This thesis presents an analysis of two payment systems: the HEP III payment system for hospitals designed to promote quality improvement, and the RUGs II payment system for nursing homes designed to encourage admissions of heavy care patients irrespective of their payer status.

The HEP III payment system is analyzed using game theory techniques. The analysis determines the equilibrium investment in quality improvement. It shows that the incentives lead to development of centers of excellence, incentives which diminish with market share. It shows that due to the zero sum design hospitals will not voluntarily agree to participate, and if forced to participate by monopsonistic payers are likely to collude to avoid investment in quality. Private information about efficiency in quality production does not affect hospitals' decisions about participation or investment levels in equilibrium.

The analysis of RUGs II, a case mix payment system for nursing homes in effect in New York State since 1986, focuses on the type of patients nursing homes will prefer to admit from a pool of waiting patients. Profit maximizing homes will prefer to admit private pay low care patients. They will accept higher complexity patients only if those patients have higher spend downs. The analysis also determines the response of not for profit nursing homes, which are assumed to prefer poor and more complex patients, and are subject to a break even constraint.

An analysis of Patient Review Instrument data for all nursing homes patients in Monroe County, New York, for the 1986-1990 period is presented. The data are used to test hypotheses about the objectives nursing homes maximize and to simulate the nursing home market in Monroe County. The simulations suggest that all for-profit homes and 30% of not-for-profit homes behave as profit maximizers. 50% of admissions to profit maximizing homes are based on spend down only, and ignore information about RUGs scores.

Policies which lead not-for-profit homes to increase admissions of Medicaid patients have only small impact on county-wide admissions. The decrease in private pay admissions to not-for-profit homes is offset by increased private pay admissions to for-profit homes.