The purpose of this dissertation is to develop an understanding of how managed care insurers? presence, market power, and use of selective contracting affect physician groups, and how state Any Willing Provider (AWP) and Freedom of Choice (FOC) laws moderate the effects by limiting selective contracting. The approach lies on the premise that different market structures have different implications for medical groups, and that these markets stimulate medical groups? responses in different ways. This research can help interpret theoretically the conventional wisdom that managed care has caused consolidation in the medical group market and that groups have grown larger. The proposed work will then empirically test this belief and measure the consequences for group costs, revenues, and efficiency. Medical groups and managed care insurers (MCOs) engage each other in a variety of market structures, ranging from strong competition in both markets to monopoly power in both markets. The approach taken in this proposal is that MCOs in many markets are able to increase the price elasticity of demand for medical services, thereby altering the size of the medical group that optimizes member-physician utility. In some markets, however, the relationship can be described better as monopoly-monopsonist bargaining. A proposed bargaining model describes the size and investment decisions that medical groups make in response to insurer monopsony power and selective contracting. The combination of state-level variation in the ability of MCOs to exclude providers with the variation in managed care penetration in different geographical markets allows research on how these differences affect medical groups. The empirical analysis considers both the direct effects that HMOs might have on the costs and prices of physician services and the indirect effects through medical groups? responses. Panel data from the Medical Group Management Association (MGMA) on medical group costs and outputs are used to achieve these ends. The first empirical work will use random effects to determine whether different levels in managed care?s presence in the market contribute to differences in either group size or investment, due to either the goal of increased efficiency or the objective of increased bargaining power, or simply the change in elasticity of demand. Group responses to the level of managed care in a market will be ascribed to efficiency, whereas responses to HMO concentration conditional on efficiency will be interpreted as driven by bargaining motives. The next step will use quasi-translog functions to find the relationship between group size and average cost and revenues, and whether HMO monopsony or selective contracting influences this relationship either directly or indirectly via group responses. Technical efficiency and its determinants are then measured using cost and revenue functions in stochastic frontier analysis.
