Quantitative risk measures applied to Alaskan commercial fisheries

Abstract

Risk measures can summarize the complex variability inherent in fisheries management into simple metrics. We use quantitative risk measures from investment theory to analyze catch and revenue risks for 90 commercial fisheries in Alaska, USA, nearly a complete census. We estimate the relationship between fishery characteristics and catch risk using nonparametric random forest regression to identify attributes associated with high or low risks. Catch and revenue risks for individual Alaskan fisheries are substantial and are higher than risks for farmed food alternatives. Revenue risks are greater than catch risks for most fisheries, indicating that price variability is an additional source of risk to fishermen. Regression results indicate that higher productivity species tend to be higher risk, and there is an increasing gradient of risk moving north and west across Alaskan waters, with the remote western Bering Sea fisheries tending to have the highest risks. Low risk fisheries generally have large catches and support larger fleets. Finally, fisheries with greater catch history under some form of dedicated access privileges tend to have lower catch risks.