Health insurance, networks, and provider competition

This dissertation contains three chapters. The first discusses the impact of Taiwan's National Health Insurance (NHI) on entry and competition in the market for dentists. The NHI was designed to ensure equal access to adequate health care for all citizens. Because the NHI is expected to increase health care demand, there must be a sufficient supply response to avoid shortages. Although the NHI resulted in regulated prices, increased utilization may have increased rather than decreased provider revenues. Higher profits made entry to rural markets in particular more attractive. It is an empirical question whether there was sufficient entry into underserved areas after the implementation of the NHI. I calculate the minimum market size required to support a given number of dentists to test whether the NHI made entry sufficiently attractive. My results support that the implementation of the NHI leads to increased entry and competition in the market for dentists.; The second chapter investigates whether managed care organizations (MCO) can affect their enrollees' choice of hospitals. I hypothesize that MCOs attempt to steer patients toward low-cost and high-quality hospitals. One must recognize, however, that MCOs may lack information on hospital cost and quality. I therefore test the hypothesis whether MCOs gradually direct patients. I find that if costs and qualities of a hospital deviate unfavorably from the market average in a period, then MCO enrollees will less likely use the hospital in the following period. The results support the presence of an MCO redirection mechanism.; The third chapter assesses whether multi-hospital health systems decrease the influence of managed care networks. Reacting against the use of networks by MCOs, hospitals can establish an alliance or a partnership. I hypothesize that horizontally integrated hospital systems reduce MCOs' ability to control cost and quality, and that such hospitals are less constrained with respect to cost and quality than those hospitals unaffiliated with any system. I identify health-system effects among hospitals in Massachusetts. The empirical results establish that hospitals that belong to a system have significantly larger and unfavorable cost and quality variations, which supports my hypothesis.