The deep-sea red shrimp Aristeus antennatus is one of the most valuable demersal resources in Catalonia (northeast Spain), and Palamós is the most important harbour for this fishery in the area. Here, a management plan published in 2013 established the use of a 40-mm square-mesh codend (40s), replacing the previously used 50-mm diamond mesh codend (50d). The objective was to decrease the amount of juveniles in the catches, but the regulation did not bring the expected results. In this study, we measured and compared the selectivity of the 40s and the 50d in this fishery. In addition, we experimented with the use of a 50-mm square-mesh codend (50s). According to our results, the 40s had a lower 50% selection length (L50) than the 50d, while the 50s had a substantially higher L50 than the other two. A transition analysis showed an increase in yield per recruit after the second year from a hypothetical implementation of the 50s. Our conclusion is that the 40s does not have a higher selectivity than the 50d, which (at least partially) explains the failure to reach the management objective in Palamós. Conversely, the use of a 50s would significantly benefit the fishery, increasing gear selectivity and yield per recruit.