This dissertation consists of three essays. The first essay studies quality increases in the medical sector. A large and growing share of income is spent on medical goods and services each year. Existing measures of the price and quantity of medical goods and services do not take changes in quality into account. Ample micro evidence suggests the quality of medical goods and services has, in fact, improved over time. This essay estimates changes in medical quality at the aggregate level. To do so, this essay develops and estimates a dynamic structural model of the demand for medical purchases. The main result of this essay is that the quality of medical goods and services has increased by 2.2 percent per year between 1996 and 2007. One implication is that, after adjusting for changes in medical quality, the relative price of medical goods and services fell by 0.5 percent per year over this period, whereas Bureau of Labor Statistics estimates suggest it rose by 1.6 percent per year. The second essay develops a method to infer the life cycle profile of the quality of medical care in accumulating of health capital and the depreciation rate of health capital. To do so, this essay develops a life cycle model of the demand for medical purchases in which individuals invest in health capital. The use of these methods is illustrated by inferring the life cycle profile of the quality of medical care and the depreciation rate of health capital for 25-84 year old males between 1996 and 2007. The third essay studies implementable outcomes in partnership games. In this setting, it is well known that contracts which satisfy budget balance cannot implement efficient outcomes. Then, it is natural to ask which outcomes can be implemented. This essay characterizes the outcomes of all budget balancing contracts. With standard regularity conditions on production and utility functions, all outcomes which can be implemented by a budget balancing contract can be implemented by a linear contract. This implies that, with respect to welfare, we can consider a compact set of implementable outcomes without loss of generality. The budget-balancing contract whose outcome maximizes welfare, therefore, exists.

This work is driven by two facts. First, the majority of households in the U.S. obtain health insurance through their employer. Second, around 20% of working age households choose not to purchase health insurance. The link between employment and health insurance has potentially large implications for household selection into employment and participation in public health insurance programs. In these two essays, I address the role of public and private provisions of health insurance on household employment and insurance decisions, the distribution of welfare, and the aggregate economy. In the first essay, I quantify the effects of key parts of the 2010 health care reform legislation. I construct a lifecycle incomplete markets model with an endogenous choice of health insurance coverage and calibrate it to U.S. data. I find that the reform decreases the fraction of uninsured households by 94% and increases ex-ante household welfare by 2.3% in consumption equivalence. The main driving force behind the reduction in the uninsured population is the health insurance mandate, although I find no significant welfare loss associated with the elimination of the mandatory health insurance provision. In the second essay, I provide a quantitative analysis of the role of medical expenditure risk in the employment and insurance decisions of households approaching retirement. I construct a dynamic general equilibrium model of the household that allows for self-selection into employment and health insurance coverage. I find that the welfare cost of medical expenditure risk is large at 5% of lifetime consumption equivalence for the non-institutionalized population. In addition, the provision of health insurance through the employer accounts for 20% of hours worked for households ages 60-64. Finally, I provide an quantitative analysis of changes in Medicare minimum eligibility age in a series of policy experiments.