From sea to shore: development of a static Computable General Equilibrium Model (CGEM) for fisheries. Example of the fishing industry in Finistère (France).

The PECHDEV * project aims to develop and apply computable gener al equilibrium models (CGEM) to the fishing industry to account for the c ontribution made by the activities of fishery and aquaculture to the development of regio ns dependent on fishing. In this paper, we mainly focus on the methodological steps that lead to the development of a static CGEM. In a first step, we present the different important data sets required to apply such a model and we particularly point out the difficulties that can be inherent when moving from a national social accounting matrix to the regional level. The main p rinciples and hypotheses of a static CGEM are then presented and the link between biology and economy is emphasized. Finally, we illustrate the modelling choices that were made to apply the CGEM to the fishing industry in Finistere (France). For this purpose, the major fis hing activities of Finistere were split into six fishing fleets according to relevant criteria. Then , the main stocks and species were selected following their importance in terms of tonnage and value landed in Finistere. Production functions based on current knowledge of fish stock dynamics were estimated following two distinct methods linked to data availability. Final ly, expected effects that should arise when simulating the implementation of a moratorium on Celtic cod fisheries were described. Effects should be observed at the fish production level as well as in other sectors of the Finistere economy through the interrelationships of the CGEM. Such a management scenario should also drive to a new biological equilibrium through fishing effort reallocation. The CGEM for fisheries could be a very useful tool to assess the consequences of alternative management scenarios at the regional scale. Its application to the 5 different case study regions of the PECHDEV project under various management scenarios should eventually confirm its usefulness and encourage a wider use of such promis ing methods.