Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

William J. Moore

Second Advisor

Randy Rice


This dissertation consists of three essays. The first essay provides an analysis of the interaction between special interest groups, public interest factors, median voters' tastes and preferences regarding the provision of medical services for the poor, and cost containment policies on one side, and drug reimbursement levels, the size of state Medicaid programs, and the size of states' drug budgets on the other. The relative strengths of special interest are shown to be one of the most important determinants of drug reimbursement levels and drug expenditures. The median voters' preferences significantly explain the size of states' Medicaid programs. The results verify the existence of substantial variation in state Medicaid programs, and point to potentially growing disparities as a result of current policies. The second essay examines alternatives to the traditional retrospective fee for service (FFS) payment mechanisms regarding Medicaid inpatient hospital services. These alternatives may be grouped into direct price or utilization controls, particularly, prospective payment and coverage limitations, and managed care, either in the form of fee for service primary care case management, or risk based enrollment in prepaid health plans or health maintenance organizations. Both special interest groups and median voter variables are shown to have significant explanatory power in the adoption of these alternative payment mechanisms. The third essay uses a system of six equations to examine the relative effects of direct and market driven cost containment policies, relating to the general inpatient hospital component of Medicaid. Direct cost containment policies, consisting of diagnosis related groups prospective payment systems and rate-of-increase control based prospective payment systems are found to be effective. To a lesser extent, Managed care principles also generate savings. Furthermore, significant substitute and complementary relationships between program components emphasize the importance of system-wide analysis.