Harmful Algal Blooms and Coastal Business: Economic Consequences in Florida

Abstract

Abstract The impacts of harmful algal blooms (HABs) on coastal businesses in the Ft. Walton Beach and Destin areas of northwest Florida were estimated for 1995–1999. Separate time-series models for the restaurant and lodging sectors revealed that HABs reduced restaurant and lodging revenues in the localized study area by $2.8 million and $3.7 million per month, respectively, which represents a 29% to 35% decline in average monthly revenues for each sector during months of red tide incidence. By comparison, a tropical storm was found to reduce monthly restaurant revenues by $0.5 million, and each inch of rainfall reduced revenues an additional $41,000. Adverse weather was not found to affect the lodging sector. While the estimates are conservative given the resolution of data, the magnitude of effects indicate that coastal communities have suffered significant revenue losses due to HABs and that these losses are larger than caused by other environmental events. Keywords: business losseseconomicFloridaharmful algae blooms Karenia brevis red tide This research was supported by the Florida Agricultural Experiment Station (FAES), Florida Sea Grant, and a grant from the Florida Harmful Algal Bloom Task Force, Florida Marine Research Institute, Florida Fish and Wildlife Conservation Commission. The authors also thank three anonymous reviewers and a coeditor for helpful comments received during the review process. Notes Note. During these months, the corresponding variable (W 1 or W 2) will have a value equal to one. In all other months the values of these variables is zero. Note. Asterisk indicates coefficient is significant at p < .05. At the time of the study, the FDOR would only provide data for which there were at least 10 reporting establishments within each zip code and for each business code. These constraints did not preclude the use of any data for this analysis, but only ensures that there were at least 20 reporting establishments in each month for each business type (10 from each ZIP code). Separate estimations were maintained (versus interaction terms with each variable to distinguish between business types) in order to improve the efficiency of the correction for autocorrelation and to facilitate the analysis of results. Since the independent variables are identical for both sectors, a seemingly unrelated regression procedure would not improve the results. Preliminary analyses also estimated separate models for each area; however, F tests revealed that the coefficient estimates were not statistically different between areas (Adams et al. Citation2002). The analysis reported in this article uses the combined data.