is no clear theoretical justifica- tion for distance decay in nonuse values. This may help explain why distance decay analyses of nonuse and nonuser values have found mixed results (Bateman et al. 2000, 2006; Hanley, Schla ¨pfer, and Spurgeon 2003; John- ston and Ramachandran 2014; Jørgensen et al. 2013; Loomis 2000; Rolfe and Windle 2012). The lack of theoretical justification for dis- tance decay does not imply that nonuse values are spatially homogeneous. Rather, it implies that spatial patterns in these values may occur in many different forms, including noncontin- uous or discrete distributions that appear at different geospatial scales. Despite this pos- sibility, there has been little development of methods able to characterize discrete spatial patterns in stated preference WTP estimates. This includes methods suited to cases in which the scale at which spatial patterns (or clustering) might emerge is unknown. For ex- ample, in some cases statistically significant spatial heterogeneity might occur at relatively small scales (e.g., within less than 20 km; cf. The authors are, respectively, director, George Perkins Marsh Institute, and professor, Department of Eco- nomics, Clark University, Worcester, Massachusetts; Ph.D. candidate, Graduate School of Geography, Clark University, Worcester, Massachusetts; econo- mist, Office of Science and Technology, Economics and Social Analysis Division, National Marine Fish- eries Service, National Oceanic and Atmospheric Ad- ministration, Silver Spring, Maryland; economist, Re- source Ecology and Fisheries Management Division, Alaska Fisheries Science Center, National Marine Fisheries Service, National Oceanic and Atmospheric Administration, Seattle, Washington.