Purpose: Arthroscopic treatment of shoulder instability has some advantages (including short surgical time, less morbidity, less postoperative pain, reduced hospitalization time, and decreased risk of complications) compared with open procedures. We performed a prospective study comparing open repair with arthroscopic repair for recurrent anterior shoulder instability. The aim was to clarify the relative effectiveness of open Bankart repair plus inferior capsular shift (OBRICS) and arthroscopic Bankart (AB) repair without augmentations with approximately 5 years of follow-up. Methods: We investigated 32 shoulders of 30 patients (24 men and 6 women) undergoing OBRICS (15 shoulders of 17 patients; two patients were bilateral) and AB (15 shoulders of 15 patients). The average follow-up was 5 years and 2.5 months (range: 60-66 months). The clinical evaluation included recurrent instability rate, range of motion, and postoperative rehabilitation. All patients were assessed using the scoring systems of Rowe and the University of California at Los Angeles (UCLA) preoperatively and during the final evaluation. Results: Recurrent instability rates were significantly different between the OBRICS (0%) and AB (26.6%) groups (p ¼ 0.022). There were fewer limitations of external rotation (ER), ER at 90 abduction, and horizontal extension for AB than for OBRICS postoperatively (p < 0.05). The mean Rowe and UCLA scores for both methods were not significantly different at final follow-up. Conclusion: Our data suggest that OBRICS leads to a lower rate of recurrent instability. However, those with AB had fewer ER and horizontal extension limitations.