Thirty years of fleet dynamics modelling using discrete‐choice models: What have we learned?

Abstract

Abstract Anticipating fisher behaviour is necessary for successful fisheries management. Of the different concepts that have been developed to understand individual fisher behaviour, random utility models ( RUM s) have attracted considerable attention in the past three decades, and more particularly so since the 2000s. This study aimed at summarizing and analysing the information gathered from RUM s used during the last three decades around the globe. A methodology has been developed to standardize information across different studies and compare RUM results. The studies selected focused on fishing effort allocation. Six types of fisher behaviour drivers were considered: the presence of other vessels in the same fishing area, tradition, expected revenue, species targeting, costs, and risk‐taking. Analyses were performed using three separate linear modelling approaches to assess the extent to which these different drivers impacted fisher behaviour in three fleet types: fleets fishing for demersal species using active gears, fleets fishing for demersal species using passive gears and fleets fishing for pelagic species. Fishers are attracted by higher expected revenue, tradition, species targeting and presence of others, but avoid choices involving large costs. Results also suggest that fishers fishing for demersal species using active gears are generally more influenced by past seasonal (long‐term) patterns than by the most recent (short‐term) information. Finally, the comparison of expected revenue with other fisher behaviour drivers highlights that demersal fishing vessels are risk‐averse and that tradition and species targeting influence fisher decisions more than expected revenue.