Quantifying moral hazard in health insurance across medical services for the near-elderly

The problem of moral hazard is a key issue in health care and health insurance reform and has been much discussed in the recent presidential campaigns and in President Obama's proposals for health reform. This thesis examines the price effect due to moral hazard and quantifies it across two medical services for a particular population: the near-elderly (individuals between the ages of 55 and 64): visits to the doctor and nights stayed in the hospital. After running specification tests, the preferred model is the zero-inflated negative binomial regression model for count data with a logit specification in the first stage. The results differ slightly depending on what type of insurance an individual has (public or private) and which dependent variable (doctor visits or nights stayed in the hospital) is used in the model. Those who have public insurance seem to be less affected by moral hazard and choices about visiting the doctor seem to be more susceptible to moral hazard than choices about spending time in the hospital.