This paper consistently estimates the structural properties of a multiple-species fishing technology. Our approach overcomes two ubiquitous features of fisheries data generating processes that invalidate classical ordinary least squares estimators: (1) unobservability, by the researcher, of the in situ fish stock and its effect on productivity, and (2) endogenous production decisions that vary with fishermen’s private information about stock abundance and other economic variables. Our estimations obtain consistency by exploiting the timing of production decisions and the information available to fishermen when endogenous production scale and scope, i.e., targeting of individual species’, decisions are made. Our estimator adopts methods used in fisheries stock assessment to control for the effects of abundance on productivity. A nonlinear instrumental variables estimator controls for endogenous scope adjustments. We present an application of our model to the Gulf of Mexico commercial reef fish fishery. Comparison with an estimator that ignores latent stock abundance effects and endogeneity in production indicates substantial bias, which can be linked to past fishery management failure.
Original Release Date: December 20, 2018
Department of Economics, Iowa State University
Weninger, Quinn; Perruso, Larry; and Bunzel, Helle, "Fishing technologies: Specification and estimation" (2018). Economics Working Papers: Department of Economics, Iowa State University. 18015.