Venue: The Fuqua School of Business, Duke University, 1 Towerview Drive, Durham, NC 27708-0120
Presentation
Taking Up or Turning Down: New Estimates of Worker Demand for Employer-Sponsored Health Insurance
In 2005, 62% of the non-elderly population obtained their health insurance through an employer source. Employers continue to alter benefit designs and increase employee cost-sharing in order to address rising health care costs. State and federal policymakers are concerned about whether health insurance is affordable to low-income workers and many states have initiated programs to subsidize the cost of employer-sponsored health insurance premiums to low-income and/or small-firm workers. In this study, we examine the factors that influence a worker’s decision to take up an offer of employer-sponsored insurance (ESI) through his or her own employer and simulate the potential effect of premium subsidies on take-up.
Two components of the Medical Expenditure Panel Survey (MEPS) for the years 1997-1999 and 2001 are used. We use the linked Household Component (HC) and Insurance Component (IC). The HC provides detailed demographic, employment, and insurance information on individuals within a sample of U.S. households. The IC collects detailed information on the number, types, and characteristics of health plans offered by U.S. business establishments. The estimation file includes a sub-sample of HC respondents who have an offer of ESI and can be linked to their respective establishments that were surveyed in the IC. We include several demographic and employment characteristics; the minimum out-of-pocket (OOP) premium available to the worker that is consistent with the coverage type (single or family) that most likely would be chosen given the worker’s household composition; and a measure of public insurance availability. We estimate the models using binary probit and instrumental variable probit techniques. For the latter, our instruments include a set of workforce characteristics and local labor market conditions.
Overall, we find a significant inverse relationship between the OOP premium and a worker’s probability of taking up coverage. Workers in households with two offers of ESI are more price-sensitive than other types of workers. The price effects across all worker types were substantially larger when we used instrumental variables estimation to deal with premium endogeneity. Additionally, we identified differences in the influence of alternative sources of coverage, health status, and income across household types.