The affordability of private health insurance: Econometric evidence from household and firm surveys

The notion that healthcare costs in the United States are increasingly burdensome for many families is not news. Empirical research, however, has lagged behind the public need for analyses showing how household finances constrain health insurance purchase, and what particular groups may be most affected by the rising costs of care. This dissertation outlines three econometric approaches to understanding whether health insurance is affordable, and if so, for whom. Chapters 1 and 3 use a choice framework to understand the relationship between employer health plan offerings and employee purchase. Chapter 2 constructs new panel data from the Kaiser Family Foundation/Health Research and Educational Trust's Annual Survey of Employer Health Benefits to better estimate the influence of insurance premiums on employee take-up (coverage) rates.In Chapter 1, a previous definition of whether insurance is affordable for the uninsured is reevaluated to incorporate the heterogeneous choice sets available to families from their employment status. Using supply-side notions including offers of employer plans and self-employment status, a revised estimate of the percentage of the uninsured who can afford insurance is developed and shown to be substantially lower than previously thought.Chapter 2 constructs a firm-level panel and applies a fixed-effects approach to a core question in health economics: how sensitive are employees to their contribution to health care premiums? Other researchers have addressed this question, but this paper finds a lower estimate of the elasticity of demand for health insurance. Another new finding of this paper is the link it establishes between the quality of and demand for insurance coverage.In Chapter 3, a multinomial choice model is used to understand the sorting of lower socio-economic status groups into managed care plans. For those of lower socio-economic status, demand for lower cost and more restrictive care is relatively large. Unlike many household surveys, these conclusions are drawn from an analysis which controls for plan premiums, cost-sharing, and the range of plan types made available by employers. The results of modeling this sorting into HMO plans have implications for public policies which encourage the adoption of particular plan types.Source: Dissertation Abstracts International, Volume: 69-07, Section: A, page: 2804.Adviser: Thomas Hertz.Thesis (Ph.D.)--American University, 2008