Many hospitals have historically used a fixed staffing policy for allocating nursing personnel, in which the daily demand in each ward is met by nurses who are permanently assigned to the specific wards. In recent years, the concept of variable staffing has been proposed as a means of increasing manpower efficiency. A variable staffing policy is one which provides for staffing adjustments to meet work load through the use of a common pool of cross-trained nurses. In this paper, a model is formulated to evaluate the relative benefits of variable and fixed staffing policies. Results from a Monte Carlo evaluation of the model demonstrate how the hospital administrator can assess the sensitivity of savings to changes in policy and operating parameters. Several criteria which an administrator might adopt for equating levels of patient care under alternative staffing schemes are suggested and studied. The proposed method of analyzing benefits of alternative allocation procedures shows promise for evaluating policy choices in hospitals as well as other service organizations with similar characteristics.