| The first chapter examines the lack of managed care penetration in rural areas in the United States. The enrollment rates in managed care organizations (MCOs) in the United States are lower in rural than urban areas. I propose a model that explains this fact by exploring the effect of imperfect competition among doctors on the type of contracts written between MCOs and doctors. A reputation-based mechanism can provide incentives for doctors to deliver high quality, but requires a positive rent for them. I show that when competition is low such as in the case of rural areas, the required rents for doctors are high. Hence, welfare of consumers in rural areas may be lower with MCOs than with fee-for-service plans.;The third chapter investigates the effects of the public provision of long-term care (LTC) insurance on individual consumption, medical care purchase decisions, and welfare. Means-tested and health-based LTC programs are studied. What are the consequences of these different types of public LTC programs on individuals' behavior? In terms of expected utilities, when the government has a limited fund, individuals with low initial health and moderate initial wealth prefer health-based programs to means-tested programs, while individuals with low levels of wealth and health status prefer means-tested programs to health-based programs.;Keywords: capitation, fee-for-service, monopolistic competition, reputation goods, health status, unemployment, long-term care.;The second chapter empirically examines the effects of unemployment on an individual's health status using the Canadian National Population Survey. Health and unemployment status clearly affect one another. To identify the causal effect of unemployment on health, I jointly estimate both outcomes allowing for correlated time-invariant random effects, correlated idiosyncratic shocks, and instrumenting unemployment to eliminate any remaining endogeneity problems. For an instrument, I use the variation in the Employment Insurance eligibility and benefits rules across regional unemployment rates. The main finding is that, for prime age male labor market participants, unemployment has a negative and large impact on self-reported health status, while it has no effect on the objective health measure. In addition, the correlations between the error terms allowed in the model are found to be large and significant indicating that health and unemployment are jointly determined via several factors. |
