Purpose The purpose of this study was to compare the cost-effectiveness of initial observation versus surgery for first-time anterior shoulder dislocation. Methods The clinical scenario of first-time anterior glenohumeral dislocation was simulated using a Markov model (where variables change over time depending on previous states). Nonoperative outcomes include success (no recurrence) and recurrence; surgical outcomes include success, recurrence, and complications of infection or stiffness. Probabilities for outcomes were determined from published literature. Costs were tabulated from Medicare Current Procedural Terminology data, as well as hospital and office billing records. We performed microsimulation and probabilistic sensitivity analysis running 6 models for 1,000 patients over a period of 15 years. The 6 models tested were male versus female patients aged 15 years versus 25 years versus 35 years. Results Primary surgery was less costly and more effective for 15-year-old boys, 15-year-old girls, and 25-year-old men. For the remaining scenarios (25-year-old women and 35-year-old men and women), primary surgery was also more effective but was more costly. However, for these scenarios, primary surgery was still very cost-effective (cost per quality-adjusted life-year, <$25,000). After 1 recurrence, surgery was less costly and more effective for all scenarios. Conclusions Primary arthroscopic stabilization is a clinically effective and cost-effective treatment for first-time anterior shoulder dislocations in the cohorts studied. By use of a willingness-to-pay threshold of $25,000 per quality-adjusted life-year, surgery was more cost-effective than nonoperative treatment for the majority of patients studied in the model. Level of Evidence Level II, economic and decision analysis.