U.S. federal law calls for an end to overfishing, but measuring overfishing requires knowledge of bioeconomic parameters. Using microlevel economic data from the commercial fishery, this paper proposes a two-stage approach to estimate these parameters for a generalized fishery model. In the first stage, a fishery production function is consistently estimated by a within-period estimator treating the latent stock as a fixed effect. The estimated stock is then substituted into an equation of fish stock dynamics to estimate all other biological parameters. The bootstrap approach is used to correct the standard errors in the two-stage model. This method is applied to the reef-fish fishery in the northeastern Gulf of Mexico. The traditional method, which uses catch-per-unit-effort as a stock proxy, significantly overstates the optimal harvest level. © 2011 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.