Short-term responses of hospitals to the New Jersey prospective payment system, which uses Diagnosis Related Groups (DRGs) to establish rates of compensation for all payers, were examined in this study. The sample consisted of 84 New Jersey hospitals that were subject to prospective payment and a comparison group of 76 hospitals. Hospitals comprising the comparison group located in eastern Pennsylvania, were reimbursed retrospectively. Regression equations, which included independent variables to control for market supply and demand conditions, were estimated for the cost per admission, cost per day, length of stay; and cases treated. The results indicate that increases in the cost per admission and cost per day were lower (P less than 0.05) in hospitals subject to the all payer DRG system than in those institutions that were reimbursed retrospectively. In addition, the results suggest that most of the cost savings attributed to the New Jersey DRG system are due to a reduction in the average length of stay. The paper concludes with policy implications.